This article about minimal risk assets is by former hedge fund manager turned author Lars Kroijer, an occasional contributor to Monevator. He also wrote Investing Demystified.
I believe any private investor can create a sound and well-diversified investment portfolio by using just two assets as building blocks – a world equity index fund, and an appropriate government bond fund.
Other assets can be added to suit. But those two assets alone can be a firm foundation for your long-term investment plans.
There have already been several articles here on Monevator about world index funds. However I’ve not spoken much about the second piece of the puzzle – the minimal or ‘risk-free’ government bond fund.
This article is one of a short series to put that right!
Today we’ll look at how you can decide upon the lowest-risk investment that will be the basis on which your riskier portfolio can be built.
Next time I will describe how and why you should match the time horizon of your minimal risk asset to your investment time horizon, how you invest in your chosen minimal risk asset, and what returns you can expect to make.
What is the minimal risk asset?
For a sterling-denominated investor, short-term UK government bonds are a good choice for your minimal risk-asset.
There is probably no genuinely riskless security in the world today. However the probability that the UK government will default on its debts is as low a risk as we can find when investing in sterling. Thus it is ‘minimal risk’.
Incidentally, cash in the bank is not entirely without risk, as I have discussed before on Monevator. But it is worth looking briefly at cash as a minimal risk asset, as there can be some particular benefits for private investors. Look out for that below.
One of these days I’ll also outline why other asset classes traditionally designated low risk – gold, property, physical assets – are not that low risk at all.
Buy government bonds in your base currency
Your choice of your minimal risk asset also depends on your base currency.
A US-based investor buying short-term UK government bonds has the same security of getting his principal back as any British investor. But they also incur additional currency risk due to exchange rate fluctuations.
If, for example, the UK government bond promised to pay the investor £101 a year hence for a £100 investment today, both investors are equally certain of receiving £101. But while the £101 would always be £101, the US dollar value of that amount will fluctuate quite a bit and is thus riskier.
The US investor would therefore be better served by choosing as their minimal risk asset short-term US government bonds. These bonds would be of a similar credit quality to the UK government bonds, but the returns would be independent of currency risk.
Similarly, a French or German investor could opt for German government bonds.
Wait, what is my base currency?
While most reader’s base currency is obvious (sterling for UK investors, dollars for US ones, and so on) and currency risk is a risk you would rather avoid, your base currency can also be a mix of currencies.
Your base is the currency that you think you will one day need to spend the money in.
For example I live in the UK and will probably have the majority of my future expenses here. But I also spend a lot of my time (and my money) in Denmark, the Eurozone, and the US. I may have future expenses for my children’s education outside the UK, and my wife and I might live or retire abroad one day.
By having my base currency as a mix of several currencies, albeit dominated by sterling, I can better match my future cash needs and reduce the risk of my being caught out by a falling currency against my future foreign-denominated expenses.
Rate my bonds
If your base investment currency is one where the government credit is of the highest quality, those government bonds will generally be a great choice for your minimal risk investment.
But how do you know if your government bonds are the good stuff?
Most people will need to turn to the professional ratings agencies. Today there are three major credit agencies that rank the creditworthiness of bonds – Moody’s, Standard & Poor’s and Fitch.
Here are the classifications these agencies use to rate long-term bonds:
| Long-term bond ratings | Moody’s | S&P | Fitch |
| Prime | Aaa | AAA | AAA |
| Investment grade | Aa1 to Baa3 | AA+ to BBB- | AA+ to BBB- |
| Non-investment grade | Ba1 to Ca | BB+ to C | BB+ to CCC |
| Default | C and lower | D | DDD to D |
Source: Author/Various agencies
The credit agencies were widely discredited after 2008 when they wrongly gave high ratings to all sorts of sub-prime garbage. In general though they give you a good indication of the credit quality of a country’s bonds.
Credit ratings change frequently. When you consider adding to your minimal risk asset, you can look up the latest credit ratings on Wikipedia by searching for ‘List of countries by credit rating’. If the government credit of your base currency is listed there as AAA then you have an easy choice for your minimal risk asset.
With the adverse environment of government debt and deficits in recent years, the list of AAA-rated countries from all agencies has shortened. That said, if your home base currency offers AA or higher-rated bonds then it would be sufficient to accept those as your minimal risk asset. If we only accepted bonds with the very highest rating, at the time of writing this would exclude bonds from major economies like the US, UK, Japan and France, which is neither practical nor desirable for many investors.
While there is obviously a reason for these countries losing the highest rating, it is worth noting that the financial markets trade these countries’ bonds at real yields that are among the most creditworthy in the world in any currency.
Beyond government bonds
I have referred repeatedly to government bonds, but what we are really after is the lowest risk investment for you, given your currency and maturity.
