Using a cashflow ladder in retirement?

OldScientist

Prism said:

BritishInvestor said:

Prism said:

westv said:

Presumably if a cash made that much difference then it would have been part of the original "4% studies".

Yes, in fact there are almost no studies that include cash. They all assume the downturn is protected by bonds. This is probably why you get quite a few questions on if/how much to have in a cash buffer and how to use it.

https://finalytiq.co.uk/wp-content/uploads/2017/02/FPA-Journal-December-1997-Conserving-Client-Portfolios-During-Retirement-Part-III.pdf

"As a final word, it is fair to conclude that cash is indeed "trash" in long-term investment portfolios, particularly when the client in seeking to maximize withdrawals."

Thanks. From chart 7 it looks like swapping some intermediate bonds (up to 50%) for cash doesn't change the withdrawal rate by much at all. In the current situation of cash typically returning more than bonds I would still likely do that. If we get back to being properly rewarded for the risk of bonds then maybe less cash. Besides, retirement cash isn't likely to be T-Bills or even instant access. Its more likely to be in savings accounts paying more interest. As some point the line between cash and bonds blurs.

Pfau (https://retirementresearcher.com/4-rule-work-around-world/) found that 50% bonds and 50% cash improved SAFEMAX in many countries including the US and UK (although he also notes that Bengen's intermediate term bonds and the bonds in the DMS dataset he is using may have different maturities - i.e. 50% bonds and 50% T-bills may have ended up with the same maturity as for Bengen's results). With both the JST and Barclays datasets adding T-bills to a UK portfolio appears to improve historical outcomes for the worst cases (except for 1937). Outcomes for the Barclay's dataset can be found at https://www.2020financial.co.uk/pension-drawdown-calculator/ for those who want to play.

Edit: Just to mention that the bonds in Barclay's/JST databases are 20 or 15 year maturity (so mixing with 50% cash brings the weighted average maturity to 8-10.5 years, i.e. intermediate term).

bostonerimus

OldScientist said:

Prism said:

BritishInvestor said:

Prism said:

westv said:

Presumably if a cash made that much difference then it would have been part of the original "4% studies".

Yes, in fact there are almost no studies that include cash. They all assume the downturn is protected by bonds. This is probably why you get quite a few questions on if/how much to have in a cash buffer and how to use it.

https://finalytiq.co.uk/wp-content/uploads/2017/02/FPA-Journal-December-1997-Conserving-Client-Portfolios-During-Retirement-Part-III.pdf

"As a final word, it is fair to conclude that cash is indeed "trash" in long-term investment portfolios, particularly when the client in seeking to maximize withdrawals."

Thanks. From chart 7 it looks like swapping some intermediate bonds (up to 50%) for cash doesn't change the withdrawal rate by much at all. In the current situation of cash typically returning more than bonds I would still likely do that. If we get back to being properly rewarded for the risk of bonds then maybe less cash. Besides, retirement cash isn't likely to be T-Bills or even instant access. Its more likely to be in savings accounts paying more interest. As some point the line between cash and bonds blurs.

Pfau (https://retirementresearcher.com/4-rule-work-around-world/) found that 50% bonds and 50% cash improved SAFEMAX in many countries including the US and UK (although he also notes that Bengen's intermediate term bonds and the bonds in the DMS dataset he is using may have different maturities - i.e. 50% bonds and 50% T-bills may have ended up with the same maturity as for Bengen's results). With both the JST and Barclays datasets adding T-bills to a UK portfolio appears to improve historical outcomes for the worst cases (except for 1937). Outcomes for the Barclay's dataset can be found at https://www.2020financial.co.uk/pension-drawdown-calculator/ for those who want to play.

But it all depends on the numbers you put into the mill before you crank the handle. With today's bond returns and inflation things would be messy.

I think partial annuitization might be interesting to look at as a replacement for fixed income, so cash, single premium or deferred annuity if you are still working and equities for long term growth. I essentially did that when I sold 90% of my fixed income and bought into a DB plan.

Right now the bonds bit of the retirement cashflow ladder would worry me.

Thrugelmir

BritishInvestor said:

Linton said:

BritishInvestor said:

Prism said:

BritishInvestor said:

Prism said:

westv said:

Presumably if a cash made that much difference then it would have been part of the original "4% studies".

Yes, in fact there are almost no studies that include cash. They all assume the downturn is protected by bonds. This is probably why you get quite a few questions on if/how much to have in a cash buffer and how to use it.

https://finalytiq.co.uk/wp-content/uploads/2017/02/FPA-Journal-December-1997-Conserving-Client-Portfolios-During-Retirement-Part-III.pdf

"As a final word, it is fair to conclude that cash is indeed "trash" in long-term investment portfolios, particularly when the client in seeking to maximize withdrawals."

