One for the experts( you know who you are!)

BritishInvestor

Mistermeaner said:

BritishInvestor said:

Mistermeaner said:

Audaxer said:

HarryGray said:

BritishInvestor said:

HarryGray said:

Well, technically the most optimal asset allocation at a 4% withdrawal rate is 100% equity. 

I'm not sure it's as clear cut as that.

Nothing is clear cut, simply stating that statistically the best retirement drawdown ignoring a clients Attitude to Risk is 100% equity at a constant withdrawal rate. There is no evidence anywhere to suggest that 'lifestyling' works 

I would have thought that if you have a portfolio of 100% equity and there is a poor sequence of returns, especially in the first decade of retirement, there would be more risk of running out of money than with a balanced portfolio?

There was a great thread somewhere recently (that I can't find) that covered this in some detail referencing various studies 

The somewhat surprising conclusion was that you are generally better off 100% in equities - or at least no worse off 

There has only been something like 2 or 3 starting years of retirement in the last 100+ years when this would have depleted your pot in 30years

Would be good to see the studies, and to be clear whether success means least (percentage of) historical failures or how long the money lasted for in the worst historical outcome.
It's also worth bearing in mind that these studies might not have included real-world issues such as various fees and investor (mis)behaviour.

Drawdown: safe withdrawal rates

Thanks, have sent you a message

jamesd

BritishInvestor said:

Mistermeaner said:

There was a great thread somewhere recently (that I can't find) that covered this in some detail referencing various studies 

The somewhat surprising conclusion was that you are generally better off 100% in equities - or at least no worse off 

There has only been something like 2 or 3 starting years of retirement in the last 100+ years when this would have depleted your pot in 30years

Would be good to see the studies, and to be clear whether success means least (percentage of) historical failures or how long the money lasted for in the worst historical outcome.
It's also worth bearing in mind that these studies might not have included real-world issues such as various fees and investor (mis)behaviour.

The standard SAFEMAX calculation is based on:
1. a chosen investment mix, possibly varying over time
2. a specified number of years
3. a specified set of drawdown rules
4. historic or randomly varying within ranges investment performance
5. 100% success rate

In the studies failure normally mean running out of money but in practice it means adjusting income down by more than provided in the rules.

Studies have then gone on to try varying investment mixtures to see how high a SAFEMAX can be made.

Studies normally ignore costs. Kitces Guyton looked at them for constant inflation-adjusted income - 4% rule basis - and found that the SWR  percentage was reduced by around 30% of costs. So deduct 0.3 from 4% if costs are 1% or use a tool that lets you specify costs.

Broadly, stick to at least 50% equities and the differences aren't large between historic and synthetic approaches.

Because the bad cases are pretty extreme it's common to look at lower success rates like 99% or 90% and there are good arguments for 75% or lower being better. But there's a catch. Most cases do better with high equities but the limiting cases can be the ones that need the bonds. Mistermeaner is right about the investments outcome but I don't know whether they have recognised that the higher than usual equity levels come about because the sample population being tested against has been modified.

Because of this I do think it's best to keep some bonds, to handle those cases better. But someone with a surplus and lots of income flexibility might be able to handle those cases with a big spending cut.

Part of my objectives with Drawdown: safe withdrawal rates was describing some of the breadth of what produces reasonably good outcomes to help people pick what they prefer within that range.

BritishInvestor

jamesd said:

BritishInvestor said:

Mistermeaner said:

There was a great thread somewhere recently (that I can't find) that covered this in some detail referencing various studies 

The somewhat surprising conclusion was that you are generally better off 100% in equities - or at least no worse off 

There has only been something like 2 or 3 starting years of retirement in the last 100+ years when this would have depleted your pot in 30years

Would be good to see the studies, and to be clear whether success means least (percentage of) historical failures or how long the money lasted for in the worst historical outcome.
It's also worth bearing in mind that these studies might not have included real-world issues such as various fees and investor (mis)behaviour.

The standard SAFEMAX calculation is based on:
1. a chosen investment mix, possibly varying over time
2. a specified number of years
3. a specified set of drawdown rules
4. historic or randomly varying within ranges investment performance
5. 100% success rate

In the studies failure normally mean running out of money but in practice it means adjusting income down by more than provided in the rules.

Studies have then gone on to try varying investment mixtures to see how high a SAFEMAX can be made.

Studies normally ignore costs. Guyton looked at them for constant inflation-adjusted income - 4% rule basis - and found that the SWR  percentage was reduced by around 30% of costs. So deduct 0.3 from 4% if costs are 1% or use a tool that lets you specify costs.

