≡ Menu

It’s not fair! Sequence of returns risk

You never know what return the wheel of fortune will deliver each year

With the mindset of a long-term investor, you can avoid a lot of the worries that afflict the frightened hordes. But you can’t really avoid sequence of returns risk.

Oh I know you’re not scared of a stock market crash.

Nor do you pile in at the top.

You have the tough-under-fire attitude of a Vietnam veteran on his third Tour of Duty. When share prices plummet and others quiver before CNBC, you go surfing, Apocalypse Now-style.

“Is that all you’ve got?” you laugh as the market falls 10%.

The average after-inflation annual return from shares globally is 5.3% per year1.

So as long as you sit tight and don’t panic, you’ll be rewarded, right?

Well… yes. Probably.2

The sequence of returns matters

But you’d better know there’s another kind of risk to think about, and it can be a doozy.

Assuming we’re both adding or taking money out of our portfolios over different periods of time, the return you get will be different to the return I get.

This would be true even if we saw the same average 5% real return from our portfolios.

Huh?

I know – it’s counter-intuitive.

It also has a clumsy name: sequence of returns risk.

Sequence of returns risk is the risk that fate will deal you a shocking hand. That the timing of bear markets and bull markets will fall worse for you than for another investor.

This danger is especially high when you’re taking money out of a portfolio during a market crash.

It’s why we’re urged to reduce our exposure to the riskiest assets as we approach retirement age.

Young versus old

Stock market falls are great news for young investors who invest more throughout the turmoil.

The best time to get a bad hand from the market is when you’re starting out as a saver.

You’ve got less money to lose in a market crash. Better yet, what you buy with new money at a lowered cost basis will grow in the good times to come. (Plus you get used to volatility early.)

In contrast, the last thing you should want the day before you retire is to have all your money in shares, only for the market to plummet.

You’re retired. You need that shrinking portfolio to live on.

How to multiply your money

You might not think it matters how the market tosses up its treats and its treacheries.

Returns from investment are multiplicative, after all. You multiply your money!

And every precocious child knows that it doesn’t matter what order you multiply numbers together. You still get the same result.

For example:

1 x 2 x 3 x 4 = 24

4 x 3 x 2 x 1 = 24

3 x 4 x 1 x 2 = 24

It’s exactly the same with investing.

When the market delivers a 20% return, it goes up 1.2 times.

When the market falls 10%, you multiply it by 0.9 times.

1.2 x 0.9 = 1.08

0.9 x 1.2 = 1.08

Okay, so why does it matter to us poor strivers exactly when the sturm und drang of a stock market crash hits us?

Well it wouldn’t if you were a member of the landed aristocracy and you were simply managing a big pile of loot before passing it onto the next generation.

But most of us are saving and investing over our lives to ensure our financial futures. We’ll have to withdraw money from our savings in retirement for spending.

And it’s because we add and subtract money from the market over time – at different times – that the sequence of returns risk can have its wicked way with us.

Here’s one we did earlier

Let’s consider a real world example.

Here are the total returns from the FTSE 100 for the five years from 2008 to 2012:

Year Return
2008 -28.3%
2009 27.3%
2010 12.6%
2011 -2.2%
2012 10.0%

Source: FTSE

Do the sums and you’ll see that’s an average annual return of 3.9% per year.

Now let’s imagine you invested £100 at the start of 2008. Here’s where your money would have stood at the end of each year:

 Year Return Investment
2008 -28.3% £71.70
2009 27.3% £91.27
2010 12.6% £102.77
2011 -2.2% £100.51
2012 10.0% £110.56

Note: £100 compounded for five years, as per the returns listed.

The first thing to note is you’ve ended up with less than you might have expected.

If you plug 3.9% into a compound interest calculator, it will spit out £121. You got £10 less.

Why? Because investment returns are geometric, rather than arithmetic. But that’s for another article…

For now remember we ended up with £110.56 after this five year run.

Investing in Bizarro World

Back to the sequence of returns risk. Let’s imagine you fell through a wormhole and ended up in an alternative reality, five years in the past.

(Stay with me here.)

Being a good Monevator reader, you shrug off your trip through time and space and head to the nearest stockbroker. People always need to save and invest for their retirement, even in Bizarro World.

But things aren’t entirely the same here.

In this alternative reality, the order of the annual returns from 2008 to 2012 are reversed:

Year Return
2008 10%
2009 -2.2%
2010 12.6%
2011 27.3%
2012 -28.3%

Source: Bizarro World Bank Headquarters broom closet.

This time the big crash comes at the end of the five year sequence. Rather than at the start as it did in our reality.

Do the maths and you’ll see the Bizarro World market averaged the same 3.9% return.

But what about a £100 investment?

Year Return Investment
2008 10.0% £110.00
2009 -2.2% £107.58
2010 12.6% £121.14
2011 27.3% £154.20
2012 -28.3% £110.56

Note: £100 compounded for five years on Bizarro World.

Because returns are multiplicative, we end up with exactly the same £110.56 in Bizarro World as we got on Planet Earth.

