the "safe" 4% drawdown rule - flexible if receiving a DB pension also ?

JohnWinder

‘Firstly there is no SWR.’

Secondly, here’s a definition of it (were it to exist):
‘A safe withdrawal rate (SWR) is defined as the quantity of money, expressed as a percentage of the initial investment, which can be withdrawn per year for a given quantity of time, including adjustments for inflation, and not lead to portfolio failure; failure being defined as a 95% probability of depletion to zero at any time within the specified period. ‘ https://www.bogleheads.org/wiki/Safe_withdrawal_rates
So, it’s a useful concept even if there’s no single useful figure for it, despite some shortcomings.

‘Its a largely a made up figure.’

Newton made up a figure for the force of gravity based on observations and some theory. Bengen made up a figure based on historical returns for US securities. You can read his papers and download his book at https://www.academia.edu/. Bengen is not as universally applicable as Newton, but the made up figures weren’t just made up.

SouthCoastBoy

Just to add to the debate, in retirement I don't see spending being linear, especially as the majority of people will have access to a state pension.

Personally I feel a detail spreadsheet is the way to go, adjusted yearly.

scobie

dunstonh said:

Have read a few threads on here discussing safe withdrawl rates of your pension pot etc, and often they become too complex for somebody like myself to follow , however I often see 4% being referred to

Firstly there is no SWR.   It is opinion and will be very much depend on your assets and your age.  4% is considered too high for UK.    2.5%-3% if you are in your 50s rising to 3% in your low 60s rising to 3.5% in your mid to late 60s is more typical for UK.

so in 4 year , if i expect my DB to then be £32k pa.  would i still use the 4% rule on the 600k (which would give 24k pa and increase that by say 2% pa), or because i am half funded by my DB pension , and we will get state pensions 10 year into our retirement  does that mean you can afford to use a higher % , say 5%, then reduce it to 3% once both get state pensions ?

Take DB and state pension out of the equation.   e.g. if you have an income need of £35,000 a year and your state pension and DB pension will provide £25,000, then your shortfall is £10,000.     You need to factor that £10,000 shortfall into the DC pension (and other savings/investments).

thanks any advice , wasn't sure how flexible this 4% scenario is meant to be

Its a largely a made up figure. So, its as flexible as any other made up figure.

And if you are doing things like "funding the gap" or have variable cost things to pay for over the years then you need to factor capital spending into it.   So, an SWR is largely pointless then.

Why would 3.5% be considered normal in your late 60s? Simple maths suggests that even with no capital appreciation your pot would last close to 30 years and the likelihood of there being no capital appreciation is zero. 

And unless you’re expecting to live to 110 this figure seems way too conservative. 

SouthCoastBoy

scobie said:

dunstonh said:

Have read a few threads on here discussing safe withdrawl rates of your pension pot etc, and often they become too complex for somebody like myself to follow , however I often see 4% being referred to

Firstly there is no SWR.   It is opinion and will be very much depend on your assets and your age.  4% is considered too high for UK.    2.5%-3% if you are in your 50s rising to 3% in your low 60s rising to 3.5% in your mid to late 60s is more typical for UK.

so in 4 year , if i expect my DB to then be £32k pa.  would i still use the 4% rule on the 600k (which would give 24k pa and increase that by say 2% pa), or because i am half funded by my DB pension , and we will get state pensions 10 year into our retirement  does that mean you can afford to use a higher % , say 5%, then reduce it to 3% once both get state pensions ?

Take DB and state pension out of the equation.   e.g. if you have an income need of £35,000 a year and your state pension and DB pension will provide £25,000, then your shortfall is £10,000.     You need to factor that £10,000 shortfall into the DC pension (and other savings/investments).

thanks any advice , wasn't sure how flexible this 4% scenario is meant to be

Its a largely a made up figure. So, its as flexible as any other made up figure.

And if you are doing things like "funding the gap" or have variable cost things to pay for over the years then you need to factor capital spending into it.   So, an SWR is largely pointless then.

Why would 3.5% be considered normal in your late 60s? Simple maths suggests that even with no capital appreciation your pot would last close to 30 years and the likelihood of there being no capital appreciation is zero. 

And unless you’re expecting to live to 110 this figure seems way too conservative. 

