Share this page:
Stay connected:
Welcome to the Citywire Money Forums, where members share investment ideas and discuss everything to do with their money.

You'll need to log in or set up an account to start new discussions or reply to existing ones. See you inside!

Notification

Icon
Error

Help understanding risk
jac79
Posted: 12 June 2017 17:09:41(UTC)
#1

Joined: 12/06/2017(UTC)
Posts: 3

Thanks: 3 times
Hello all,

I am new to investing, and am in the process of setting up a SIPP for retirement. For simplicity I have decided to go for a Vanguard LifeStrategy fund, so now all I need to do is determine what ratio of equity to bonds I would like, i.e. the level of risk I can tolerate.

My previous understanding, having read 2 books on investments, was that bonds reduce the chances of losses over the short and long term. But then I came across this tool by Vanguard, which shows the benefit (or not) of moving away from 100% equity to include bonds. It appears to show that bonds only reduce volatility: the stocks never actually dip below the level of return offered by bonds.

Is this particular to this dataset, do you think, or is it usual behaviour?

The reason I'm asking is because the argument against 100% equity is that there could be a big dip in the markets just before you plan to retire, so you could feasibly have to wait another 15 years or whatever to retire until returns get back to normal. According to the Vanguard tool, however, you would appear to be no better off owning a portion of bonds.

Or am I missing something?

Any help much appreciated!

Jon
Stephen B.
Posted: 20 June 2017 15:58:20(UTC)
#2

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
In effect there are two different kinds of risk. What traditionally gets called risk in the financial world is short-term volatility, i.e. the fact that share prices move a lot from day to day, and sometimes drop rather sharply, e.g. you could easily see a 20% fall in the space of a few weeks - indeed if you invest long enough you almost certainly will see such episodes. Fundamentally, if you're investing for retirement and that's a long way away you shouldn't care about that kind of short-term effect. However, you may still find that you're affected psychologically, it can generate a lot of anxiety to see big falls but it depends on your personality. Unfortunately you probably won't really know how you'll react until it happens ... anyway, having a significant weighting to bonds or cash reduces the impact of such falls, but you need to have a big weighting to have a significant effect, e.g. 50% in cash would reduce a 20% fall to a 10% fall.

On the other hand you have a long-term risk related to what you want to achieve at retirement, i.e. how much money you think you'll need. Most people drastically underestimate that - as a rough rule of thumb you need about 20-30 times the annual income you would like. The effect of variation around that are also quite asymmetrical - if you don't have enough to live on that's very bad, whereas having more is nice but not essential. Long-term returns are very uncertain, so the best you can do is monitor how things are going as you go along and adjust your plans if necessary. However, particularly in the current market I think it's almost certain that equities will have a significantly higher long-term return than bonds, so a high bond weighting may well reduce your final amount substantially.

The other thing you raise is what to do as retirement approaches. Traditionally people would cash out and buy an annuity, so it made sense to go progressively into safer investments as you got close to that to protect against a crash at the end. However, annuity rates are now very poor (also partly related to low bond yields) and life expectancies are long, so there's a good case to stay invested in retirement and draw money gradually. In that case you may want to make your investments somewhat safer but it's less critical. It's also easier these days to delay retirement if necessary.
2 users thanked Stephen B. for this post.
jac79 on 20/06/2017(UTC), Mike L on 22/06/2017(UTC)
jac79
Posted: 20 June 2017 17:49:15(UTC)
#3

Joined: 12/06/2017(UTC)
Posts: 3

Thanks: 3 times
Hi Stephen, thanks very much for taking the time to respond—it's really helpful. But I am still confused by the Vanguard tool I placed a hyperlink to in my original post. That tool, which depicts the volatility for different biases towards bonds, appears to show that bonds only reduce the size of the peaks, and leave the troughs untouched. In effect, the message of the graph appears to be: in the worst case, the return from high equity or high bonds will be the same, but in the best case high equity will be better. So who wouldn't go for high equity?

The more I think about it, the more I think the Vanguard tool must be wrong, or at least choosing an atypical dataset as its basis.
King Lodos
Posted: 20 June 2017 18:04:25(UTC)
#4

Joined: 05/01/2016(UTC)
Posts: 1,427

Thanks: 231 times
Was thanked: 1682 time(s) in 719 post(s)
One thing the Vanguard tool's not showing you is how damaging year-on-year volatility *can* be..