In many countries, there are other kinds of domestic bonds related to the sovereign issuer, such as government-guaranteed regional, city, or municipal bonds. Those and similar bonds could be reasonable alternatives as minimal risk assets, particularly if there are tax or other advantages to investing in them. However, you need to make sure that the government guarantee is bulletproof, even in distress.
If you get a superior yield from these alternative bonds compared to the standard government bonds, you are probably taking additional credit risk.
Also, be careful in thinking that adding these kinds of bonds provides you with additional safety. They are typically poor diversifiers of risk, because they tie back to the same creditworthiness as the domestic government bonds.
In the end, government bonds will often be the best choice for the minimal risk asset.
What about cash?
Everything I’ve written so far would apply today to funds and to most very wealthy individuals.
However what if you’re a more typical private investor, with maybe a few tens of thousands of pounds invested in your ISA or SIPP – or even a larger portfolio but with your non-equity allocation still amounting to a low-ish six-figure sum?
Well, these days you may decide it is better to keep your minimal risk asset as cash instead of government bonds.
There are a couple of reasons for this. For one thing, after years of easing by central banks, government bonds yields are currently very low. Depending on which country you’re based in, you may well be able to find cash savings accounts that pay a higher interest rate than you’ll get on similar government bonds.
- For instance, in the UK you can currently get almost 2% for cash that you lock away for three years, provided you choose from the Best Buy accounts.
- In comparison, short-term UK government bonds (gilts) are yielding less than 0.5%.
To trust that your money in the bank is safe, there has to be a reputable credit insurance scheme in place. In the UK that’s the Financial Services Compensation Scheme (FSCS), which guarantees deposits in authorised banks to the tune of £85,000.
Now, while the FSCS was set-up by the UK government, it operates independently of UK lawmakers. The compensation scheme is funded by compulsory levies on financial firms authorised by two UK regulators – the Prudential Regulation Authority (PRA) and the Financial Conduct Authority (FCA).
The scheme has paid out billions in compensation since being set up by Act of Parliament in 2000, and means cash in a bank that’s covered qualifies as ‘minimal risk’ under any sensible definition of the term. However you might wonder what would happen if the FSCS ran out of firepower due to a financial crisis?
In practice, while the FSCS operates independently of the UK government, the latter provides an implicit backstop. For example, in 2008 the authorities and the FSCS worked together to guarantee the deposits of 2.5 million customers of the Bradford and Bingley Building Society, which saw the FSCS receive an initial top-up loan from the Bank of England and later from HM Treasury.1
Effectively then, provided you choose an FSCS-covered bank for your savings, you have the same credit risk as with gilts – because both are backed by the UK government – but you’re getting a higher interest rate.
That’s attractive, and not an option for institutions.
Remember that if you decide to use cash as your minimal risk asset and you have more than £85,000 to find a home for, you’ll need to open more than one account with different FSCS-protected firms – operating under separate banking licences – to ensure all your money is covered.
You will also have to keep moving your cash as your higher-rate terms come to an end.
At some point as your wealth grows you might decide it’s easier to keep your money in bonds, not least for the diversification benefits – government bonds tend to go up when shares fall – and also because the yield on government debt may improve in the next few years, and so reduce this ‘free rider’ gap between cash and short-term government debt.
Remember that any credit insurance backing a bank is only as good as the government providing it. Besides the details of the credit insurance scheme itself, you’ll want to evaluate how the credit markets perceive your government’s standing when deciding whether you can trust such schemes.
Also remember that in a really dire scenario, it’s possible a country’s government may decide to reduce the compensation limits, which might mean it was ‘fake insurance’ – and not there when you need it.
However for the UK and US I believe this is very unlikely indeed (the UK actually extended its coverage to Icelandic banks in the last crisis).
Help! I don’t trust my government!
For all their undoubted economic successes over the past decades, countries like Brazil, Mexico and India do not have highly rated government bonds – they are all BBB rated or lower at the time of writing.
If your base currency is one without a highly rated bond available, you face a tougher choice.
For example as an Indian you could buy Indian short-term government bonds, which would not be minimal risk, or else you could buy highly-rated government bonds in one or a couple of foreign currencies, which introduces currency risk.
Depending on the credit rating of your base currency government, you may choose to take the credit risk of the domestic government bonds instead of taking the currency risk of highly rated foreign bonds, or perhaps even keep money in cash deposits in the local bank if that is considered a superior credit option to domestic government bonds.
Incidentally, the absence of a great local currency minimal risk asset is one reason why an asset like gold is the de facto ‘money under the mattress’ asset in places like India.
Older people in certain parts of the world, such as India, undoubtedly remember previous eras of domestic economic turmoil. The thought of buying local government bonds as their minimal risk asset will seem like heresy to them.
And they are right. These investors do not have essentially risk-free bonds in their local currency – however far the government has come. Perhaps one day the credits of these governments and many like them will grow in esteem to the point that they become the lowest-risk bonds in the world, but not today.