You need to factor in inflation 

OldScientist

SouthCoastBoy said:

scobie said:

dunstonh said:

Have read a few threads on here discussing safe withdrawl rates of your pension pot etc, and often they become too complex for somebody like myself to follow , however I often see 4% being referred to

Firstly there is no SWR.   It is opinion and will be very much depend on your assets and your age.  4% is considered too high for UK.    2.5%-3% if you are in your 50s rising to 3% in your low 60s rising to 3.5% in your mid to late 60s is more typical for UK.

so in 4 year , if i expect my DB to then be £32k pa.  would i still use the 4% rule on the 600k (which would give 24k pa and increase that by say 2% pa), or because i am half funded by my DB pension , and we will get state pensions 10 year into our retirement  does that mean you can afford to use a higher % , say 5%, then reduce it to 3% once both get state pensions ?

Take DB and state pension out of the equation.   e.g. if you have an income need of £35,000 a year and your state pension and DB pension will provide £25,000, then your shortfall is £10,000.     You need to factor that £10,000 shortfall into the DC pension (and other savings/investments).

thanks any advice , wasn't sure how flexible this 4% scenario is meant to be

Its a largely a made up figure. So, its as flexible as any other made up figure.

And if you are doing things like "funding the gap" or have variable cost things to pay for over the years then you need to factor capital spending into it.   So, an SWR is largely pointless then.

Why would 3.5% be considered normal in your late 60s? Simple maths suggests that even with no capital appreciation your pot would last close to 30 years and the likelihood of there being no capital appreciation is zero. 

And unless you’re expecting to live to 110 this figure seems way too conservative. 

You need to factor in inflation 

And sequence of returns since, for fixed real withdrawals, large drops in portfolio value early on in the 30 year or so retirement period can have a significant effect on outcomes and this is why working with constant returns can be misleading. This is one of the reasons Bengen did his research. It is interesting to note that the lowest 30 year real annualised return on UK stocks was about 1.2%, for bonds it was -4.7%, for bills (i.e. a cash proxy) -2.4%, and for a 60/40 portfolio it was about 0.2% (own calculations from data at https://www.macrohistory.net/ ).

If you want to remove the effects of sequence of returns, then two alternatives are

1) investing in an inflation linked bond ladder with, currently, a real yield of about 0.5%, would result in a payout rate of about 3.5% for a 30 year ladder (taking a 65 year old to 95yo, with roughly 10% chance of exceeding this for a single person and 20% that at least one of a couple will live longer). For a 35 year ladder (giving about 8% chance of at least one of a 65 yo couple being alive for longer) would reduce the payout to 3.1% (ONS longevity data from https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandlifeexpectancies/articles/lifeexpectancycalculator/2019-06-07 )

2) purchasing an inflation linked annuity would currently provide payout rates of about 4.2% and 3.5% for single and joint with 50% survivor benefits, respectively.

However,  the OP may already have sufficient guaranteed income (i.e., DB pension, state pensions) to cover their needs without obtaining any more. Hence, a third alternative is to use percentage of portfolio withdrawal approaches (and there are a number available) instead of inflation adjusted 'safe' withdrawal methods.

Linton

scobie said:

dunstonh said:

Have read a few threads on here discussing safe withdrawl rates of your pension pot etc, and often they become too complex for somebody like myself to follow , however I often see 4% being referred to

Firstly there is no SWR.   It is opinion and will be very much depend on your assets and your age.  4% is considered too high for UK.    2.5%-3% if you are in your 50s rising to 3% in your low 60s rising to 3.5% in your mid to late 60s is more typical for UK.

so in 4 year , if i expect my DB to then be £32k pa.  would i still use the 4% rule on the 600k (which would give 24k pa and increase that by say 2% pa), or because i am half funded by my DB pension , and we will get state pensions 10 year into our retirement  does that mean you can afford to use a higher % , say 5%, then reduce it to 3% once both get state pensions ?

Take DB and state pension out of the equation.   e.g. if you have an income need of £35,000 a year and your state pension and DB pension will provide £25,000, then your shortfall is £10,000.     You need to factor that £10,000 shortfall into the DC pension (and other savings/investments).

thanks any advice , wasn't sure how flexible this 4% scenario is meant to be

Its a largely a made up figure. So, its as flexible as any other made up figure.

And if you are doing things like "funding the gap" or have variable cost things to pay for over the years then you need to factor capital spending into it.   So, an SWR is largely pointless then.