That was true even though the sequence of returns was reversed.

We expected that. So far so good.

Now add sequence of returns risk

The complication comes if you are saving or taking money from your investment over the years.

Let’s say you add £20 at the end of each year to your portfolio.

The result in our reality on Planet Earth:

Year Return Investment
2008 -28.3% £91.70
2009 27.3% £136.73
2010 12.6% £173.96
2011 -2.2% £190.14
2012 10.0% £229.15

Note: £100 initially invested, then £20 added at the end of each year.

What about in the alternate reality, where the sequence of returns was reversed?

Here you got a different result:

Year Return Investment
2008 10.0% £130.00
2009 -2.2% £147.14
2010 12.6% £185.68
2011 27.3% £256.37
2012 -28.3% £203.82

Note: Again, £100 in, then £20 added each year. Alternative return sequence.

As you can see, you’re left with a different sum. Falling through the trouser leg of time3 and investing in Bizarro World reduced your final total by around 10%.

Now I don’t know how much things cost on Bizarro World. But I’m sure anyone would rather have that extra spending money.

More seriously, this is exactly what happens in real-life to different investors with different saving and withdrawing schedules.

The sequence of returns varies over time. And so two regular savers with the same general strategy but investing over different periods will see different sums accumulated by the end.

Even if they enjoyed the same average annual return.

People who retired in the mid-1990s as the stock market soared were laughing.

People who retired in 2001 in the midst of a severe market decline?

Not so much.

The science bit: As well as the multiplication, we now have addition in our sums. So the order now matters.

Withdrawal symptoms

As I mentioned – and more scarily – all this is equally true when you’re withdrawing money as when you’re saving.

I say ‘more scarily’ because there’s not much you can do about it once you’ve stopped earning.

At least if you see a bear market while you’re accumulating money, you can try to find more cash to invest before you retire. You might even enjoy a market rebound on that extra cash you put in.

Once you’re retired though, you’ve no new money. So maybe you’ll have to spend less and cancel your subscription to Caravan Monthly.

Withdrawal method

Imagine you had £100,000 in 2008.

For the sake of this example let’s say you put it all in the stock market.

You withdraw £4,000 a year.

In the table that follows the third column shows how £100,000 would fare if you kept all your money invested. The fourth column shows the impact of withdrawing £4,000 at the end of each year:

Year Return Hands off With withdrawal
2008 -28.3% £71,700 £67,700
2009 27.3% £91,274 £82,182
2010 12.6% £102,775 £88,537
2011 -2.2% £100,514 £82,589
2012 10.0% £110,564 £86,848

Note: £100,000 in with no withdrawals, versus £4,000 taken out each year.

No great surprise. Taking £4,000 out a year reduces how much money you’re left with, compared to keeping it all invested.

But let’s now turn our telescope to Bizarro World. How does the alternative sequence of returns play out with the same £4,000 withdrawal rate?

Year Return Hands off With withdrawal
2008 10.0% £110,000 £106,000
2009 -2.2% £107,580 £99,668
2010 12.6% £121,135 £108,226
2011 27.3% £154,205 £133,771
2012 -28.3% £110,564 £91,914

Note: Alternate sequence for retirement assets on Bizarro World.

Column three reruns the £100 example. With no withdrawals, the hands-off portfolio of £100,000 compounds to the same £110,564 in both instances.

However in Bizarro World with £4,000 per year withdrawals, its sequence of returns proves more favourable for the retiree.

She ends up with £5,000 more in the pot in 2012 than her spirit sister here on Planet Earth.

Hard luck

Interestingly, the result for the retiree on Bizarro World is the opposite of what we saw with its regular savers. Savers did worse there over the same five-year period than we did.

But let’s not get hung up on these specific numbers.

The point is that sequence of returns makes a difference to how your retirement plays out. But we cannot know what returns we’ll get in advance.

Which makes it an especially thorny problem.

Don’t risk doing badly

Can you do anything to sidestep the sequence of returns risk?

Not a lot. Its impact is mainly down to luck.

You might try to guess if various markets are cheap or expensive. You could try to shift your investments accordingly.

But many people – probably most – will do worse using such active strategies than if they had just saved and rebalanced automatically.

I think the main response to sequence of returns risk should be:

  • To de-risk your portfolio by rebalancing towards safer assets as you approach retirement

All investing involves risk. But by diversifying your portfolio and playing a bit safer, you can reduce the role of luck, and increase the odds of your plan working out.

Next week we’ll consider what to do if you think sequence of returns risks is about to savage your retirement. Subscribe to make sure you see it!

  1. Source: Credit Suisse Global Yearbook 2022 []
  2. A handful of markets, such as pre-communist Russia, have been completely wiped out. Never put all your eggs in one basket. []
  3. Apologies to the late Terry Pratchett. []
{ 46 comments… add one }
  • 1 oldthinker June 13, 2013, 9:53 pm

    A simple way of explaining this effect would be by looking at several charts that all start at the same point and arrive to the same point but take different routes there. If you keep accumulating, you want the chart representing the return from your portfolio to stay as low as possible for as long as possible, so that you keep buying low.