Thanks. From chart 7 it looks like swapping some intermediate bonds (up to 50%) for cash doesn't change the withdrawal rate by much at all. In the current situation of cash typically returning more than bonds I would still likely do that. If we get back to being properly rewarded for the risk of bonds then maybe less cash. Besides, retirement cash isn't likely to be T-Bills or even instant access. Its more likely to be in savings accounts paying more interest. As some point the line between cash and bonds blurs.

" If we get back to being properly rewarded for the risk of bonds then maybe less cash."

But aren't you then getting into the realms of tactical asset allocation. For example, should large-cap growth shares also not be excluded given currently lofty valuations?

The difference between bonds and equity is that with (safe) bonds you know for certain, barring end of the world scenarios, what your total return will be from the day you bought until maturity.   So you are able to sensibly judge one outcome against another and make "tactical decisions" appropriate for your particular situation.

Equity comes with no guarantees so any tactical asset allocation must be pased on your guesses which, given a perfect market, could just as well turn out wrong.

As to asset allocation with equity: the important thing in my view is to avoid over-reliance on any one factor.

I guess the point I am trying to make is why introduce complexity/fiddling when there is little evidence to suggest that the plain vanilla, periodically rebalanced, 60/40 portfolio is"broken".

What are long dated Government bonds now yielding to maturity?  

OldScientist

bostonerimus said:

OldScientist said:

Prism said:

BritishInvestor said:

Prism said:

westv said:

Presumably if a cash made that much difference then it would have been part of the original "4% studies".

Yes, in fact there are almost no studies that include cash. They all assume the downturn is protected by bonds. This is probably why you get quite a few questions on if/how much to have in a cash buffer and how to use it.

https://finalytiq.co.uk/wp-content/uploads/2017/02/FPA-Journal-December-1997-Conserving-Client-Portfolios-During-Retirement-Part-III.pdf

"As a final word, it is fair to conclude that cash is indeed "trash" in long-term investment portfolios, particularly when the client in seeking to maximize withdrawals."

Thanks. From chart 7 it looks like swapping some intermediate bonds (up to 50%) for cash doesn't change the withdrawal rate by much at all. In the current situation of cash typically returning more than bonds I would still likely do that. If we get back to being properly rewarded for the risk of bonds then maybe less cash. Besides, retirement cash isn't likely to be T-Bills or even instant access. Its more likely to be in savings accounts paying more interest. As some point the line between cash and bonds blurs.

Pfau (https://retirementresearcher.com/4-rule-work-around-world/) found that 50% bonds and 50% cash improved SAFEMAX in many countries including the US and UK (although he also notes that Bengen's intermediate term bonds and the bonds in the DMS dataset he is using may have different maturities - i.e. 50% bonds and 50% T-bills may have ended up with the same maturity as for Bengen's results). With both the JST and Barclays datasets adding T-bills to a UK portfolio appears to improve historical outcomes for the worst cases (except for 1937). Outcomes for the Barclay's dataset can be found at https://www.2020financial.co.uk/pension-drawdown-calculator/ for those who want to play.

But it all depends on the numbers you put into the mill before you crank the handle. With today's bond returns and inflation things would be messy.

I think partial annuitization might be interesting to look at as a replacement for fixed income, so cash, single premium or deferred annuity if you are still working and equities for long term growth. I essentially did that when I sold 90% of my fixed income and bought into a DB plan.

Right now the bonds bit of the retirement cashflow ladder would worry me.

Although in the UK we have a way to go to match to 20% inflation rate that occurred at the end of the first world war (and that would have caused problems for retirees who were relying on a portfolio of stocks/bonds/cash). Of course, in reality only a very small proportion of the population at that time actually retired.

I agree, in the absence of other significant (relative to expenditure) guaranteed income, then partial annuitization might very well prove to be a good decision. While I've modelled this for the US, I have yet to finish working stuff out for the UK - will get around to it eventually!

OldScientist

bostonerimus said:

BritishInvestor said:

k6chris said:

I think this type of investement plan, as well as things like 'natural yield' investing, are as much about giving people a warm(er) sense of security as they are about maximising returns.  I would argue that is not a bad thing, since sleeping well and not overly fretting about investments means you are less likely to make a stupid and harmful (investment) decision.  There is no 'do it this way' plan that suits everyone.  My view (which is worthless) is keep a 2 year cash buffer and invest the rest at a level that allows you to sleep well, which for me is about 70% equity and 30% bonds / non equity.  Rebalance every year.  Good luck

I can't see how a natural yield approach will give peace of mind.

