Retirement drawdown: Use cash buffer first or drawdown proportionally?

Pat38493

Linton said:

NoMore said:

Your assumption that a no bucket strategy is about chasing more gains is wrong. You reduce risk in a no bucket strategy by your overall asset allocation. 

You’ve already admitted previously that your whole portfolio is probably equivalent to a 60 40 equity split, now I’m not sure if you mean equivalent in the actual split or in the performance of but that essentially doesn’t matter So the argument is instead of bucketing just set your whole portfolio to a 60 40  asset allocation and rebalance on withdrawal. 

People seem to have an aversion to selling equities in a down turn so cash buffers and buckets are designed to help alleviate that aversion but that’s just a psychological problem, According to the no bucket strategist it doesn’t actually have any material benefit beyond that. 

Note I didn’t really want to criticise any one individuals portfolio, at the end of the day people should do what is right for themselves.

i am still currently on the fence about what to do when it comes to drawdown but are leaning towards the simple overall portfolio allocation and re balance on withdrawal method. We will see,  perhaps I am not as mentally tough as I think I am when a downturn happens. 

Selling equities in a downturn is far more than purely a psychological problem.  It is a major factor behind the concepts of the SWR and the Sequence of Returns risk, and the reason for expenditure reducing strategies like Guyton Klinger. 

Risk is controlled very broadly by overall allocations and more specifically by ones management strategy for dealing with equity volatility. The objective of any bucket strategy is to reduce the long term effects of short and medium term equity volatility. My proposals aim to take this as far as possible by reducing the need for continuous selling of equity to a minimum.

I think the point is that those who have done detailed historical simulations on this type of thing, find that using a bucket strategy didn't actually make any different to the end outcome unless you applied foresight to the withdrawals that would not have been known at the time.  Mathematically it works out about the same, because what you lost in selling a portion of your equities in a downturn, you gained in additional returns of having more equities during the entire time including good times.

As discussed, if you successfully time the situation at least to some extent, then the bucket strategy could provide a benefit, but the experts were unable to find and actual "rule" that could have been applied with existing knowledge at any moment - it would have required judgement calls based on the current political or economic global situation.

If you have a pot that is large enough that your needs are met just from income on your middle bucket, all this is largely irrelevant because you will be fine anyway.

You need to have a good attitude to risk here and you need to be able to trust that the markets will behave roughly similarly that they did during known economic history of capitalism because you could see some pretty big balance reductions in the first 5 years in SOR scenario, but you should still be ok - this is actually one of the main benefits of the bucket strategy - that you don't have to "trust" as much that your entire fund is down a lot but it will recover like it did in the worst case historical scenario.  Can you look at a scary graph that applies to you personally and trust that ecomonic reality will still apply into the future?  

Reality is though that if markets did continue to go down and down beyond anything that's ever happened before, given the same pot size, you would get into trouble anyway whether or not you have buckets.

jim8888

Qyburn said:

jim8888 said:

Hi Gazza the way I tend to look at it is by asking myself when I need to dip into the market and cash some shares to keep my cash pot with at least a year's coverage in it. At the moment, I reckon I've at least a year before I need to do that, but I am now drawing down from a SIPP and taking income from a DB pension too so I suppose technically I'm already cashing in some equities to do that. 

How are you drawing down, are you specifying which assets to cash in?

Yes, although it's quite simple for me as all my SIPP pension is held in one fund, a Vanguard 80/20 Lifestrategy. It's held in Fidelity, so they automatically sell and pay it out on a monthly basis.
If I held multiple funds, I'd have to sell off whichever fund I chose to ensure I had the cash in the SIPP for my monthly withdrawal. 

Qyburn

One point about rebalancing. When "cash buffer" was I was assuming we were talking about cash outside the pension, in which case you can't actively rebalance from cash to growth assets.  

Linton

Pat38493 said:

Linton said:

NoMore said:

Your assumption that a no bucket strategy is about chasing more gains is wrong. You reduce risk in a no bucket strategy by your overall asset allocation. 

You’ve already admitted previously that your whole portfolio is probably equivalent to a 60 40 equity split, now I’m not sure if you mean equivalent in the actual split or in the performance of but that essentially doesn’t matter So the argument is instead of bucketing just set your whole portfolio to a 60 40  asset allocation and rebalance on withdrawal. 

People seem to have an aversion to selling equities in a down turn so cash buffers and buckets are designed to help alleviate that aversion but that’s just a psychological problem, According to the no bucket strategist it doesn’t actually have any material benefit beyond that. 