- If you lose 10% across your portfolio, you need an 11% return to get back to even;
- If you lose 50%, you need a 100% return;
- If you lose 80%, you need a 400% return.

And obviously if markets go to zero (which has happened in Russia) or a stock goes to zero, you'd need an infinite return to break even.

So while a degree of volatility has to be tolerated, it makes sense to avoid significant losses to capital beyond a point .. Whether this increases your Total Return vs 100% stocks depends on the markets you get, but a definition of risk could be: The probability of a favourable outcome.

For an example of how bad things can get, consider the Japanese stock market .. Second largest market in the world in 1990, yet over the next 20 years it lost 80% .. Anyone who'd stayed long stocks would be very unlikely to get a favourable outcome .. Similarly, in the 19th century, bonds outperformed stocks for over 70 years straight.

Bonds look very expensive today, but so do stocks .. Diversification is trickier than ever .. As a buy-and-hold investor, personally I'd hold a little cash and a little gold too .. It may be low returns are unavoidable .. But one advantage of bonds is that when value flows out of stocks, it usually finds its way into bonds, which at least avoids those large losses to capital that can make a favourable outcome unlikely.
1 user thanked King Lodos for this post.
jac79 on 20/06/2017(UTC)
Stephen B.
Posted: 20 June 2017 18:07:34(UTC)
#5

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
I hadn't looked at the tool, but having done so I don't understand your question. If I move the slider towards bonds it does indeed reduce both peaks and troughs - for example in 1990, with a high equity mix the return is negative but below about 50% the return becomes positive. It won't necessarily happen everywhere because there may be years where both equity and bond returns were negative, but in general it seems correct - even 2008 goes positive with 90% bonds. You should however be aware that bond yields were much higher in the past, so in this case past performance is definitely not a good guide to the future except in a rather general sense.
Stephen B.
Posted: 20 June 2017 18:16:49(UTC)
#6

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
As a response to King Lodos - it's true that catastrophic losses are different. However, if you have a diversified worldwide portfolio that can only happen if essentially every country has some kind of simultaneous disaster. If you think about the kind of scenarios that might produce that, e.g. nuclear war, major pandemic etc, I don't think it makes a lot of sense to try to find financial ways to protect against them, if things are that bad the financial system will have stopped functioning anyway. I think the 2008 financial crisis was about the worst case for a "normal" disaster and equities still recovered fairly rapidly from that.
Stephen B.
Posted: 20 June 2017 18:24:56(UTC)
#7

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
Ah, maybe I see your point, I guess you mean the cumulative return graph. But indeed if you choose a different starting point you can get a different result, e.g. start from 2001 or 2007 and the bond returns are higher, and starting in 1990 they aren't much different.
1 user thanked Stephen B. for this post.
jac79 on 20/06/2017(UTC)
Stephen B.
Posted: 20 June 2017 18:30:43(UTC)
#8

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
In fact starting in 1990 (to date) a 50/50 equity/bond mix is higher than either pure equities or pure bonds. However, 1990 was the start of a long bull market for bonds ...
jac79
Posted: 20 June 2017 18:34:28(UTC)
#9

Joined: 12/06/2017(UTC)
Posts: 3

Thanks: 3 times
Quote:
Ah, maybe I see your point, I guess you mean the cumulative return graph. But indeed if you choose a different starting point you can get a different result, e.g. start from 2001 or 2007 and the bond returns are higher, and starting in 1990 they aren't much different.


Yes, I think I understand now. The point I should be concerning myself with is what could happen, which the Vanguard tool, using a historical dataset, doesn't show. In theory, all the markets could go down in the very long term, it's just that this hasn't happened before. But it could, and that's what I should consider protecting myself against.

I like to think that I'm of a technical nature and should therefore approach short term losses with the cool-headedness they deserve. But I guess, as you say, one can never know until a big loss hits!

Thanks for all your help guys.
Stephen B.
Posted: 20 June 2017 18:47:14(UTC)
#10

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
Well, as King Lodos pointed out it has happened in some places, e.g. Russia and China had communist revolutions that basically wiped out everything. The UK has had a capitalist economy (of sorts) since the 18th century so all historical data will show things turning out OK in the end, but that doesn't necessarily mean it will always be true.

