Donate to Remove ads

Got a credit card? use our Credit Card & Finance Calculators

Thanks to johnstevens77,Bhoddhisatva,scotia,Anonymous,Cornytiv34, for Donating to support the site

Sequence of returns risk?

Including Financial Independence and Retiring Early (FIRE)
kempiejon
Lemon Quarter
Posts: 3488
Joined: November 5th, 2016, 10:30 am
Has thanked: 1 time
Been thanked: 1145 times

Re: Sequence of returns risk?

#439750

Postby kempiejon » September 4th, 2021, 11:25 pm

Hariseldon58 - cheers for the reply, good to hear from a real life worked example as it were.
I'm juggling with similar sorts of numbers and multiples for spending and equity portfolio sizes. I hope I'm already FI and would like to stop permanent full time employment in a couple of years. It's nice to hear my theoretical isn't too far adrift, first glance I thought 6+ years insurance in cash-like is quite a bit bigger than I was thinking about but 3 years of comfortable that's in the ball park. I'll look at my minimal, I have figures as I did the exercise following redundancy a few years back and do update the numbers to stay current. Spending twice that seems more profligate than comfortable but I guess lifestyles and spending styles are very different person to person and without the restrictions of work I might need more money to fill my time comfortably...
Ta for sharing.

xeny
Lemon Slice
Posts: 450
Joined: April 13th, 2017, 11:37 am
Has thanked: 233 times
Been thanked: 154 times

Re: Sequence of returns risk?

#440551

Postby xeny » September 8th, 2021, 10:13 am

Itsallaguess wrote:There's clearly a number of different ways that we might look to counter Sequence-of-Return risks, and there's a good 4-page article here that goes into some depth regarding how important it is to consider the long-term effects of not doing so -

https://www.ftadviser.com/pensions/2020/07/27/how-to-manage-sequence-of-returns-risk/?page=1


After some thought, I think the graph on the first page of that article is misleading. It presents the two retirees as having equal wealth, just starting their retirement at different times.

1973 was the start of a significant bear market, so if the second retiree started 1974 with the same size portfolio that the first retiree entered 1973, then the extreme likelihood is that it was a substantially larger portfolio at the start of 1973, so it's not unexpected that it would last significantly longer.

I'm not saying SoR doesn't exist, but that graph would be more "honest" if retiree 2 was shown starting with a portfolio after it had been impacted by the losses of 1973.

Alaric
Lemon Half
Posts: 6033
Joined: November 5th, 2016, 9:05 am
Has thanked: 20 times
Been thanked: 1399 times

Re: Sequence of returns risk?

#440572

Postby Alaric » September 8th, 2021, 11:07 am

xeny wrote:I'm not saying SoR doesn't exist, but that graph would be more "honest" if retiree 2 was shown starting with a portfolio after it had been impacted by the losses of 1973.


Even more so, if they had equal wealth in 1970 and one retired in 1973 and the other in 1974, it would depend how much buying or selling either felt obliged to do.

I suggest that the idea to follow is that if intending to take an income by selling growth, have enough of a cash or near cash buffer, that it isn't necessary to sell when there isn't any growth.

EthicsGradient
Lemon Slice
Posts: 571
Joined: March 1st, 2019, 11:33 am
Has thanked: 33 times
Been thanked: 231 times

Re: Sequence of returns risk?

#440644

Postby EthicsGradient » September 8th, 2021, 3:05 pm

I think the FT Adviser article shows that wanting a return of 5% above inflation, over decades, can be risky. If you start that at a market high point, you could run out of money. I'd hope that few people plan for 5% plus inflation anyway. 4% is sometimes quoted (and effectively what the FT Adviser article says their "dynamic spending" method could have achieved), but that may be ambitious too.

Consider that World GDP growth has, since 1990 (fall of the USSR, China properly into capitalist production) averaged around 3.3% a year (I got that from some online IMF figures). It seems risky to assume you can do better than that, year after year; you're relying on being the rich capitalist who has control of the money and can thus come out ahead of the workers who don't have such power in the markets. Still true, but governments may decide that's not sustainable, and other considerations like climate change (or work to avoid it) may cut GDP growth.

Lootman
The full Lemon
Posts: 18681
Joined: November 4th, 2016, 3:58 pm
Has thanked: 628 times
Been thanked: 6563 times

Re: Sequence of returns risk?

#440651

Postby Lootman » September 8th, 2021, 3:20 pm

EthicsGradient wrote:I think the FT Adviser article shows that wanting a return of 5% above inflation, over decades, can be risky. If you start that at a market high point, you could run out of money. I'd hope that few people plan for 5% plus inflation anyway. 4% is sometimes quoted (and effectively what the FT Adviser article says their "dynamic spending" method could have achieved), but that may be ambitious too.

The quoted rates that you often see for a safe withdrawal rate, such as 5% or 4% a year, are typically quoted before inflation, not after it. That said the long term return from equities is something like 4% or 5% annually after inflation, i.e. a real return.

EthicsGradient wrote:Consider that World GDP growth has, since 1990 . . averaged around 3.3% a year. It seems risky to assume you can do better than that, year after year.

But again returns from shares have greatly exceeded GDP growth for a century or more. A 3.3% annual rate of return over the same period would have been miserably low in comparison to the actual returns from equities.

One guess as to why is that global GDP growth includes many countries that have performed badly in economic terms, and which were not historically investable.

hiriskpaul
Lemon Quarter
Posts: 3852
Joined: November 4th, 2016, 1:04 pm
Has thanked: 682 times
Been thanked: 1489 times

Re: Sequence of returns risk?