As a result, investors with a less-creditworthy domestic government often make their base investment currency the US dollar, because of its status as the global reserve currency. They then choose US government bonds as their minimal risk asset.
While the lower credit ratings of some countries’ government bonds mean that their bonds yield more, this is not a good reason to have them as your minimal risk or safe asset. If you want to add returns to your portfolio, you can do so by adding broad exposures of equities instead. These have the added benefit of both being geographically diversified and adding expected returns.
Consider diversifying even the very low risk that your domestic government might fail
Investing in sub-AA credit ratings is a question of degrees. Some investors would be happy to invest in their BBB-rated local currency government bonds whereas others would rather invest abroad with currency risk than have an AA domestic-rated government bond.
The choice partly depends on your situation and your sensitivity to currency risk versus domestic credit risk.
For those inclined to accept sub-AA domestic government bonds as your minimal risk asset, I would encourage you to think about what else would happen in your portfolio if your domestic government defaulted. In many cases, a domestic government default could have a catastrophic effect on your portfolio and on your general life. If you had diversified some of the domestic risk away by having your minimal risk asset as highly rated foreign bonds, such as German, UK, or US government bonds, then you would at least have some respite when the domestic calamity hit.
Some investors believe that having all your minimal risk assets invested in the bonds of just one government, however creditworthy, is a bad idea. Those investors argue that while the government bonds of Britain or Germany are highly rated today, there is always some risk that they could fail – perhaps even spectacularly and quickly2. Because of this possibility, they argue investors should diversify their minimal risk asset into a couple of different, highly rated government bonds, even if this means taking a currency risk for those bonds that are not in your base currency.
My own view is that if you are invested in government bonds that are among the most highly rated in the world, the probability of a sudden default is so low that for practical purposes it is a risk you can feel safe taking.
Minimal risk assets and you
Here are my own recommendations for minimal risk assets for various base currencies:
| Base currency: | Suggested minimal risk asset: | Alternative minimal risk asset: |
| US dollar | US government bonds | Mix of world-leading government bonds (take a currency risk) |
| Euro | German or AAA/AA Eurozone government bonds | Mix of world-leading government bonds (take a currency risk) |
| UK Sterling | UK government bonds | Mix of world-leading government bonds (take a currency risk) |
| Other currency with AAA/AA government credit | Domestic government bonds | Mix of world-leading government bonds (take a currency risk) |
| Other currency with sub-AA domestic government credit | One or a mix of world-leading government bonds (take a currency risk) | Domestic government bonds (take a credit risk) or bank deposits if a strong credit bank (or other) |
Source: Author
As you can see, I believe your minimal or ‘safe’ asset is not necessarily your domestic government bond.
Consider a Spanish investor who is after the lowest risk asset, and does not want to take a currency risk. This investor should not be buying Spanish government bonds that are relatively lowly rated, but rather should buy German government bonds that are also euro denominated.
If this investor did not want the minimal risk to be the bonds of just one government, he could diversify by either adding other euro-denominated government bonds, or he could accept the currency risk of investing in highly rated non-euro government bonds from the US or UK.
Watch out for more on risk
Below you’ll find a video that recaps some the things I’ve discussed in this article. (You will also find other interesting videos on my YouTube channel).
In my next article I’ll explain why and how you should match the time horizon of your investment to the minimal risk asset, what returns you can expect from these minimal risk investments, and how you can go about buying them.
Lars Kroijer’s book Investing Demystified is available from Amazon. He is donating any to medical research. He also wrote Confessions of a Hedge Fund Manager.
- In the end, B&B customers were transferred to rival Abbey and compensation was not required. [↩]
- For those who don’t think government bonds can default I would encourage you to read This Time is Different: Eight Centuries of Financial Folly by Carmen Reinhart and Kenneth Rogoff (Princeton University Press, 2011). The authors make a mockery of the belief that governments rarely default and that we are somehow now protected from the catastrophic financial events of the past. [↩]
Comments on this entry are closed.
I’m reading Lars’ book at the moment and recently increased the Government and Corporate Bonds in my pension (I was invested a little too heavily in equities for my risk tolerance/timescales really).
I see another “Bond Bloodbath” is being predicted in the financial press (again), as interest rates look headed back to some sort of normalcy. I know people have been predicting a slump in bond prices for some time, which hasn’t come. It obviously will at some point I guess, and they certainly have a fair way to fall back based on their gains over the last few years.
I guess as Rational Investors we don’t pay attention to such noise, but I can’t help but think Bonds seem set to get a bit more exciting than we’ve seen in the recent past. ;o)
http://www.bbc.co.uk/news/business-39325794
Can you please provide some age related thoughts for holding bonds.
It seems to me that the older we get there is less distinction in holding cash or bonds. At 70 my peak spending in retirement my last 10-15yrs. After that capapability to travel, get insurance and possible care fees will then require greater drawdown in all asset classes.