It may also be worth pointing out that unlike equities, bonds have no upside - the best possible case is just that you get the redemption yield, and at the moment that's almost nothing. Conversely the downside is complete loss just as it is with equities.
1 user thanked Stephen B. for this post.
King Lodos on 20/06/2017(UTC)
King Lodos
Posted: 20 June 2017 21:49:27(UTC)
#11

Joined: 05/01/2016(UTC)
Posts: 1,427

Thanks: 231 times
Was thanked: 1682 time(s) in 719 post(s)
The thing with stocks is when the US goes down, most other markets go down too..

The worst drawdown in the S&P500 was about 83% in the 1930s – luckily it was brief, but when they do go down, there's never a guarantee they're going back up .. At that moment, your net worth is depleted – and if you're all in stocks, you can't buy at the bottom, when it might be a generational buying opportunity.

You can diversify with foreign stocks, but then you welcome less regulated markets, fewer quality companies, and less consistent outcomes – e.g. I believe French, German and Japanese stocks have experienced negative real returns over periods from 50-60 years.

This is for me the simple logic of the market:

- A US 30-year Treasury Bond yields about 2.9% today .. So you can have a virtually guaranteed (nearly) 3% for the next 30 years, essentially risk-free (safer than money in the bank).

- Stocks on P/E ratios of 25 (in a very crude calculation: 1/25) give you an earnings yield of 4%.

So it's a risk-free 3% vs a riskier 4% .. This is why I don't think it's entirely obvious that stocks are cheaper than bonds relative to their respective levels of risk (as they shouldn't be) .. Both could experience additional upside in the medium-term if demand for long-term bonds or stocks increased (if fear or greed respectively grows) .. Stocks may be a better bet against inflation, but inflation-linked bonds (or gold) might be a better bet than stocks .. Either way, I think you'd buy bonds today with the intention of rebalancing against stocks – rather than just holding to maturity.

2 users thanked King Lodos for this post.
Jon Snow on 20/06/2017(UTC), Mike L on 22/06/2017(UTC)
Stephen B.
Posted: 20 June 2017 22:18:47(UTC)
#12

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
You shouldn't just be mentioning inflation in passing, over long periods it makes a big difference even at fairly low rates. Shares, like property and even gold, are real assets which over long periods hold their value against inflation - high inflation may be connected to economic problems which reduce the value of shares, but that's a separate issue. So your virtually guaranteed 2.9% is only 0.9% real return if the Fed hits 2% inflation on average. (One might also wonder how guaranteed it really is, given that US debt is now about $20 trillion according to wikipedia and most US governments haven't shown great ability to limit it.) Also, if you buy US bonds you have a currency risk if you're in the UK, and gilts are yielding under 2% so negative in real terms.

In any event when looking at retirement you must take inflation into account - e.g. a £30k income might seem OK in current money, but in 30 years at 2% inflation it would only be worth about £16k, and even moderately higher inflation could reduce it a lot more.
3 users thanked Stephen B. for this post.
Jon Snow on 20/06/2017(UTC), King Lodos on 21/06/2017(UTC), Jenki on 21/06/2017(UTC)
Stephen B.
Posted: 20 June 2017 22:30:25(UTC)
#19

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
Also for shares, it's true that a P/E of 25 is on the high side, but in the long run company earnings should grow with GDP, not just inflation, so the 4% earnings yield is maybe 6.5% real or 8.5% nominal. That is of course uncertain, but if GDP growth isn't reasonably positive over a period of decades it would imply economic and political trouble at a level which would certainly put the safety of gilts in question too.
Jon Snow
Posted: 20 June 2017 22:56:41(UTC)
#20

Joined: 02/03/2014(UTC)
Posts: 981

Thanks: 642 times
Was thanked: 730 time(s) in 394 post(s)
Some very insightful posts here.

All I'd add is a practical example, in my real world SIPP (in drawdown) I hold roughly -

32% bonds, investment grade and high yield
18% UK CTY
32% Far East HFEL
10% Property SLI
8% other GCP

No USA, no tech, no ETFs.