Note I didn’t really want to criticise any one individuals portfolio, at the end of the day people should do what is right for themselves.

i am still currently on the fence about what to do when it comes to drawdown but are leaning towards the simple overall portfolio allocation and re balance on withdrawal method. We will see,  perhaps I am not as mentally tough as I think I am when a downturn happens. 

Selling equities in a downturn is far more than purely a psychological problem.  It is a major factor behind the concepts of the SWR and the Sequence of Returns risk, and the reason for expenditure reducing strategies like Guyton Klinger. 

Risk is controlled very broadly by overall allocations and more specifically by ones management strategy for dealing with equity volatility. The objective of any bucket strategy is to reduce the long term effects of short and medium term equity volatility. My proposals aim to take this as far as possible by reducing the need for continuous selling of equity to a minimum.

I think the point is that those who have done detailed historical simulations on this type of thing, find that using a bucket strategy didn't actually make any different to the end outcome unless you applied foresight to the withdrawals that would not have been known at the time.  Mathematically it works out about the same, because what you lost in selling a portion of your equities in a downturn, you gained in additional returns of having more equities during the entire time including good times.

As discussed, if you successfully time the situation at least to some extent, then the bucket strategy could provide a benefit, but the experts were unable to find and actual "rule" that could have been applied with existing knowledge at any moment - it would have required judgement calls based on the current political or economic global situation.

If you have a pot that is large enough that your needs are met just from income on your middle bucket, all this is largely irrelevant because you will be fine anyway.

1) What do you mean by "it worked out the same" and the "end outcome"?  Are you talkng about % returns? Wealth at death? Or are there other metrics?

The standard bucket strategies as I understand them are really waterfall models with switching between selling equities and using reserves.  Yes, in those you must overall sell to feed in at the top at the same rate as you take out of the the bottom which without unusually successful market timing of the switching could be pointless or even counter productive. Presumably this can be and has been back tested   But it is irrelevent since that is is not what is being proposed here.

2)I am not sure I have successfully explained to you what the strategy is.  You talk about selling equities in a downturn.  With my current income needs I never sell equities for cash in a downturn or any other time on an ongoing basis.   Were I to need significantly more income then it could necessary to regularly sell equities but the % sold would be far less than with a single pot model and so would lead to a larger SWR.

When money is moved from the growth bucket  to the income bucket equities may well be sold to buy different equities so there is no market timing involved.

Bostonerimus1

Rather than chronological buckets I think about assets and sources of income, but they are just another sort of bucket and it comes down to personal preference. In order of which to access first my asset buckets go:

income from part time work;
rental income;
SP, DB pensions and annuities;
dividends and interest;
cash buffer;
capital held in equities and bonds;
home equity.

Also rather than a fixed withdrawal rate, studies have shown that by varying the withdrawal rate between upper and lower guardrails according to market conditions you can increase your likelihood of success. Rebalancing can also be implemented using guardrails like a +/-10 deviation from your set asset allocation ratios.

coyrls

Linton said:

GazzaBloom said:

Linton said:

GazzaBloom said:

@Linton I'm still unclear as to how your approach works in practice, you say you use the middle mid term bucket to generate income, does that need to generate enough income to cover topping up a full years drawdown from the cash buffer? If so, at say 5% income generation it would need to be a very large bucket. To avoid the risk of having to sell stocks as they fall your income bucket needs generate enough income to cover your annual expenses doesn't it?

for example, let's say you need to draw down £30K a year, how would you position assets and how much in each?

Bucket 1 £90K  - 3 years worth of cash or cash equivalent
Bucket 3 £600K - income generating stocks, bonds? Giving say 5% (£30K)
Bucket 3 ?? - growth stocks for inflation matching/beating?

Yes because  of the bottleneck of income generation for higher % withdrawals it will be necessary to have some growth portfolio to cash transfer possibly taken annually. However this would be at a much lower level than if all income was taken from one portfolio. Taking your example a drawdown of £30k could require a pot of £900k.

So one could split the portfolio as say 

Bucket 1 £100k generating £2.5k interest
Bucket. 2 £450 k generating £22.5k income
Bucket  3 £350k generating £5k to ensure that bucket 2 increases with inflation.

Bucket 1 need only or mainly cover the money coming from growth since income is assumed to be safe over the short to medium trrm. So ignoring inflation Bucket 1 could cover say >10 years, or more usefully significant one-off expenditure.

This is from just an example from a few minutes thought. There is a wide range of flexibility in the numbers.

Thanks.

So, a portfolio allocated (roughly using your example) 10% cash / 50% income (bonds/income stocks) / 40% growth stocks.