Thanks for such an informative overview.
Equities and cash are more or less a no brainer as far as I can see but what to do about bonds, that is the question that nobody seems to be able to answer. I currently only hold cash and short term bonds on the fixed income side but I feel like I should have some VGOV for the long haul. Should I buy or should I wait?
Lars- great article.All very clear
What do you think of UK investors diversifying their Bond portfolio in the same way as their Equity portfolios and for the same reasons.
For instance Vanguard has a World Bond Index fund GPB hedged
Seems to be made up of mostly “safe” short term Govt Bonds-mainly US therefore -plus a few coporate bonds.
Hedging controls the currency risk
xxd09
To follow on from Malcolm’s comments above, I currently have my bond portfolio split between three Vanguard funds; Global Bond Index (GBP) Hedged, UK Govt Bond Index and UK Inflation-Linked Gilt Index. I’ve done this purely to diversify the bond allocation.
I’ve been investing since my late teens but have only recently (2 years ago) began to take control and move towards a passive lower cost stance. I’ve no real idea if bond index diversification is needed and perhaps one of the three funds would suffice?
Just wondering if anyone has any thoughts on people living in the UK choosing not to hold UK Government bonds given the UK state pension? I.e. treat the state pension as a pseudo bond fund given it is government backed?
Would the state pension be considered a minimal risk asset?
I live in Poland and I’m still not sure about which minimal risk asset should we choose in here. Government credit rating isn’t so good (S&P BBB+, Fitch A-, Moody’s A2) so it’s not so safe to invest in our government bonds.
What about global bond physical currency unhedged ETFs? Like iShares Global AAA-AA Government Bond UCITS ETF (distributing, ticker SAAA) or iShares Global Inflation Linked Govt Bond UCITS ETF (accumulating, ticker SGIL)? The former is much more diversified between countries (the latter invests >75% of assets in US and UK government bonds).
@Fly By Night
I treat the state pension, my defined benefit pension, NS&I Index-linked certificates and cash as the bond component in my overall (not just investment) financial models. I don’t own any “proper” bonds.
I don’t get too hung up on the various definitions, but I see those items above as (to a greater or lesser extent) low risk, Government-backed and inflation-proofed. As such they are the under-pinning of my retirement fund asset base. So when I do invest, I invest 100% in equities.
Various people can, will and have said that the state pension cannot be relied upon. I suppose that depends on your age and whether you think the universal state pension will be replaced by something either means-tested or by an alternative (e.g. auto-enrolment pensions). I’m happy to include the State Pension in my models as I don’t believe what I have accrued so far will be taken from me retrospectively. The new flat rate state pension is c£8,000 pa. It seems to me that ignoring it from my calculations is taking prudence too far given that it will cover my core spend (utilities, Council Tax, insurance, basic provisions etc) in retirement.
So my answers to your questions are yes and yes.
Cheers
Richard
hmm – treating state pension as bond component
thats interesting, my initial gut feeling is that its prob also sensible
i’d agree a ‘golidlocks’ amount of prudence is usually whats required, not too much, not too little. I was never convinced, for example, by RITs routine dismissal of state pension in his calculations, but maybe that was something to do with his overseas ambitions?
I view the state pension as here to stay, in some form at least.
I suspect that the age at which you can access it will get higher and higher though.
Any government that tried to get rid of it , would ( in the words of sir Humphrey ) be making a very “courageous” decision.
Means testing would be hideously complicated too.
it might get less generous when auto enrolment pensions start being taken though.
Interesting thought with regard to using it as a bond substitute though.
Calling pension and cash bond exposure misses the point that government bonds out-perform in a crisis. Look at the performance of US 10 years in 09 for example. This happens due to a flight-to-safety and is my main reason for holding gov bonds.
Lars says you should hold foreign currency when you think you might one day spend in that currency. This is true but even living in the UK you spend on imported foreign goods. These inflate in price when GBP falls (as we’re about to see) so perhaps everyone should have some foreign currency exposure.
but you can get that FX exposure from *any* asset type denominated in a different currency. doesn’t have to be govt bonds..
@John
I guess we’re back to definitions. I invest in equities in the hope of making a long-term real gain (with lots of – potentially major – ups and downs along the way).
I have pensions, ILSC and cash as my “minimum risk assets” as (baring Armageddon) they should always, by and large, retain their real value with no ups and and downs.
To me bonds have more risk (i.e. volatility) than that.
I’m not equating those other assets with bonds in any technical sense. I’m simply saying (for my own investing philosophy) that if I want minimum risk (of a real loss) I’m happy to use pensions, ILSC and cash rather than bonds.
For me, the two arms of my investing philosophy are, and the hope is:
Equities = make me money;
Pensions, ILSC and cash = don’t lose me money.