(But agreed that peace of mind is important).

My equity index funds produce around 2% dividends each year. If that's enough to cover your spending then you can live without spending capital and that should be a warm and fuzzy feeling.

My goal has always been to have a self sustaining retirement income pot that will not fall in value through retirement so I don't have to worry about outliving my money and will have money to leave to my heirs.

While Abraham Okusanya is rather scathing of natural yield as a withdrawal strategy (https://finalytiq.co.uk/natural-yield-totally-bonkers-retirement-income-strategy/), it only shares some of the same weaknesses as any percentage of portfolio approach - namely that the real income can vary considerably from year to year and become quite small (of course, with a strong base of guaranteed income, this variation is not so important). Historical outcomes for a 60/40 USA portfolio, 30 year retirement, annual data from 1872 to 2015, and JST data set are plotted below (the lines are the percentiles of historical cases, i.e. '50' is the median, '1' represents the worst 1% of historical cases, etc.) - in the worst cases, the withdrawal rate does fall to about 2% of the initial portfolio value in real terms. Not shown as a graph, but the worst case final portfolio value was just over 50% of the initial portfolio (in real terms), so a reasonable preservation of capital.
 

For a UK portfolio (60/40) the results are even more fun, with a real withdrawal rate at the 1st percentile of close to 1.5% after about 5-10 years (the worst case is 1.1% - not far from Okusanya's value for a 50/50 portfolio - the datasets are also slightly different) and a worst case final portfolio value of just over 20% (so not as much capital preservation as one might have liked).


While the worst cases were not much fun, the median retirements were better than choosing a SAFEMAX approach (in the UK case, that was actually true for more than 75% of retirements) and there is money still in the pot after 30 years which means the longer lived retiree won't be left without funds or something to leave as legacy.

GazzaBloom

To put all this worry about 30-40 year retirement and having enough money into context, I have just learned today that a former colleague and friend of of mine has died at age 63...that's 8 years left for me should I suffer the same fate.

Retirement planning must be a balance between saving, sweating investment asset allocation and balancing fear of long term low returns, market crashes, too much exposure to market volatility, preservation of capital, etc, etc... and a sudden curtailment or derailment of plans.

This is why I have taken my DB pension at the earliest opportunity at age 55, with tax free lump sum and reduced annual payment so there can be some living today as no-one knows what tomorrow holds or how many tomorrows there will be. I would be happy to be 80 and living a modest life having not worked longer than I really wanted, travelled, bought a really nice guitar or two, relaxed and fixed up my house earlier in life, rather than having lived in fear of running out of money fretting this mortal coil and suddenly departing.

bostonerimus

GazzaBloom said:

To put all this worry about 30-40 year retirement and having enough money into context, I have just learned today that a former colleague and friend of of mine has died at age 63...that's 8 years left for me should I suffer the same fate.

Retirement planning must be a balance between saving, sweating investment asset allocation and balancing fear of long term low returns, market crashes, too much exposure to market volatility, preservation of capital, etc, etc... and a sudden curtailment or derailment of plans.

This is why I have taken my DB pension at the earliest opportunity at age 55, with tax free lump sum and reduced annual payment so there can be some living today as no-one knows what tomorrow holds or how many tomorrows there will be. I would be happy to be 80 and living a modest life having not worked longer than I really wanted, travelled, bought a really nice guitar or two, relaxed and fixed up my house earlier in life, rather than having lived in fear of running out of money fretting this mortal coil and suddenly departing.

We must talk in averages and distributions when it comes to planning and predictions, hen you can make sensible assumptions. But that's all they are. Your colleague's death is a single event and while obviously sad and psychologically meaningful to you, it should have no effect on your planning. You could just as well be one of the people that forms the other end of the longevity distribution tails. Taking the DB pension is a good way to insure for your longevity. I did just the same and took my DB pension at 55

GazzaBloom

bostonerimus said:

GazzaBloom said:

To put all this worry about 30-40 year retirement and having enough money into context, I have just learned today that a former colleague and friend of of mine has died at age 63...that's 8 years left for me should I suffer the same fate.

Retirement planning must be a balance between saving, sweating investment asset allocation and balancing fear of long term low returns, market crashes, too much exposure to market volatility, preservation of capital, etc, etc... and a sudden curtailment or derailment of plans.

This is why I have taken my DB pension at the earliest opportunity at age 55, with tax free lump sum and reduced annual payment so there can be some living today as no-one knows what tomorrow holds or how many tomorrows there will be. I would be happy to be 80 and living a modest life having not worked longer than I really wanted, travelled, bought a really nice guitar or two, relaxed and fixed up my house earlier in life, rather than having lived in fear of running out of money fretting this mortal coil and suddenly departing.