I still can't get my head round any real advantage over drawing 5% proportionally from each allocation and rebalancing annually other than being able to strategically choose timing when you drawdown from the growth stocks allocation.

Yes to the asset allocation. So if the income portfolio was 50% equity that would give rather higher than 60% equity overall.  Also cash is not necessarily cash.  In my case perhaps 50% is currently Wealth Peservation funds though now that 5% interest is available I am considering moving over to fixed interest. 

Some of the advantages for me are:
1) All ongoing activity can be automated.  In particular income from dividends/interest is paid directly from an S&S ISA into my current account, which is regarded as part of the cash bucket. So in practice not very different to extra pension.  Top level management is only required every few years.
2) A large amount of near to cash available for flexibility in one-off expenses.  This is thanks to the relative safety of income investments.  So we can take major holidays when we feel like it without pre-planning finances or waiting to have saved up enough money.
3) There is no direct link from selling investments to day-to-day expenditure.  All income both generated and from guaranteed pensions goes into the cash bucket and all expenditure comes from the cash bucket.  This means that selling of investments is never required to maintain income over the medium term.  This is the "no sleepless nights" benefit as the short/medium term performance of the growth bucket can be largely ignored as can the capital value of the income bucket. 
4) Since buckets are designed for a specific job their asset allocations and sizes are pretty much predetermined.The overall asset allocation is automatically what it needs to be to meet all one's requirements.  On what basis do you decide on the %'s in a single portfolio intended to do everything?

One downside is that it takes significant work to set up in the first place in designing and sizing the buckets to meet their objectives.

 - 

What is your current drawdown rate, ie what percentage of the total of all your buckets are you taking as income?  I suspect this is the metric that has the biggest effect on success, regardless of the nuances of how investments are categorised.  You mention "expenditure reducing strategies like Guyton Klinger" but these reducing strategies were devised to allow a higher initial drawdown rate.  If your initial drawdown rate is sufficiently low, there is no requirement for expenditure reducing strategies.

Bostonerimus1

coyrls said:

Linton said:

GazzaBloom said:

Linton said:

GazzaBloom said:

@Linton I'm still unclear as to how your approach works in practice, you say you use the middle mid term bucket to generate income, does that need to generate enough income to cover topping up a full years drawdown from the cash buffer? If so, at say 5% income generation it would need to be a very large bucket. To avoid the risk of having to sell stocks as they fall your income bucket needs generate enough income to cover your annual expenses doesn't it?

for example, let's say you need to draw down £30K a year, how would you position assets and how much in each?

Bucket 1 £90K  - 3 years worth of cash or cash equivalent
Bucket 3 £600K - income generating stocks, bonds? Giving say 5% (£30K)
Bucket 3 ?? - growth stocks for inflation matching/beating?

Yes because  of the bottleneck of income generation for higher % withdrawals it will be necessary to have some growth portfolio to cash transfer possibly taken annually. However this would be at a much lower level than if all income was taken from one portfolio. Taking your example a drawdown of £30k could require a pot of £900k.

So one could split the portfolio as say 

Bucket 1 £100k generating £2.5k interest
Bucket. 2 £450 k generating £22.5k income
Bucket  3 £350k generating £5k to ensure that bucket 2 increases with inflation.

Bucket 1 need only or mainly cover the money coming from growth since income is assumed to be safe over the short to medium trrm. So ignoring inflation Bucket 1 could cover say >10 years, or more usefully significant one-off expenditure.

This is from just an example from a few minutes thought. There is a wide range of flexibility in the numbers.

Thanks.

So, a portfolio allocated (roughly using your example) 10% cash / 50% income (bonds/income stocks) / 40% growth stocks.

I still can't get my head round any real advantage over drawing 5% proportionally from each allocation and rebalancing annually other than being able to strategically choose timing when you drawdown from the growth stocks allocation.

Yes to the asset allocation. So if the income portfolio was 50% equity that would give rather higher than 60% equity overall.  Also cash is not necessarily cash.  In my case perhaps 50% is currently Wealth Peservation funds though now that 5% interest is available I am considering moving over to fixed interest. 