Cheers
Richard
True, it’s a separate issue. The question of the correct exposure it a difficult one. Clearly owning FTSE100 in GBP also gives FX exposure as the recent Brexit debacle showed. Perhaps that’s enough.
@ADR: perhaps you should count your house as equity too, of a particularly undiversified, illiquid, near-indivisible but tax-favoured kind.
As for “minimum risk” doesn’t it all depend on your time scale? Over a millennium govt bonds are maximum risk: the government will no longer exist or will have defaulted. On that timescale precious metals are probably your best bet for minimum risk. If you can keep them safe from the looters. The question is, what about a timescale of a few decades?
@dearieme
My house is a difficult one. It’s in my financial model but only really comes into play if I’m modelling care home fees or a potential inheritance for someone. Otherwise it doesn’t feature as – bar downsizing – I can’t get any retirement income from it (the estimation of which is the main purpose of this financial model). The hardest thing about modelling my house is what real rate of growth should I apply to its value? If I use 5-10% pa then I won’t be going into a care home when I get doddery and sell up, but into the Ritz. Unless care home fees rise at the same rate… 🙁
Timescales for me are a handful of decades, and I expect to be investing through my retirement (so I don’t use any element of “lifestyling” or risk reduction as I get older). As noted above my state pension should cover the (very) basics and my DB pension the other essentials. Thus – hopefully – by the time I get to pension age my equity investments will be play money. I’m intending to draw down my SIPP (for tax reasons) to cover the time between retiring and drawing my pensions. ILSC and cash are then there as a buffer in case of “problems” in the equity market.
But so much depends on individual circumstances. I’m risk averse so would probably hardly use equities if I had – or will expect to have – enough cash to make things work out. Sadly I don’t, so the stock market it is…
Cheers
Richard
I own my age in cash and short term bonds (split equally) and the balance in a low-cost global equity index fund.
@ADR: Short dated gov bonds are “minimum risk/volatility” but kinda like cash. My argument is for long dated government bonds which have significant volatility. My point is that this volatility is usually a good thing when markets crash as it should be to the upside. This negative correlation is the same reason that I hold gold and would have CTAs if they were cheaper (Don’t tell TI I said something not entirely negative about hedge funds). It’s certainly not the case that I think these assets will outperform equities in the long term, or that they are “min risk”. It’s that they will likely make me feel better when things go wrong.
Thanks for the article Lars. Perfect timing for my current needs. Look forward to the coming articles.
Isn’t there an element of a play on words here? When talking about equities, professionals use “risk” to mean volatility. But talking about bonds and cash, Lars means “risk” in the layman’s sense, i.e. the risk of permanent loss of capital and income.
Interesting article Lars, but why should the British people trust the UK government any more than the Indian people trust theirs?
A few quick stats:
Uk debt/GDP: 90%
Indian debt/GDP: 69%
UK current deficit: 4.4%
Indian current deficit: 3.5%
It also seems to me that the India economy has a much greater potential for income growth (improving the tax base) than the UK. Both control their own currency, so I’d be interested as to why India is considered BBB compared to AA for the UK?
Ratings agencies are notorious for getting right – after the event happens!
@ Brain Unwashed
“I feel like I should have some VGOV for the long haul” BU
Early in the article Lars states :-
“For a sterling-denominated investor, short-term UK government bonds are a good choice for your minimal risk-asset”.
VGOV is not short-term!!
Whether we go from short (e.g. IGLS) to mid-long term such as VGOV to seek enhanced -ve correlation with stocks, but maybe more potential downside risk is an interesting point; picked up on by John in comment 18.
@magneto – agreed. It’s the duration of VGOV and the low interest rate environment that we’re in that put’s me off buying it. However, I doubt IGLS will offer much in the way of offset should stocks take a dive, whereas VGOV might. There must be a sweet spot between -ve correlation with equities on the one hand and interest rate risk on the other but it’s a tough one. Some mix of cash/IGLS and VGOV (or longer) to lower the overall duration I suppose. Barbell time?
Nice article and some interesting comments. Like some of the early comments, I’m a bit surprised Lars doesn’t mention currency hedging as a way of mitigating that “(take a currency risk)” in the table of recommendations. Currency hedging of bond holdings has been well gone over on Monevator before (most recently http://monevator.com/currency-hedged-etfs/ I think, and I remember some discussion on another article last year which drew my attention to the fact all Vanguard’s bond funds/ETFs were currency hedged)… so I assume Lars has some specific reasons for apparently preferring unhedged or thinking the hedging is defective somehow… be quite nice to know what they are?
One other thing: one of the most entertaining things I’ve read in the last year is Lionel Shriver’s “The Mandibles: A Family 2029-2047” (2029 is 100 years on from the Great Depression of course) – a future history of the aftermath of a full scale bond default by the USA, describing the impact on the several generations of a family affected by it. Highly recommended for visceral thrills to anyone with an interest in “safe” assets.