  • 2 freebird June 13, 2013, 9:59 pm

    Perhaps a simpler illustration is to imagine a 30 year return where all of the change is concentrated in either the first year or in the last year. If share prices are cut in half, the early drop means losing only half of your first installment while the late drop loses half your nest egg. It works in reverse as well, an early pop means profit on only your first installment. It’s difficult to convince a young person that the worst thing that can happen is a large early gain, assuming constant long-term returns.

    The advantage from dollar cost averaging can be mirrored when cashing out during retirement if you switch to share proceeds averaging. In other words sell a fixed number of shares every year. The downside is fluctuating income, a concept that I imagine is familiar to most who work in private industry, especially at the higher income levels. For the uninitiated this means setting aside an emergency cash reserve to buffer the lean years.

    I don’t see a straightforward way to mitigate the sequence of returns problem at low cost. One can rebalance every year to leave only one installment exposed, but in a growing market, this is a recipe for severe underperformance. Some amount of rebalancing is certainly appropriate, however, because I see giving up wage income with a 100% equity exposure as far too risky.

    In my own case I have kept a 50/50 balance throughout my life– and made up for my underperforming early years by doubling my savings rate. The way I see it I gave up some frivolous spending in my youth to sleep better in my middle age, and I must say it was well worth the price.

  • 3 dearieme June 13, 2013, 10:02 pm

    In my years of being interested in equities I scored as follows.

    1989: started buying equities, mainly investment trusts in PEPs and a pension.

    1999: equities having gone mad, sold the lot, buying gilts and convertibles. Excellent decision.

    2003: in the Spring read that equities were cheap but did nothing about it because my attention, time and energy were exhausted by the consequences of illness and death in the family. (I imagine that this sort of thing is a known curse of DIY investment.) Failure to reflect and act proved costly in terms of opportunity lost.

    2008: decided that the condition of developed economies would justify further falls so didn’t buy equities. Obviously I completely underestimated the effect of government monetary policies, so perhaps an opportunity lost. But somewhere in there I put a fair lump into the Physical Silver ETF in an ISA which has done very nicely indeed, even now. So that compensates for not buying equities.

    2010: bought a Physical Gold ETF
    2013: sold it to help pay off the mortgage.

    If the market would obligingly crash as our Cash ISAs mature, I’d expect to buy equities again. Have not the first idea when to sell the silver ETF. If don’t know, won’t trade.

    The annual sums involved have been modest but they do add up.

  • 4 Greg June 13, 2013, 10:22 pm

    As you’ve touched on, the current generation who are retiring now had a huge bull market for stocks while they were in them when they were younger, then a huge bull market in Gilts when they were in them when older. (In addition to being able to get a job easily in the aforementioned stock bull market and safely bedded in when things went pear shaped.)

    Couple that with running up an amazing level of public debt and the same again of off balance PFI debt, as well as the fact they hived off all our public industries and used all our gas, they might have a bit more sympathy with us youngsters.

    Oh wait, there’s more! Want to educate yourself so you won’t be made ‘not economically viable’ by people in emerging markets, or stuck doing tough, menial, poorly paid work? Why that will be £30,000! Want to live in a gloomy ex council flat in London? Why that will be £350,000 please!

    In Camden, not exactly the fanciest place in the world, median house price divided by median salary is sixteen. Sixteen!
    http://www.retirementinvestingtoday.com/2013/05/valuing-london-property-at-borough-level.html

    I suspect that the people retiring now will have the best deal of any retirees for the next 50 years. (And then we’ll have global warming to worry about!) Yet they still aren’t happy!

    (I realise the ones that are fuel poor probably aren’t the ones that messed everything up, but that makes it worse, not better.)

    :@

    More related to the article, the volatility is such that one might massively under-perform for one’s entire life. The range of final returns produced by a simple Monte-Carlo is amazing. Enough to make a mockery of the nice, smooth curves in all the guides!

  • 5 Grumpy Old Paul June 14, 2013, 9:08 am

    @Greg,

    I didn’t vote for “running up an amazing level of public debt and the same again of off balance PFI debt, as well as the fact they hived off all our public industries and used all our gas” and opposed all of those policies at the time when they were very fashionable. So please don’t blame me Guv!

    However, I am appalled by the selfishness and lack of sympathy by my contemporaries and elders to the plight of younger generations. My hackles rise whenever I hear an ignorant old bigot say “I’ve paid taxes all my life”, claims to be entitled to free everything and still whinges about being hard done by. Typically, the total of that individual’s lifetime tax and NI contributions probably cover just the cost of a few years of their State Pension and NHS usage.

    I’m rather surprised that there hasn’t been more inter-generational tension.

    @The Investor,

    Good article making points I’ve never seen before and highly relevant to an old buffer like me planning to run down my savings and investments during a dissolute retirement! I’ll really have to give this some serious thought.