#440833

Postby hiriskpaul » September 9th, 2021, 1:04 am

Lootman wrote:
EthicsGradient wrote:I think the FT Adviser article shows that wanting a return of 5% above inflation, over decades, can be risky. If you start that at a market high point, you could run out of money. I'd hope that few people plan for 5% plus inflation anyway. 4% is sometimes quoted (and effectively what the FT Adviser article says their "dynamic spending" method could have achieved), but that may be ambitious too.

The quoted rates that you often see for a safe withdrawal rate, such as 5% or 4% a year, are typically quoted before inflation, not after it. That said the long term return from equities is something like 4% or 5% annually after inflation, i.e. a real return.

When drawing an income from a portfolio, long term returns are not the whole story. Sequence of returns really matters as well. As an example, using data from the Barclays Equity Gilts study, a portfolio consisting of 100% UK equities produced a total real (RPI adjusted) return of about 6.4% between 1968 and 1998. A 60/40 portfolio of equities/T bills managed only 5.4%. So simplistic total return logic might suggest that the 100% equities portfolio would be the best performer when drawing down over the 30 year period. However, with a 4% SWR, that portfolio would have run out of money after 20 years. In comparison, the lower performing 60/40 portfolio still had 28% of the money left after 30 years. That's inflation adjusted, ie 28% in 1968 pounds.

2 things are important to note here:
1) The highest performing portfolio ran out of money even though the real returns were 2.4% greater than the drawdown rate;
2) The lower performing portfolio was the better choice for drawdown.

1nvest
Lemon Quarter
Posts: 4323
Joined: May 31st, 2019, 7:55 pm
Has thanked: 680 times
Been thanked: 1316 times

Re: Sequence of returns risk?

#440928

Postby 1nvest » September 9th, 2021, 11:55 am

hiriskpaul wrote:
Lootman wrote:
EthicsGradient wrote:I think the FT Adviser article shows that wanting a return of 5% above inflation, over decades, can be risky. If you start that at a market high point, you could run out of money. I'd hope that few people plan for 5% plus inflation anyway. 4% is sometimes quoted (and effectively what the FT Adviser article says their "dynamic spending" method could have achieved), but that may be ambitious too.

The quoted rates that you often see for a safe withdrawal rate, such as 5% or 4% a year, are typically quoted before inflation, not after it. That said the long term return from equities is something like 4% or 5% annually after inflation, i.e. a real return.

When drawing an income from a portfolio, long term returns are not the whole story. Sequence of returns really matters as well. As an example, using data from the Barclays Equity Gilts study, a portfolio consisting of 100% UK equities produced a total real (RPI adjusted) return of about 6.4% between 1968 and 1998. A 60/40 portfolio of equities/T bills managed only 5.4%. So simplistic total return logic might suggest that the 100% equities portfolio would be the best performer when drawing down over the 30 year period. However, with a 4% SWR, that portfolio would have run out of money after 20 years. In comparison, the lower performing 60/40 portfolio still had 28% of the money left after 30 years. That's inflation adjusted, ie 28% in 1968 pounds.

2 things are important to note here:
1) The highest performing portfolio ran out of money even though the real returns were 2.4% greater than the drawdown rate;
2) The lower performing portfolio was the better choice for drawdown.

If the day after starting withdrawals the portfolio value halves, then in effect a 4% SWR has risen to a 8% SWR relative to the ongoing portfolio value. For those accumulating then stocks halving is a good thing as new money buys twice as many shares. Your higher reward example is perhaps suggestive that a bad sequence of returns occurred i.e. accumulators did well due to stock declines.

50/50 initial stock/bonds where bonds are spent first transitions to 100/0, averages 75/25. Conceptually both the bucket (50/50 -> 100/0 and constant weighted 75/25 might be expected to yield similar overall rewards, however the bucket approach is safer in having lower early years sequence of returns risk. In later years more often good gains followed by a decline are less harmful as that might be just giving back some of the gains rather than eating ones own capital.

Combining additional factors such as relative valuations and averaging (average in over 2 timepoints a year apart or whatever), and you can bolster the 4% suggested historic worst case SWR upward. 4% SWR is based on the worst case 30 year period, all others by definition were better.

75/25 constant weighted is said to have had a 99% chance of supporting a 4% SWR for 30 years

Image

With buckets and relative valuations that's likely increased to a 100%, IF forward time is within historic limits. For many the risk of not seeing another 30 years is a far greater risk. IIRC even in the worst cases where 4% SWR failed it still served 20+ years prior to failure.

Some might target 6% SWR with a reasonable chance of success, when so however you ideally also need a plan B in case of failure, such as a small separate pot that is set aside to accumulate and cover longevity (such as if the main pot does fail). Perhaps selecting asset(s) for that growth pot that may do well under the circumstances that drove the main pot to fail. That plan B could perhaps even be as simple as just having combined state and occupational pensions to cover (perhaps lower later life) spending that are yet to be received. If I'm 60 and newly retired, and at 68 will have combined £20K/year state and occupational pensions and could live off that alone, whilst also owning a home that provides late life care home cost cover then even if all my liquid wealth was blown by age 68 I'd be OK, such that I could push for a 14% SWR. In that context a 4% or even 5% SWR is pretty low. My actual SWR is however lower still, as I have heirs in mind, investing more for their benefit than myself.


Return to “Retirement Investing (inc FIRE)”

Who is online

Users browsing this forum: No registered users and 12 guests