5.18% yield, I draw about 50% and reinvest the rest for the future.






5 users thanked Jon Snow for this post.
King Lodos on 21/06/2017(UTC), Keith Cobby on 21/06/2017(UTC), Mickey on 21/06/2017(UTC), Jenki on 21/06/2017(UTC), Mike L on 22/06/2017(UTC)
King Lodos
Posted: 21 June 2017 00:46:41(UTC)
#13

Joined: 05/01/2016(UTC)
Posts: 1,427

Thanks: 231 times
Was thanked: 1682 time(s) in 719 post(s)
Stephen B.;48112 wrote:
You shouldn't just be mentioning inflation in passing, over long periods it makes a big difference even at fairly low rates. Shares, like property and even gold, are real assets which over long periods hold their value against inflation - high inflation may be connected to economic problems which reduce the value of shares, but that's a separate issue. So your virtually guaranteed 2.9% is only 0.9% real return if the Fed hits 2% inflation on average. (One might also wonder how guaranteed it really is, given that US debt is now about $20 trillion according to wikipedia and most US governments haven't shown great ability to limit it.) Also, if you buy US bonds you have a currency risk if you're in the UK, and gilts are yielding under 2% so negative in real terms.

In any event when looking at retirement you must take inflation into account - e.g. a £30k income might seem OK in current money, but in 30 years at 2% inflation it would only be worth about £16k, and even moderately higher inflation could reduce it a lot more.


Absolutely – which is why portfolios like Ruffer's and Trojan's use just inflation-linked bonds .. Some very long, some short .. (I think the short-duration would be more suitable protection for most investors – they protect duration risk on their long bonds with options .. I think Ruffer Total Return could be a good choice for inflation protection.)

But you do have to start from the assumption markets are correctly priced .. Yields on inflation-linked bonds are generally around 0 or slightly negative – so the amount of inflation we expect is pretty much baked into prices .. To overweight linkers, like Ruffer, you need a macro prediction.

If you look at expected returns on Research Affiliates' site – the broad asset classes: US stocks, treasuries, TIPS, are all estimated around 0% real return .. I think long-duration bonds are still likely to offer diversification, just because there are situations (esp if inflation looks to remain very low) where yields could be bid down significantly.

https://interactive.researchaffiliates.com/asset-allocation.html#!/?currency=USD&model=ER&scale=LINEAR&terms=REAL&_k=w1wt5q


Ludditeme
Posted: 21 June 2017 07:27:24(UTC)
#16

Joined: 14/11/2016(UTC)
Posts: 33

Thanks: 33 times
Was thanked: 25 time(s) in 11 post(s)
King Lodos;48117 wrote:
[quote=Stephen B.;48112]

Absolutely – which is why portfolios like Ruffer's and Trojan's use just inflation-linked bonds .. Some very long, some short .. (I think the short-duration would be more suitable protection for most investors – they protect duration risk on their long bonds with options .. I think Ruffer Total Return could be a good choice for inflation protection.)



I'm interested in people's opinions about bond/defensive investing generally. I currently have about 10% of my SIPP invested in bond (or bond substitute) funds, and have approximately 17% in RCP, PNL, and RICA. My view is that these IT's can expend more research time & experience than myself in making calls on when/where to invest, with a focus on capital protection over outright growth.

The alternative I am considering is the Lifestrategy 20% Equity. Does anyone have an opinion about the merits (or not) of this simple low cost option?

Thanks
1 user thanked Ludditeme for this post.
Mickey on 21/06/2017(UTC)
Stephen B.
Posted: 21 June 2017 08:03:44(UTC)
#14

Joined: 26/09/2012(UTC)
Posts: 91

Thanks: 1 times
Was thanked: 98 time(s) in 49 post(s)
[quote=King Lodos;48117
But you do have to start from the assumption markets are correctly priced ..
[/quote]

Well, that's the point, I don't think that government bonds are correctly priced, yields have been driven down by QE. In more normal times the real return should be something like 2-3%. It may be that US shares are also expensive at the moment, certainly we've had a substantial run-up in prices driven by vague hopes about what Trump might do. However I think there are still plenty of markets around the world which look reasonable. Also even for the US, while the short-run returns might be low if the current P/E were to unwind, I'd still expect underlying earnings growth to follow GDP and hence returns would beat Treasuries in the longer run.