We must talk in averages and distributions when it comes to planning and predictions, hen you can make sensible assumptions. But that's all they are. Your colleague's death is a single event and while obviously sad and psychologically meaningful to you, it should have no effect on your planning. You could just as well be one of the people that forms the other end of the longevity distribution tails. Taking the DB pension is a good way to insure for your longevity. I did just the same and took my DB pension at 55

Agreed, I calculated that it would take 20+ years for taking the DP pension at 65 rather than 55 with lump sum and reduced payment before leaving it until 65 would would be more beneficial, that would put me at 85 before it before it became the more lucrative option. Will I be alive and need the extra money at 85?

bostonerimus

OldScientist said:

bostonerimus said:

BritishInvestor said:

k6chris said:

I think this type of investement plan, as well as things like 'natural yield' investing, are as much about giving people a warm(er) sense of security as they are about maximising returns.  I would argue that is not a bad thing, since sleeping well and not overly fretting about investments means you are less likely to make a stupid and harmful (investment) decision.  There is no 'do it this way' plan that suits everyone.  My view (which is worthless) is keep a 2 year cash buffer and invest the rest at a level that allows you to sleep well, which for me is about 70% equity and 30% bonds / non equity.  Rebalance every year.  Good luck

I can't see how a natural yield approach will give peace of mind.

(But agreed that peace of mind is important).

My equity index funds produce around 2% dividends each year. If that's enough to cover your spending then you can live without spending capital and that should be a warm and fuzzy feeling.

My goal has always been to have a self sustaining retirement income pot that will not fall in value through retirement so I don't have to worry about outliving my money and will have money to leave to my heirs.

While Abraham Okusanya is rather scathing of natural yield as a withdrawal strategy (https://finalytiq.co.uk/natural-yield-totally-bonkers-retirement-income-strategy/), it only shares some of the same weaknesses as any percentage of portfolio approach - namely that the real income can vary considerably from year to year and become quite small (of course, with a strong base of guaranteed income, this variation is not so important).

Just dividends probably aren't going to fund many retirements because people's pots are big enough compared to their spending. One of my tactics has always been frugality l as I was never sure what the size of my retirement pot would be, but I've always been able to control my spending. A smaller budget just means that the retirement pot has to do less work. Early in my planning I included capital gains in with dividends, but always avoided spending down capital because of sequence of returns risk and my desire to leave money to my heirs. I actively sought out ways to reduce what I had to withdraw from my investment portfolio and so took a DB pension when I could and also bought a rental property. When those are combined with my relatively low income needs I end up actually depositing money into my investments each year, so I have a negative retirement withdrawal rate....of course I laid out a lot for the DB pension and the rental flat while I was working. But adding to my pot while I'm retired is a great feeling and of course my dividends get reinvested and my capital gains accumulate.

bostonerimus

GazzaBloom said:

bostonerimus said:

GazzaBloom said:

To put all this worry about 30-40 year retirement and having enough money into context, I have just learned today that a former colleague and friend of of mine has died at age 63...that's 8 years left for me should I suffer the same fate.

Retirement planning must be a balance between saving, sweating investment asset allocation and balancing fear of long term low returns, market crashes, too much exposure to market volatility, preservation of capital, etc, etc... and a sudden curtailment or derailment of plans.

This is why I have taken my DB pension at the earliest opportunity at age 55, with tax free lump sum and reduced annual payment so there can be some living today as no-one knows what tomorrow holds or how many tomorrows there will be. I would be happy to be 80 and living a modest life having not worked longer than I really wanted, travelled, bought a really nice guitar or two, relaxed and fixed up my house earlier in life, rather than having lived in fear of running out of money fretting this mortal coil and suddenly departing.

We must talk in averages and distributions when it comes to planning and predictions, hen you can make sensible assumptions. But that's all they are. Your colleague's death is a single event and while obviously sad and psychologically meaningful to you, it should have no effect on your planning. You could just as well be one of the people that forms the other end of the longevity distribution tails. Taking the DB pension is a good way to insure for your longevity. I did just the same and took my DB pension at 55

Agreed, I calculated that it would take 20+ years for taking the DP pension at 65 rather than 55 with lump sum and reduced payment before leaving it until 65 would would be more beneficial, that would put me at 85 before it before it became the more lucrative option. Will I be alive and need the extra money at 85?

Yes, I did the same assessment. Not surprisingly the numbers usually work out that the break even point is just before average mortality age.