Broadly, stick to at least 50% equities and the differences aren't large between historic and synthetic approaches.

Because the bad cases are pretty extreme it's common to look at lower success rates like 99% or 90% and there are good arguments for 75% or lower being better. But there's a catch. Most cases do better with high equities but the limiting cases can be the ones that need the bonds. Mistermeaner is right about the investments outcome but I don't know whether they have recognised that the higher than usual equity levels come about because the sample population being tested against has been modified.

Because of this I do think it's best to keep some bonds, to handle those cases better. But someone with a surplus and lots of income flexibility might be able to handle those cases with a big spending cut.

Part of my objectives with Drawdown: safe withdrawal rates was describing some of the breadth of what produces reasonably good outcomes to help people pick what they prefer within that range.

Re costs, in my analysis, I'm seeing 0.5% reduction in SWR which ties in with what Abraham is seeing (which isn't surprising as I use his tool )
https://finalytiq.co.uk/impact-of-adviser-fees-on-withdrawal-rates-in-retirement-portfolio/
Agreed on keeping bonds, and I would guess the "typical" advised split is circa 60/40. The "optimal", investor behaviour aside, was around 90% in the quick test I ran, but obviously very dependent on parameters.

jamesd

This:

"The two methods I suggested for MC are only using unadulterated historic data.  The first was drawing at random based on a normal distribution using the mean and SD.  The second was drawing at random with replacement from historic data.  If you think these are incorrect assumptions, then historic window has already failed."

is something of an intellectual trap.

Neither of the approaches described is actually historic because the real historic sequences exclude non-viable year on year state changes like say 15% inflation to 10% deflation to 10% inflation. They also assume that the humans can handle arbitrary variation and I don't think that real human behaviour supports this

The historic data may well contain all three values but history is being adulterated by permitting non-viable changes.

What this doesn't do is make the analysis approach invalid. But what it should do is cause clear distinctions in the description between historic sequence approaches and historic variance approaches. It's totally correct here to use either historic sequences or random sequence within historic variance, just try to be clear about which is being used. Each way has its own strengths and weaknesses.

The practical effects of using random rather than  historic sequences tend to be something like:
1. SWR 0.2 to 0.3 lower
2. higher equity component, often 100%.

As with the effect of lower success rates the differences arise from changes in the sample and like lower success rates either approach is well supported and entirely reasonable for retirement planning. Just pick your preferred combination and stick to the resulting SWR.

BritishInvestor

jamesd said:

This:

"The two methods I suggested for MC are only using unadulterated historic data.  The first was drawing at random based on a normal distribution using the mean and SD.  The second was drawing at random with replacement from historic data.  If you think these are incorrect assumptions, then historic window has already failed."

is something of an intellectual trap.

Neither of the approaches described is actually historic because the real historic sequences exclude non-viable year on year state changes like say 15% inflation to 10% deflation to 10% inflation. They also assume that the humans can handle arbitrary variation and I don't think that real human behaviour supports this

The historic data may well contain all three values but history is being adulterated by permitting non-viable changes.

What this doesn't do is make the analysis approach invalid. But what it should do is cause clear distinctions in the description between historic sequence approaches and historic variance approaches. It's totally correct here to use either historic sequences or random sequence within historic variance, just try to be clear about which is being used. Each way has its own strengths and weaknesses.

The practical effects of using random rather than  historic sequences tend to be something like:
1. SWR 0.2 to 0.3 lower
2. higher equity component, often 100%.

As with the effect of lower success rates the differences arise from changes in the sample and like lower success rates either approach is well supported and entirely reasonable for retirement planning. Just pick your preferred combination and stick to the resulting SWR.

"1. SWR 0.2 to 0.3 lower"
I had a quick test earlier and I'm seeing a much larger difference. 

jamesd

BritishInvestor said:

jamesd said:
 Guyton looked at them for constant inflation-adjusted income - 4% rule basis - and found that the SWR  percentage was reduced by around 30% of costs. So deduct 0.3 from 4% if costs are 1% or use a tool that lets you specify costs.

Re costs, in my analysis, I'm seeing 0.5% reduction in SWR which ties in with what Abraham is seeing (which isn't surprising as I use his tool )
https://finalytiq.co.uk/impact-of-adviser-fees-on-withdrawal-rates-in-retirement-portfolio/
Agreed on keeping bonds, and I would guess the "typical" advised split is circa 60/40. The "optimal", investor behaviour aside, was around 90% in the quick test I ran, but obviously very dependent on parameters.

Correction it was Kitces, not Guyton that I was referring to.