Why would 3.5% be considered normal in your late 60s? Simple maths suggests that even with no capital appreciation your pot would last close to 30 years and the likelihood of there being no capital appreciation is zero. 

And unless you’re expecting to live to 110 this figure seems way too conservative. 

Apart from your calculation ignoring inflation:

If you aim for a 100% success rate on the very small amount of historical data available the SWR will be driven by the 2 cases where crashes and/or inflation linked drawdown  severely decrease the size of your pot in the early years of retirement and your pot fails to recover sufficiently before being hit by the next crash and so you run out of money.  This is known as Sequence of Returns (SOR) risk.

You need sufficient spare savings to live through the worst SOR situation or some strategy whereby you can avoid drawing down too much from your pension pot when SOR threatens your future.  There is much discussion what is the best such strategy.

MK62

....plus, as we've seen in recent years, there's the sequence of inflation to consider too......that can often be even more insidious, as unlike stock market crashes, it's seldom recovered......it's usually pretty much baked in for the rest of your retirement.

Albermarle

eastcorkram said:

Not really relevant to the thread I know, but I'm amazed how anyone somehow ends up already in receipt of a DB pension paying £29k a year at the age of 53 

Mick 70 has started many threads, if you want to check back, and as far as I remember he had to take this DB pension at 50, due to the unusual conditions of the scheme.

Pat38493

Linton said:

Apart from your calculation ignoring inflation:

If you aim for a 100% success rate on the very small amount of historical data available the SWR will be driven by the 2 cases where crashes and/or inflation linked drawdown  severely decrease the size of your pot in the early years of retirement and your pot fails to recover sufficiently before being hit by the next crash and so you run out of money.  This is known as Sequence of Returns (SOR) risk.

You need sufficient spare savings to live through the worst SOR situation or some strategy whereby you can avoid drawing down too much from your pension pot when SOR threatens your future.  There is much discussion what is the best such strategy.

OK but in a previous post, you made the same point, but it seemed to imply that using historical scenarios and then an SWR results in you saving up too much money because 99% of the time you will end up as the richest person in the graveyard, or maybe I misunderstood your previous point?

This is something I also thought about quite a bit and other than having a large cash float, there isn't much obvious you can do about it as you won't know whether you are in a 1920s or late 60s scenario until after you are already retired.

This raises another issue which is the danger of running historical scenarios too often - imagine you ran your historical scenario today and it shows 100% success, so you flip the finger to your boss and exit stage left. 

The following week, there is a big crash of 30% on the markets.  

You re-run your historical simulation and it says your success is now 70%.

Should you
a) Panic, call your boss, beg forgiveness and try to go back to work.
b) Open a beer on the reflection that at least one of the historical scenarios you modelled the prior week included an early 30% crash and worse - therefore you are now inside your own scenario that showed you are perfectly fine.

The historical models like cfiresim and so on, don't have any concept of where we are in the long term market cycle, and the historical data seems to show that the stock market is not random, but very unpredictable in the short term - it however responds to real world events in a somewhat predictable way, in that you can say pretty confidently that if news x is announced as a surprise, markets will go down.  You have no way of knowing how much or for how long, but it's not truly random.

Further, markets seem to turn in cycles that lasts a decade to 15 years or so - if there has recently been a 40% crash in the markets, it's very unlikely there will be another 40% crash next week.  However these historical simulations don't take this into account.

The only approach I've seen that attempts to correct for this in advance is to use CAPE or some kind of adjusted CAPE figure as a variable in your withdrawal strategy.  (CAPE as I understand it, is an attempt to assess whether the overall market is currently under or over valued based on the combined long term earnings of all the companies in the relevant market for any posters not aware).

Interestingly, this approach also attempted to compensate for the issue you raised that when you use cfiresim or many others, if you model a 40 year retirement, the last cohort that you have results for will be 1993, so you will not see any effects of the 2000 dot com bubble, or the 2008 financial crisis.  This was also attempted to be corrected by including these scenarios and using monte carlo approach for future periods not yet known in the same model.

Albermarle

For someone with a good basic guaranteed pension income, that covers main regular spending,  I am not sure it is necessary to get too tied up in knots about SWR's on a decent sized DC pot, and I would tend to agree with @german_keeper's comment.
I think what I mean is that in the position you are in, unlike those who only have a DC pot, you have plenty of leeway just to take things as they come rather than plan to the nth degree.