  • 6 vanguardfan June 14, 2013, 9:28 am

    Hi
    A really pertinent post on a topic that gets far too little attention. When you think about it, the idea of basing any projection of investment returns on simple linear assumptions, as thought they behaved like bank deposits, is so misleading its amazing that the FCA allows the industry to get away with it.
    A probabilistic approach is necessary, such as the FIRECalc model (though based on US data so will give an overly optimistic picture for UK residents) or the data in Tim Hale’s book, looking at the expected frequency of returns over all the possible 10 or 20 year periods for which we have data.
    Its salutory to note that even the best laid plans cannot ever be certain – we cannot control (or even have much idea) whether our particular 20, 30 or 50 year investment period will turn out to be favourable.

  • 7 Curious-Sarah June 14, 2013, 9:57 am

    Tell me why do I only ever read about things like this on Monevator?

    (Thank you I am not complaining — just saying I never read about this once in years of reading The Motley Fool books or website etc).

    Well another thing to worry about then. I mean another thing to overcome! 🙂 🙂

  • 8 Monk June 14, 2013, 10:00 am

    I remember sitting opposite an irate pensioner on the long seat on the number 48 bus on route to Victoria way back when I was tender of age. She was criticising the length of skirts worn by the majority of the mini skirted female passengers while indifferent to the gaping chasm between her knees that foisted sight of her drawers on the unwilling sets of eyes seated on the other side of the bus.

    There’s always been inter-generational tension.

    And claims that we’ve/they’ve/he’s/she’s never had it so good regularly feature as far back as I can remember.

    The blame game is a long and arduous one and one that’s never really produced an outright victor. It’s one of those games that’s best avoided, a bit like find the lady where you’re convinced you know you’re right long after the deck has been packed away and you’re out of pocket.

  • 9 L June 14, 2013, 10:14 am

    An important and comprehensive post TI.

    I think there is one thing you can add. Namely, the importance in saving towards a large savings/investment pot and then prudence in the addition and drawing of cash.

    Let me explain. If you keep your withdrawals constant, as your pot becomes larger each withdrawal becomes a smaller the proportion of your net worth. This is importance as this limits the exposure of your total net worth to the sequence (arithmetic compounding) of returns.

    In the most extreme example, take a billionaire who withdraws only £1 each year. The effect of the above is tiny. However, imagine that that billionaire decides to withdraw all(!) of his net worth into cash at one point in time. The effect is massive. Can you time exactly when to do this? Empirical evidence says it’s difficult, and why take the risk? Even though this is an extreme example, it’s not necessarily far from reality. For example, the 25% tax free pension withdrawal. As always, it’s important to plan ahead. It won’t solve the problem. But it will make it less of a problem.

  • 10 Luke June 14, 2013, 10:21 am

    ‘People who retired in the late 1990s as the stock market soared were laughing.

    People who retired in 2003 after several steep market declines?

    Not so much.’

    Very true – I find one of the most annoying trends in personal finance blogs being an almost universal failure to acknowledge luck. There are quite a few names that spring to mind.

    It’s one reason why I find the Monevator blog a breath of fresh air – the Investor admits his mistakes (I assume the Accumulator has made none 😉 )

  • 11 dearieme June 14, 2013, 11:51 am

    “they hived off all our public industries”: “public” my eye, they were government-owned industries. Flogging them off was wise.

  • 12 The Investor June 14, 2013, 12:39 pm

    Hey chaps, can we please stick to the topic for this post. It’s an interesting discussion but I’d rather the comments were focussed on this subject rather than politics. Much appreciated!

  • 13 Paul Claireaux June 14, 2013, 1:00 pm

    Good piece.
    Yes returns are indeed path dependent.
    Very people know this and it’s worth knowing.
    BTW this issue is also behind the flaws in that famous Dalbar survey

  • 14 Colin June 14, 2013, 1:32 pm

    Arithmetic average is never appropriate for this type of calculation , only geometric is correct, which rather defeats the first point of the article.

    Secondly if you have a good run of returns you want to do it when you have lots of assets , if you are adding funds over time, then you get the maximum return if you have a lot of money, conversely if you intend to retire with your assets in shares, and why not , you may live longer in retirement then work, by derisking you lose the opportunity of getting the maximum benefit from stock market rises. If rackets double I want it in year 30 not year 1

  • 15 BeatTheSeasons June 14, 2013, 1:52 pm

    It’s a long time since I studied mathematics, but I think there’s something wrong with these calculations. Surely you can’t just add up percentages and multiply by the number of years?

    I think the issue is that you’re using a set of periods with arbitrary length of 365 or 366 days. If you moved to Mars (and found a wireless connection) you might start tracking your investments in 669 day periods to match Martian years. That would give you a different set of percentages but the same total return.

    I agree that the timing of capital inflows and outflows makes a difference, but isn’t that just the same as saying your return is a weighted average according to how much you owned and when?

    Seeing as there is no way to predict when the ups and downs will happen, this is surely why it’s recommended to use pound cost averaging as Freebird mentions.

    And given that the long term direction is hopefully upwards there is truth in the mantra about ‘time in the market’ rather than ‘timing the market’.