I generally like Monevator articles but this one seems somewhat weak. Why should short-dated bonds (say 1-5 years) be more effective than a duration weighted barbell of cash and longer-dated bonds (20-30 years)? Less cash used, higher convexity, higher correlation to equity falls etc. Worse, he seems to be recommending exposure to short-dated foreign government bonds on a currency unhedged basis. By construction such as fund will have little duration risk or yield, so returns will be driven by GBP/XXX volatility. Essentially one would taking a view on a basket of 2-year fx forward outrights. In what manner is that a risk minimal asset? Currency hedged should be the default unless you have a view on GBP or are investing in higher yielding EM government bonds (which is hardly risk minimal). I also think that you need to bear in mind that the short-term bonds of many developed countries are deeply distorted by unconventional policy measures. Do you really want to buy German 2-year bonds (Schatz) at a yield of -0.76%. It’s at that yield due to scarcity issues associated with the ECB QE capital key, the negative depo rate and a small implied probability of Euro-area breakup if Le Pen wins.
If you want a truly risk minimal asset, surely domestic T-bills are the correct solution (or cash in an NSI account!)
Presume the idea of a risk free asset is that the possible loss in value over the timescale of interest is bounded and much less than the risk associated with equities, commodities, commercial bonds etc. The risk of holding cash is inflation and default by the holding institution. With gilts held in a platform you still have the remote risk of the holding trustee going under and in that case I believe the safety net is 50k cf 85k with cash.
@mr-flibble — Eh? From the article, for a UK Sterling investor Lars recommends “UK Government Bonds”. Exactly as you suggest. The rest of the article is an explanation of the pros and cons of going off-piste (including a sizeable discussion about cash).
@all — I agree there could have been some useful mention about hedged bond funds; on the other hand this article is well over 2,000 words already! 🙂 Perhaps I can persuade him to cover hedging in a future article, by popular demand.
I can’t speak for Lars but one reason he may not favour hedged government bond funds is as returns/yields are currently so low already, even relatively low hedging costs are going to eat a chunk of returns.
That said, I’ve noticed over the years people/readers have generally looked to overseas bonds (/corporate bonds, infrastructure, etc) as a way of chasing higher returns from their ‘minimal risk’ component of their allocation (as opposed to say diversifying their government credit risk).
But in the Lars’ worldview, the low risk part of the portfolio is simply not there to deliver returns, not above whatever the market is giving you at the moment. If today that return on offer is a low/negative real yield, so be it. Returns come from your equities, in the Lars’ scheme of things. (And many other writers of course).
We haven’t had a big crash for a while. (Heck, we didn’t have a 1% decline in the US market for several months until this week). Crashes are when you’re glad you had some of these lousy low-yielding non-equity assets around. The rest of the time they feel like dead weight — obviously so at the moment, but sometimes to many investors even in times when cash/gilts were yielding 6%+. (For example in the late 1990s, when holding cash/bonds seemed nuts with equities going up 20% a year and techs multiples of that).
For the best chance at the highest returns, hold only equities (and don’t look at your portfolio for 30 years). Adding bonds (/cash) is about risk-adjusted returns, and downside protection. 🙂
But always keep in mind that when I started this blog, *many* people were saying investing in equities was crazy, and that suggesting people do so was irresponsible. (Yes, really). We’re in the good times. They may last for days, months, or years, but they won’t last forever.
Thanks for the many comments and to Monevator for explaining my view. My view is that it makes sense for a lot of investors to add both corporate and non-home government bonds to a portfolio if you are ok with a slightly more complex portfolio (which this forum would be, but far from all investors are). The reason for simplicity in my view is the most important factor has to do with costs. Any complexity tends to add layers of advice, time, tax, etc. that quickly can eliminate any advantage that adding these asset classes bring.
This piece was really about finding the lowest risk investment, not making returns, and it really does depend on your currency and time horizon (in next blog I’ll explain that more). I’m not a big believer in currency hedging, mainly due to costs and the fact that it is hard to do very accurately (leaving the exposure a bit unknown). I remember having to hedge a relatively simple set of hedge funds in only a couple of currencies, and perhaps I’m biased as a result. But of course if you want to hold foreign bonds as minimal risk assets it can make sense (but please look closely at costs).
See some thoughts on duration and age related investing and would be happy to do a piece on that. There is a chapter in my book on it too.
Cheers
On the age-related front, here’s one Mr Kroijer wrote for us earlier:
http://monevator.com/how-should-you-invest-for-your-age/
Hi, the time horizon is a key point. If you have to sell longer dated gilts – or a longer dated gilt fund – early you risk crystalising a loss.
For my SIPP I’m now of the mind to buy individual gilts that run until my 55th birthday, that will make up at least a good part of the 25% of assets I cash in to take my tax free lump sum.