No surprise and not unreasonable. I forget the details of Kitces version that I was thinking of but the differences are probably:

1. Kitces using US investor and investments vs UK and international
2. Kitces using 30 years of 4% rule with many sequences vs 41 years and one sequence
3. longer bond duration in the usual US mix

Using a variable spending strategy like Guyton-Klinger would mask the effect of ignoring costs unless you experienced something near the worst case because they show up in investment performance.

jamesd

BritishInvestor said:

jamesd said:

The practical effects of using random rather than  historic sequences tend to be something like:
1. SWR 0.2 to 0.3 lower
2. higher equity component, often 100%.

"1. SWR 0.2 to 0.3 lower"
I had a quick test earlier and I'm seeing a much larger difference. 

I expect more like 0.2-0.3 for common mixtures but if you run the comparison for some specific conditions that's OK. I was trying to give some idea of what to expect, not cover all possibilities.

Audaxer

jamesd said:

BritishInvestor said:

jamesd said:
 Guyton looked at them for constant inflation-adjusted income - 4% rule basis - and found that the SWR  percentage was reduced by around 30% of costs. So deduct 0.3 from 4% if costs are 1% or use a tool that lets you specify costs.

Re costs, in my analysis, I'm seeing 0.5% reduction in SWR which ties in with what Abraham is seeing (which isn't surprising as I use his tool )
https://finalytiq.co.uk/impact-of-adviser-fees-on-withdrawal-rates-in-retirement-portfolio/
Agreed on keeping bonds, and I would guess the "typical" advised split is circa 60/40. The "optimal", investor behaviour aside, was around 90% in the quick test I ran, but obviously very dependent on parameters.

Correction it was Kitces, not Guyton that I was referring to.

No surprise and not unreasonable. I forget the details of Kitces version that I was thinking of but the differences are probably:

1. Kitces using US investor and investments vs UK and international
2. Kitces using 30 years of 4% rule with many sequences vs 41 years and one sequence
3. longer bond duration in the usual US mix

Using a variable spending strategy like Guyton-Klinger would mask the effect of ignoring costs unless you experienced something near the worst case because they show up in investment performance.

jamesd, do you know of any information or studies regarding Safe Withdrawal Rates for retirement portfolios with funds aiming to provide growing natural income from dividends?

jamesd

Audaxer said:

jamesd, do you know of any information or studies regarding Safe Withdrawal Rates for retirement portfolios with funds aiming to provide growing natural income from dividends?

No but I expect markedly lower SWRs, maybe 1-1.5% lower than 4.5% as a crude guess.

Consider that Bengen found a useful increase in SWRs - 4% to 4.5% - from adding small caps, which tend to pay low or no dividends but increase total return. 4.5% used 35% large cap, 20% small cap and 45% intermediate duration bonds and Bengen clearly prefers this because he writes about his 4.5% rule in preference to 4%.

While I expect lower SWRs from lower total returns an individual might still want to use living off native yield approaches that growing dividends can complement. If the person has ample income that way and a substantial inheritance desire it can be a good fit even with lower SWR. Sleeping more easily can be a benefit.

BritishInvestor

Audaxer said:

jamesd said:

BritishInvestor said:

jamesd said:
 Guyton looked at them for constant inflation-adjusted income - 4% rule basis - and found that the SWR  percentage was reduced by around 30% of costs. So deduct 0.3 from 4% if costs are 1% or use a tool that lets you specify costs.

Re costs, in my analysis, I'm seeing 0.5% reduction in SWR which ties in with what Abraham is seeing (which isn't surprising as I use his tool )
https://finalytiq.co.uk/impact-of-adviser-fees-on-withdrawal-rates-in-retirement-portfolio/
Agreed on keeping bonds, and I would guess the "typical" advised split is circa 60/40. The "optimal", investor behaviour aside, was around 90% in the quick test I ran, but obviously very dependent on parameters.

Correction it was Kitces, not Guyton that I was referring to.

No surprise and not unreasonable. I forget the details of Kitces version that I was thinking of but the differences are probably:

1. Kitces using US investor and investments vs UK and international
2. Kitces using 30 years of 4% rule with many sequences vs 41 years and one sequence
3. longer bond duration in the usual US mix

Using a variable spending strategy like Guyton-Klinger would mask the effect of ignoring costs unless you experienced something near the worst case because they show up in investment performance.

jamesd, do you know of any information or studies regarding Safe Withdrawal Rates for retirement portfolios with funds aiming to provide growing natural income from dividends?

I think it would be worth understanding the downsides vs a total return approach before considering SWR
https://www.vanguard.co.uk/documents/adv/literature/total-return-investing.pdf