  • 16 Greg June 14, 2013, 2:04 pm

    One plot I’d like to see is a fan plot of the returns over a ten-year period starting over each of the previous 50 years or so. (Giving 50 little lines, which might have to be merged into a density plot to be clear). I realise finding an index that has been around might be tough.

    The same could be done for various indexes: Stocks, Gilts, cash, cash at rates that small savers can get, various mixes of the above and even a changing mix (e.g. 90-10 the first year falling each year.) Including annual contributions would be another nice addition.

    That would prepare people for the (scary) investment journey ahead of them quite well.

    Heh, I might actually do all the above at some time in the medium future, unless you fancy it?

  • 17 Greg June 14, 2013, 3:01 pm

    @BTS

    There are two main points in the article:
    1) If you had one big lump sum and thought you could find out the returns by taking the arithmetic mean of the yearly return and then multiplying by the number of years, you are wrong. For annualised returns you must use the geometric mean. (This is the main difference between CPI and RPI, not housing as the dim-witted media go on about.)

    As TI touches on, this is easy to understand if one of the years is -100%. Imagine the following annual returns:
    +25%, +25%, +25%, -99.99999%, +25%
    An arithmetic mean would suggest you are break even, but clearly, you have effectively lost the lot.

    2) Most people don’t just have one lump sum. clearly, if you are adding money each year, the later years matter more! Bizarrely, most guides just don’t cover this!

  • 18 The Investor June 14, 2013, 3:47 pm

    @BTS — I see Greg has kindly replied, but anyway I’m not doing what you’re saying to compound those returns. Perhaps I wasn’t clear, as I’m not sure where you’ve got that from. (Perhaps the aside about the 3.9% arithmetic average?).

    You’re right though that the point of the article isn’t spectacularly deep! 🙂 It’s just massively overlooked and misunderstood.

    Was it you and I who had the big discussion about dollar cost averaging? I forget. Anyway as I’ll touch on in part two, DCA-ing doesn’t exactly beat sequence of returns risk, although it would flatten the range of possible outcomes (good and bad). Academic research shows that over more periods than not you’d be best off just investing a lump sum right away if you’ve got one if you wanted the best chance of the best return.

    The reasons to DCA are mainly emotional/psychological (and those are very good reasons!) and also to potentially avoid rare and catastrophically bad outcomes — there may also be certain instances when one is able to see it’s a terrible time to invest a meaningful lump sum. (e.g. 1999, but is that only with hindsight? By definition yes for most people, since most investors were fully invested, which is why we had a bubble in the first place…)

  • 19 oldthinker June 14, 2013, 6:14 pm

    @The Investor

    > Academic research shows that over more periods than not you’d be best off just investing a lump sum right away if you’ve got one if you wanted the best chance of the best return.

    This is an answer to the wrong question. For many (most?) investors the goal is not to maximize the mathematical expectation of their return but to maximize the probability of achieving their target return, because their financial security in retirement is dependent on achieving it. This is a very different goal, especially when the target is reasonable, and pursuing it may require a different strategy. For example, I would not take a chance of overshooting my target by 20% if the downside was an equal chance of undershooting the target by 10%; if I was interested in maximizing the expected return only, I would, of course, take such an opportunity.

  • 20 The Investor June 14, 2013, 6:39 pm

    @oldthinker — I agree, and allude to as much in the line after the one you’ve quoted. 🙂

    More thoughts here: http://monevator.com/risk-and-investment/

    But I was replying to a comment about sequence of return risk, and DCA being a potential solution. It’s not really, IMHO.

  • 21 Paul Claireaux June 14, 2013, 7:06 pm

    To The Investor
    I agree that for the regular investor / saver their returns are highly path dependent. We should all be aware of this when reading the oft quoted surveys about how returns are lost through timing mistakes (Dalbar)

    And to old thinker (from another old thinker)

    I take your point about meeting our goals being more important than achieving some grand total return.

    But I’m not sure how anyone can go about maximising the probability of achieving their target return without taking a view on valuation.

    You can use individual gilts to guarantee nominal returns but what if inflation in the cost of your target starts to go through the roof? You will then want to be out of that gilt and into something with better inflation protection properties. Trouble is at that time it’s too late and your gilt holdings are already smashed.

    I’m not sure there’s any simple answer to investing but keeping an eye on valuations is a good starting point w.r.t. protecting what you have.

  • 22 ermine June 15, 2013, 9:42 am

    @Greg > clearly, if you are adding money each year, the later years matter more!

    I agree with that, but everybody else says the early years matter more due to compound interest 😉

    My experience of lifecycle saving was that it was really hard at the start and got much easier later on, and tax advantages etc were better later on (because of the 40% tax relief making savings punch much higher later).

  • 23 The Investor June 15, 2013, 10:24 am

    @ermine — Hi! 🙂 Remember that “matters more” and “will be hit by down years in the market more” are two different things. If you are “all in” equities, at one extreme, when you are 60 and want to retire at 65, then the later years clearly matter an incredible amount and you are running a large degree of risk. But if you are 60 with say a 60/40 diversified portfolio minimum (and ideally something with cash, property, maybe a small amount of gold, etc) then you have a different risk profile.