1 user thanked Stephen B. for this post.
Mr Helpful on 21/06/2017(UTC)
King Lodos
Posted: 21 June 2017 08:26:41(UTC)
#18

Joined: 05/01/2016(UTC)
Posts: 1,427

Thanks: 231 times
Was thanked: 1682 time(s) in 719 post(s)
Ludditeme;48123 wrote:
I'm interested in people's opinions about bond/defensive investing generally. I currently have about 10% of my SIPP invested in bond (or bond substitute) funds, and have approximately 17% in RCP, PNL, and RICA. My view is that these IT's can expend more research time & experience than myself in making calls on when/where to invest, with a focus on capital protection over outright growth.

The alternative I am considering is the Lifestrategy 20% Equity. Does anyone have an opinion about the merits (or not) of this simple low cost option?

Thanks


I think the combination of RCP, PNL and RICA is smart .. Hardcore passive investing monks might tell you the chances of those active managers being right about anything is as good as a coin toss – and therefore the only certainty is higher fees .. But, right or wrong, they do risk manage intelligently, and they do diversify you in ways that might be difficult to otherwise achieve.

Yale, Princeton, MIT (some of the most academic endowment portfolios on the planet) often have a 15-20% allocation to these kind of funds – and often very little directly in bonds.

When it comes to Lifestrategy 20, my gut reaction goes back to bonds being expensive and not being very likely to deliver returns much above inflation .. Bonds got nearly this expensive in 1946, and did spend most of the next 30 years delivering close-to-0% real returns .. I kind of prefer Lifestrategy funds when they're more 60:40, as it's the rebalancing that I think makes holding bonds like that worthwhile – and the fact they rebalance systematically means there's a good chance they'll be buying stocks and selling bonds when they should be (when more emotionally driven investors might be doing the opposite).

3 users thanked King Lodos for this post.
Ludditeme on 21/06/2017(UTC), Jenki on 21/06/2017(UTC), Mike L on 22/06/2017(UTC)
King Lodos
Posted: 21 June 2017 08:41:46(UTC)
#15

Joined: 05/01/2016(UTC)
Posts: 1,427

Thanks: 231 times
Was thanked: 1682 time(s) in 719 post(s)
Stephen B.;48125 wrote:
Well, that's the point, I don't think that government bonds are correctly priced, yields have been driven down by QE. In more normal times the real return should be something like 2-3%. It may be that US shares are also expensive at the moment, certainly we've had a substantial run-up in prices driven by vague hopes about what Trump might do. However I think there are still plenty of markets around the world which look reasonable. Also even for the US, while the short-run returns might be low if the current P/E were to unwind, I'd still expect underlying earnings growth to follow GDP and hence returns would beat Treasuries in the longer run.



It has been one of the big question marks in recent years: is the stock market pricing correctly or the bond market?

A surprising thing with bonds is real rates aren't particularly low .. In the 70s, you'd have been buying 10yr treasuries with real yields as low as -7%, and in the 80s around -3% – with coupons as high as 16% .. So if you bought then, looking certain to lose value to inflation, you'd have been securing risk-free >16% returns for the next 10-30 years.

Yields have been driven down, but if markets have been pricing efficiently, stocks and other assets should reprice to adapt .. The Fed Model always values stocks relative to the risk-free rate, and although it's not perfect, I think that explains most market behaviour (what is a stock worth relative to what I can get risk-free?).
Keith Cobby
Posted: 21 June 2017 08:58:31(UTC)
#21

Joined: 07/03/2012(UTC)
Posts: 291

Thanks: 121 times
Was thanked: 380 time(s) in 171 post(s)
I cannot see a big downside to equities while interest rates are so low. And how can they be raised when global debt is 325% GDP. I also cannot see QE being reversed and bonds being returned to the market. It's no good now comparing equity valuations with historical norms. This time is different and unless the debt is inflated away (hard to see how this can happen in a deflationary world with average ages increasing) then interest rates will be low as far into the future as any of us can see. In this new environment stocks are not over valued and have further to go. I am 100% in equity ITs as always.
2 Pages12Next page
+ Reply to discussion

Markets

Other markets