Now for me personally these gilts are intermediate dated at present (over 5 years but under 10).
This will give me more yield than say the IGLS ETF, with less costs as well. Just hold till redemption and no interest rate risk so no nominal loss possible. Only inflation risk which is the one sour bit.
As for assets held in a drawdown portfolio with an unlimited time horizon I can see why people are more reluctant to hold gilts, at least until they start yielding 3-4% again – if that ever happens!
Fascinating comments. Its interesting to see what assets people are holding. I’ve read loads about portfolio construction and I get snow-blind between whether to go for a complex arrangement of currency hedges/bond duration/equity types. Or alternatively just ditch the entire lot and buy a Vanguard 60/40 fund.
The more i read about this area, I conclude the way to really manage risk is to keep a couple of years outgoings in cash and have a letter in the study to my future self explaining why i must not sell out in a panic.
I’ve read a study explaining that keeping a couple years of cash rather than investing it leads to lower returns, but it will help me sleep so im going to do it. Beyond that, every asset class carries risk, but the biggest risk for me is baling out when prices tank 50%.
Its worth reading the Bogleheads forum during 2009. The posts from people who had hung on through 12 – 18 months of losses and then capitulated are frightening.
@SurreyBoy — Do you have any direct links to relevant posts from Bogleheads? Always interesting to read that sort of thing. I’ve often wondered if reports of passive investors bailing out are over-exaggerated, and I’m sure I saw Vanguard data to that effect, though the last time I looked for it I could only find stats from Dimensional Fund Advisors (who actually managed to encourage their investors to put more money to work in the bear market…)
Thanks for this article for reminding me why I should counter my rational self sometimes! I have to reallocate a chunk of money in an ISA for my wife and have spent hours thinking about this and that whilst forgetting the objective was capital preservation first second and third. So through clenched teeth I shall allocate a significant proportion to a 10 year or so IL gilt and leave my oh so significant insights to another occasion.
I liked “have a letter in the study to my future self explaining why i must not sell out in a panic.” I find that going through my old notes and cuttings tends to do the same job, but more time-consumingly.
“capital preservation first second and third. So through clenched teeth I shall allocate a significant proportion to a 10 year or so IL gilt”: but those have negative real yields, guaranteeing losses compared to RPI-inflation.
“but those have negative real yields”: aha, I now see your point. You are hoping that in spite of their defects their contribution to the balance of your portfolio will prove advantageous.
I’m afraid that that reminds me of an argument I saw made when I disparaged that dreadful US warplane, the F-35. Someone argued that it might indeed be lousy but it is intended to be part of a system, and that the system would prove to be good.
Yes indeed. My rational self reckons negative yield is bad. However there are potential scenarios were this unpromising outlook proves a lifesaver. After all are we to think everyone buying gilts on current yields are misguided and they know less than me 🙂
Ps, I work for the company that makes the F35 so you wouldn’t expect me to comment would you 🙂 🙂
“in the UK you can currently get almost 2% for cash that you lock away for three years, provided you choose from the Best Buy accounts.”
That’s not cash. That’s a 3 year bond, with worse liquidity than equivalent corporate or government bonds.
@theta — It’s not a bond, it’s cash in a fixed-term savings account that is termed a ‘bond’ for marketing reasons. Most do impose a lock-up, so I agree about the liquidity though. On the other hand, the value won’t fluctuate over the term like a bond does.
Not only in name, but also functionally it’s a bond, as your money is tied up until expiry, or put differently its duration is not zero as in the case of cash. Therefore its value does fluctuate, except you don’t really see it because you hold to maturity. If rates go up, its value goes down (and then increases more rapidly in line with the new higher rate), and vice versa if rates go down.
Moreover, the point of having cash as part of your portfolio is not only for diversification purposes but also for its optionality, the ability to take advantage of opportunity as and when it arises. With money locked up for 3 years, these savings accounts are arguably worse in this regard than proper bonds.
@theta — Well we’re going to have to agree to disagree. It is a cash account with a fixed term. It doesn’t do many of the things a bond does. The only reason, really, anyone would call it a bond is that banks call it a bond (to make us feel good). As you yourself said in your initial post, real bonds are tradeable securities. A fixed term cash account is not.
It’s value doesn’t fluctuate. It’s cash. Thinking of it as a bond may be confusing you. Legally it’s cash. If the bank went bankrupt halfway through and presuming you were FCSS protected you’d get the cash back. Many of these fixed-rate savings accounts allow withdrawals in case of bereavement etc. When they do, you get the cash back.
Fully agree with your final paragraph that it doesn’t give the short-term optionality of instant access cash. And you’re right that in some cases holding a bond instead (/as well as, which is always the point) may be better. However that doesn’t mean it isn’t cash. It is cash.