    You can spend hours messing about with spreadsheets to model all sorts of scenarios. I suspect there have been periods when investing tuppence (figuratively speaking) in the market at 20 and investing hard for 10 years then switching to cash or bonds for the rest of your life saw compound interest lift you clear. Other times dithering about with the stock market for a decade badly lagged cash (say an index investor looking at their results in 2010, though dividends would have helped, and as we’ve discussed in the article and in these comments a regular saver throughout the decade would have been buying at the bottom as well as the top…)

    Bottom line, it’s a *risk*. By definition, risks can’t be totally accurately modeled, and there’s no one size fits all rule except in hindsight.

    Indeed I sometimes think that’s where more scientific / mathematically inclined and brainy people go wrong when investing. They yearn for certainty, and move from one failed relationship with an asset class or strategy to another. A messy orgy of all the above is the best way to mitigate risk! 😉

  • 24 Paul Claireaux June 15, 2013, 10:44 am

    To The Investor.
    Very fair point there – about the mathematically minded (me included) who definitely do yearn for certainty.
    It’s only in recent years that I’ve learned there isn’t any way of having it – or anything close to it.
    Yes, a mix of assets is best once you’ve accumulated something -but with the most informed tilt you can give it towards better value markets.
    I like CAPE for this but we need to go easy on that metric too or we’ll end up filling our boots with Greek shares! And we have to remember that when economies truly collapse there is a really risk of communist government confiscating the lot!
    For younger savers – who have their contingency funds in place – i think pure equity investment is just fine – but again try to accumulate in markets with better growth prospects longer term.
    My personal feeling is that the USA is already on very stretched valuations and we could see a long Japan style decline there so I’d not allocate much there right now.
    Maybe after a big correction though . . .

  • 25 SemiPassive June 15, 2013, 11:58 am

    Even plumping for markets with “better growth prospects” is flawed, look at the lousy performance of emerging markets indices compared to the “busted, indebted” Western markets over the last year.Maybe an emerging markets tracker will trounce a world tracker over the next 20 years, but there is no guarantee.
    And at the other end of the age scale, “de-risking” into govt bonds is anything but.
    I think in terms of what I hold in SIPP and Shares ISAs I will probably stay at 60% equities (rest in index-linked gilts and corp bonds) forever now.
    Any reduced exposure to the stockmarket will probably be achieved
    by gradually increasing the % of net worth held outside of SIPP, eg cash and property, and maybe even a little gold rather than ramping up and down equity exposure within my pension wrapper itself.

  • 26 The Investor June 15, 2013, 12:16 pm

    @SemiPassive — For somebody who holds zero government bonds, I seem to spend an awful lot of time talking about why they’re not as bad as the worst critics say.

    You write:

    And at the other end of the age scale, “de-risking” into govt bonds is anything but.

    Au contraire, my friend, moving equities into gilts is absolutely de-risking your portfolio.

    Have a look at this table of UK gilt yields. Look at the “yield” column on the far right, and ignore the perpetuals (undated) bonds at the bottom.

    What you will see is that every single gilt has a positive “yield to maturity”. This means if you bought and hold to maturity any UK government gilt, you would get all your money back, plus a bit more.

    Are the yields pathetically small? Yes, especially at the short-end.

    Are any of those with less than 10 years to run to maturity likely to beat inflation? Nobody knows, but I’d say the chances are not.

    Are they less risky than an index tracker fund that can and will at some point fall at least 25% in a year and maybe as much as 50%?

    Absolutely. Swapping equities for gilts absolutely reduces your risk.

    Some people seem to equate “less risky” with “more attractive”. I agree government bonds are not more attractive right now — personally I’d say far from it. I hold none, and expect equities to grossly outperform over the next 10-20 years.

    But government bonds ARE less risky.

  • 27 SemiPassive June 15, 2013, 12:35 pm

    I was thinking in terms of gilt funds at todays yield/valuations, but if you’re talking holding individual gilts for their remaining duration then I see what you mean – inflation risk aside.
    I intend holding some individual corp bonds for their duration, but for the index linked gilts (which rightly or wrongly I see as less risky) I use a fund.
    On a kind of related note I took advantage of a fixed term 1 year cash offer at 2% in my H-L SIPP last October, instead of buying conventional gilt fund due to my aversion.

  • 28 The Investor June 15, 2013, 8:06 pm

    @semipassive — Nice move on the SIPP cash IMHO. I agree holding cash is a decent substitute for bonds for private investors (not an exact swap, but on today’s tiny yields not far off except at the longest and riskiest end! 🙂 ).

  • 29 oldthinker June 15, 2013, 10:23 pm

    @The Investor

    > Au contraire, my friend, moving equities into gilts is absolutely de-risking your portfolio.

    This depends on what kind of risk we are talking about. I am predominantly interested in the risk of not achieving the target return. My target is realistic but absolutely not achievable by investing in gilts only (I would not be invested in equities otherwise). So switching from equities to bonds would, beyond a certain point, increase the risk that matters to me. If the uncertainty of the outcome was the only risk that mattered, moving the whole portfolio into physical cash and then burning it would be a jolly good way of de-risking 🙂

  • 30 dearieme June 17, 2013, 4:45 pm

    I advocate opportunism. When Index-Linked Savings Certificates were available from ns&i, “everyone” should have bought some: they offered a unique combination of security, real return, nominal return and tax efficiency – so much so that they would obviously be withdrawn from sale eventually. And they were.

    Similarly, quite recently Building Societies were offering fixed interest Cash ISAs at excellent rates, with not too onerous penalties for early encashment – my wife opened a five-year one paying 6% p.a. That opportunity too seemed unlikely to last long, and so it proved.

    Such investments don’t feature in the usual discussion of cash vs equities vs bonds, and they aren’t available to the fund managers who compete vs benchmarks, but by golly they are handy for the private investor of modest means.

  • 31 Troy rieck August 4, 2013, 7:35 am

    If you want to look at someone doing serious work in this area of finance, look for Prof Michael Drew at Griffith University in Australia.

    The underlying point is very insightful precisely because it isn’t rocket science, just completely absent from conversations between almost all investors and planners until the last five years or so.

    Investors should reduce their exposure to valuation risks over time (as their cash flow profile turns from inflows to outflows). They should be sensitive to valuations at any point in time as well – own more equities when thir prospective returns are generous compared to when they are miserly etc. And consider the costs and benefits of certain forms of tail risk hedging strategies: valuations risk for assets, inflation risk, longevity risk etc – you don’t have to use them, but you should be aware of the costs of buying greater certainty over time.

  • 32 Planner September 5, 2013, 5:58 pm

    You can completely eliminate sequence of returns risk in the spending phase by spending the same percentage of your remaining portfolio balance every year. You can see the math at my blog post. “Clarifying Sequence of Returns Risk”. Just Google it.

  • 33 The Investor September 6, 2013, 8:52 am

    Hi Planner — Just read your interesting post. I think a better word would be to say can ‘replace’ or ‘substitute’ the risk, and in my opinion that doesn’t even really cover it.

    I do understand what you’re saying and I do see how the maths works. As you say in your post there’s an element of semantics or opinion here about what the actual risk is.

    A retiree who spent 4% of a larger number one year and 2% of a smaller one the next is clearly falling foul of risks of some sort, we can both agree! 🙂

    Anyway, I’ll try to include your article in my weekend reading post tomorrow as it’s good food for thought.

  • 34 Balanced Planner December 5, 2018, 6:54 pm

    So we’re 5+ years on from this post and the comments.

    How’s everyone’s retirement going? Has anyone changed their views or found a workable solution?

  • 35 The Investor March 15, 2022, 10:37 am

    The post above has been updated, so comments above here will reflect earlier times.

    On that score welcome to 2022! Things worked out okay after all, eh? 😉 (And no, government bonds weren’t a disaster…)

    I’ve already written the follow-up to this piece, which is inspired by the current market maelstrom. But I linked back here and then had a tidy, and decided I liked the post so much I decided to re-up it for those who arrived after – gulp – 2013…!

  • 36 Jon March 15, 2022, 5:24 pm

    A lower SWR can help mitigate sequence of returns. I myself will be using an income & capital strategy. Income from single Paid off rental, Dividend Growth Portfolio of shares, VHYL. Capital from a standard type of asset portfolio. When the Capital portfolio declines in a particular year, I will live off Income portfolio so I never deplete the Capital portfolio.

  • 37 xxd09 March 15, 2022, 6:49 pm

    As an old retiree (75) who retired at 57 in 2003 I was “saved” by John Bogle of Vanguard fame who suggested that the percentage of bonds in your Asset Allocation should equal your current age
    I finally settled on a 30/65/5-equities/bonds/ cash allocation which I have maintained through thick and thin for many years
    Obviously the amount saved and the withdrawal rate chosen are also crucial but a stockmarket crash at the beginning of retirement when all your earning power has gone and you need to withdraw cash to live on could do so much damage to a portfolio that it might never recover
    A scenario to avoid!
    xxd09

  • 38 uhm March 15, 2022, 8:31 pm

    Just about to retire, as from 1st April – I feel a bit of a fool now, but it has been a 5 year process and a bit too late to turn back now.

  • 39 AndyP March 16, 2022, 12:54 am

    Very interesting. I think that in retirement one could employ a sort of reverse dollar cost averaging to mitigate the effects of SRR. Just looking at the annual stock market returns hides the shorter time scale fluctuations. Say your financial year ended in March 2020, then you’d show something like a 30% drop in value for the year. However, if you withdrew your fixed living expenses monthly throughout the year up to March 2020, you’d only have a couple of months at the lower valuations and you’d experience a much smaller impact to your sequence of returns. Since most stock markets relatively quickly returned to their pre-crash values, the impact of the crash on monthly withdrawals would have been fairly modest. Any sort of emergency fund would have completely mitigated this type of short term crash. I guess a long period of low returns at the start of retirement would be much harder to deal with than the headline grabbing flash crashes.
    Flexibility is also key I think. Even a part-time job at minimum wages in the early years of retirement could make all the difference should investment returns be hit.