Let’s agree to disagree then. [That’s what makes a market anyway 🙂 ]
I agree about the point regarding FCSS protection though, so indeed you don’t have the credit risk of a bond (provided we are talking about sums under the deposit protection limit of course).
I insist on the point about its value fluctuating. Just because you don’t see it (you don’t have mark-to-market) it doesn’t mean that in reality its value doesn’t change. Considering extreme scenarios illustrates this point. If base rates suddenly jump to 5%, having your cash locked away at 2% becomes a bad deal. In effect your position has taken a 10%+ hit, even though you don’t see it in any account statement. Likewise, if rates go to -1% you have made a gain that again won’t show anywhere (except gradually in carry over time until maturity). So the value does change with interest rate changes, just like a bond. But because it’s not tradeable you don’t see it.
And this is another point where I agree with you, the fact that it’s not tradeable is a key differentiating factor compared to a real bond.
It may be just semantics anyway. But the key point that I wanted to make, naming aside, is that with a fixed term account you have the interest rate risk of a bond. As long as this is understood then all is well.
@theta — I’m very happy to agree to disagree, and I’d certainly agree that you have interest rate risk with a fixed rate cash account. That point was perhaps worth noting, for those who didn’t think of it.
So I am not really debating with *you* from here, it’s more because I don’t want other readers to get confused. (My perennial problem as a blog owner! 🙂 )
I still think you’re not stating the situation correctly with this idea of cash losing ‘value’. I think you’re talking about something else, like real interest rates or opportunity cost. However even that is perhaps misleading when it comes to the comparison with bonds, and their mooted desirability over cash, which is where we started.
Let’s think about the three-year fixed-rate cash account. I put my money in at a fixed rate of 2%, say, and I know exactly what return I’ll get. (Assuming FSCS limits haven’t been breached, so certainty of return). This means I know exactly what my money will be worth in three years… initial sum compounded by 2% every year for three years. I can also calculate exactly what I have in £ terms at any point in time, by the same compounding maths.
True, I cannot access that sum — as we’ve agreed and is clear in the fixed rate terms, there’s a liquidity constraint — but that’s a cost of the certainty of holding cash here.
Now compare with a vanilla issue UK government bond that has three years left to run. When I buy this gilt, I can work out (or more likely look up) the exact yield to maturity I will get *provided* I hold it to maturity (and provided I reinvest the coupon). I could if I chose to treat it like the fixed-rate cash account, not look at it for three years, and again I’d know how much money I’ll have from it in three years.
But of course as you’ve noted a bond offers liquidity — so I can also sell up at any time.
The cost of this liquidity is the bond is subject to market pricing, as you know. So at any point between now and maturity, I DO NOT have certainty about what my holding in the bond is worth. The £ value of my investment fluctuates with the market value of the bond. This is not the same as with holding my money in cash. There the £ value does NOT fluctuate.
So finally we get to this concept of the notional value of the cash changing, as in your example when interest rates jump up 5%.
What we’re really saying here I think is that the three-year cash account turned out to be a bad investment (as you say, a bad deal) not that the value of the cash has changed. Rather, there’s an opportunity cost (because I’m locked in and can’t access the new 5% rate) or perhaps we could say the *real* interest rate has gone negative (presuming that the 5% coincided with a rise in inflation that lowered the real interest rate on cash, which arguably does not need to be true just because a better offer shows up, as I’ll touch on below). But that is a different statement from saying it fluctuates in value, as I see it.
Moreover, even if we follow your reasoning, the situation is even more dramatic in the case of the bond. With the bond there is not the ‘implied’ loss of value that you seem to be positing (which I think is describing something else) but rather there will be a very real loss of value in £ terms!
True, the bond’s liquidity might now be more desirable (because I can now sell to access the 5% interest rate). But any implied fall in the real interest rate of my bond would very quickly be reflected in a fall in the market price of the bond. So my holding would soon be worth less in £ terms.
i.e. Who’d want to hold a low-yielding three-year bond when you could get 5% in cash? Nobody. So the bonds would sell off until the yield to maturity rose to reflect the higher rates on offer elsewhere. As you know this process is what happens every second in the bond market.
Finally, there are plenty of times when different deals or opportunities become available, and in a reductio ad absurdum way I think this shows why it’s not helpful to think of the value of the cash in a fixed rate account falling in value because of them.
E.g. My brother comes to me, desperate for cash, and offers to sell me his £5,000 car for £2,000. I know he wouldn’t shaft me — this £5,000 really is the value of the car. But I can’t buy the car, because my cash is locked away. Is that annoying, and regretful? Yes. Has the value of my cash fallen? No. You can think of all sorts of lost opportunities like this, where the macro economics don’t change, and nor does the spending power of my cash.
Anyway, cheers for the discussion! 🙂
Just a thought, as I’ve seen no mention of laddering cash fixed term bonds. This can give better averages and at least partially offset the disadvantages of rising interest rates