  • 40 jon March 16, 2022, 7:15 am

    Agree with Andy P. Remember a part time job earning £1,000 per month is equivalent to having an asset of £300K earning 4%.

  • 41 David March 16, 2022, 8:57 am

    What a great post. We have just started withdrawing an income from our investments and wanted some idea of what our capital pot might look like in 20 years. Our withdrawal rate is very low – just over 2%. The general assumption is you should be able to take out 4% and with a fair wind, your capital should be intact at the end of the reirement period. I set up a simple spreadsheet using Monte Carlo Simulation to generate returns over a 20 year period. (There are very helpful videos on You Tube).
    This factors in sequence of return risk and for us the results were quite sobering. We have just a 36% chance of maintaining (or increasing) our capital in real terms. Thankfully after running the model 100 times, we never used all our savings, but it was very close.
    Sequence of return risk is something that should be made more expicit. You cannot just assume that you will make the average return every year.

  • 42 Gentleman's Family Finances March 17, 2022, 3:45 pm

    Wow – that’s a great article that sums up what is quite complex for some to understand even visualise.
    Psychologically, we see a drop in prices as a bad thing but if you are young and saving, it’s the best opportunity you can get to increase your long term wealth (presuming you can save your job and keep earning!)

    I’m now at the point in my life where work income matches outgoings – and possibly stopping work for myself/the Lady which means we won’t be net buyers of assets and instead will either sell or eat our dividends.
    SORR is a risk if you go full cold turkey and stop all forms of work – but who really does this?
    If you really want to be retired or to quit the rat race – you need more courage than a spreadsheet can give you!

  • 43 Hariseldon March 18, 2022, 9:27 pm

    I lived this time frame!! retiring in November 2007 at 49. It was uncomfortable, as I was almost 100% equities, however it worked out by some judicious adjustments to the portfolio, market falls provide opportunities…

    I don’t think I would recommend this approach to my younger self if we could have a chat and the Investor provides an excellent article to encourage people to think about this.

    It’s worth considering that the risk does not go away, 15 years on I am still retired but still a few years away from the state pension age, the problem can reoccur, the last few weeks …maybe it has ! One could consider that financing retirement is always restarting every year, you have a lump sum that has to last (hopefully) a long time.

    My earlier 2007 experience of sequence of returns risk ended happily, my portfolio adjustments went well but that could be pure luck rather than good judgement.

    Look forward to reading the follow article, my thoughts are to have ‘safe’ assets to manage a few years living costs, be prepared to adjust spending in difficult times.

    The good news is that having survived the first sequence of returns incident and subsequent good years, you build a bigger pot, one can ‘afford’ to hive off that pot of safe assets to cover down years.

  • 44 CMacK March 19, 2022, 10:38 pm

    Thanks for this post Investor. If anyone REALLY wants to go down the rabbit hole on this one there is a US based (German national) blogger called Big ERN (the blog is Early Retirement Now) who has written a multi-part series on the Safe Withdrawal Rate in which he does a lot of analysis on sequence of return risk and potential strategies to mitigate it. As someone who had to read the maths about 4 times in the post above before it made sense to me, big ERN’s maths (he has a PHD in Economics, was an Asset Manager and trades options for his retirement money) in many of his posts gets very technical for me, but there is some really great stuff in there. I’m not close to drawdown yet, but i will absolutely go back and read the entire series from about 5 years out.

  • 45 Dave S March 20, 2022, 12:24 pm

    I agree that this is an important and under-discussed subject. One thing I’ve never seen discussed, though, is the effect of SoR on the entirety of the period covering both the accumulation leading up to retirement and the decummulation period afterwards.

    Simplistically, someone suffering the worst sequence from the point of retirement will most likely have enjoyed an excellent sequence in the period prior to that, and will therefore have a bigger starting pot than would otherwise be the case. In contrast, someone else may have had a lousy sequence running up to retirement, so would be starting with a smaller pot, but would then be much more likely to enjoy a good sequence during the early years of retirement.

    I would therefore expect the ‘before’ and ‘after’ periods to work together to smooth out some of the variability in outcomes. But that’s not much more than a hunch, and I’d love to see it explored in detail.

    I raised this with TA in an article last year, and he pointed me at the calculator on ERN (which needed a few tweaks to support a period of accumulation followed by the usual decummulation). I spent some time running scenarios to explore my own situation (around 7 years to retirement), and I think it supported my hunch, though perhaps not to the extent I’d hoped. But I found it complicated to do, and so I’m not that sure of my findings.

  • 46 xxd09 March 20, 2022, 1:16 pm

    I would argue that the “elephant in the room” is that the retiree once committed cannot afford to fail-the consequences would be disastrous
    Sequence of Returns are important but very secondary to maintaining the integrity of the Retirement pot
    Balancing Risk and Growth is very personal but unless you are a multimillionaire a retiree approaching retirement with a large percentage of equities in his retirement portfolio has arguably not saved enough
    xxd09

Leave a Comment