Some of the advantages for me are:
1) All ongoing activity can be automated.  In particular income from dividends/interest is paid directly from an S&S ISA into my current account, which is regarded as part of the cash bucket. So in practice not very different to extra pension.  Top level management is only required every few years.
2) A large amount of near to cash available for flexibility in one-off expenses.  This is thanks to the relative safety of income investments.  So we can take major holidays when we feel like it without pre-planning finances or waiting to have saved up enough money.
3) There is no direct link from selling investments to day-to-day expenditure.  All income both generated and from guaranteed pensions goes into the cash bucket and all expenditure comes from the cash bucket.  This means that selling of investments is never required to maintain income over the medium term.  This is the "no sleepless nights" benefit as the short/medium term performance of the growth bucket can be largely ignored as can the capital value of the income bucket. 
4) Since buckets are designed for a specific job their asset allocations and sizes are pretty much predetermined.The overall asset allocation is automatically what it needs to be to meet all one's requirements.  On what basis do you decide on the %'s in a single portfolio intended to do everything?

One downside is that it takes significant work to set up in the first place in designing and sizing the buckets to meet their objectives.

 - 

What is your current drawdown rate, ie what percentage of the total of all your buckets are you taking as income?  I suspect this is the metric that has the biggest effect on success, regardless of the nuances of how investments are categorised.  You mention "expenditure reducing strategies like Guyton Klinger" but these reducing strategies were devised to allow a higher initial drawdown rate.  If your initial drawdown rate is sufficiently low, there is no requirement for expenditure reducing strategies.

You can set your GK guard rails where you like. If you are lucky enough to have a very sustainable WR like <2% then withdrawal algorithms become a bit redundant and you should be able to implement the age old strategy of living off dividends and interest and you can ignore total return and let the capital gains compound and be immune to most capital losses. GK is useful for many people who don't have vast pension pots and ~4% drawdown would cause them hardship. Implementing a variable withdrawal strategy can safely increase their drawdown amounts, but they have to accept there might be some years of real hardship and have strategies in place to cope - discipline is important.

Linton

coyrls said:

Linton said:

GazzaBloom said:

Linton said:

GazzaBloom said:

@Linton I'm still unclear as to how your approach works in practice, you say you use the middle mid term bucket to generate income, does that need to generate enough income to cover topping up a full years drawdown from the cash buffer? If so, at say 5% income generation it would need to be a very large bucket. To avoid the risk of having to sell stocks as they fall your income bucket needs generate enough income to cover your annual expenses doesn't it?

for example, let's say you need to draw down £30K a year, how would you position assets and how much in each?

Bucket 1 £90K  - 3 years worth of cash or cash equivalent
Bucket 3 £600K - income generating stocks, bonds? Giving say 5% (£30K)
Bucket 3 ?? - growth stocks for inflation matching/beating?

Yes because  of the bottleneck of income generation for higher % withdrawals it will be necessary to have some growth portfolio to cash transfer possibly taken annually. However this would be at a much lower level than if all income was taken from one portfolio. Taking your example a drawdown of £30k could require a pot of £900k.

So one could split the portfolio as say 

Bucket 1 £100k generating £2.5k interest
Bucket. 2 £450 k generating £22.5k income
Bucket  3 £350k generating £5k to ensure that bucket 2 increases with inflation.

Bucket 1 need only or mainly cover the money coming from growth since income is assumed to be safe over the short to medium trrm. So ignoring inflation Bucket 1 could cover say >10 years, or more usefully significant one-off expenditure.

This is from just an example from a few minutes thought. There is a wide range of flexibility in the numbers.

Thanks.

So, a portfolio allocated (roughly using your example) 10% cash / 50% income (bonds/income stocks) / 40% growth stocks.

I still can't get my head round any real advantage over drawing 5% proportionally from each allocation and rebalancing annually other than being able to strategically choose timing when you drawdown from the growth stocks allocation.

Yes to the asset allocation. So if the income portfolio was 50% equity that would give rather higher than 60% equity overall.  Also cash is not necessarily cash.  In my case perhaps 50% is currently Wealth Peservation funds though now that 5% interest is available I am considering moving over to fixed interest. 

Some of the advantages for me are:
1) All ongoing activity can be automated.  In particular income from dividends/interest is paid directly from an S&S ISA into my current account, which is regarded as part of the cash bucket. So in practice not very different to extra pension.  Top level management is only required every few years.
2) A large amount of near to cash available for flexibility in one-off expenses.  This is thanks to the relative safety of income investments.  So we can take major holidays when we feel like it without pre-planning finances or waiting to have saved up enough money.
3) There is no direct link from selling investments to day-to-day expenditure.  All income both generated and from guaranteed pensions goes into the cash bucket and all expenditure comes from the cash bucket.  This means that selling of investments is never required to maintain income over the medium term.  This is the "no sleepless nights" benefit as the short/medium term performance of the growth bucket can be largely ignored as can the capital value of the income bucket. 
4) Since buckets are designed for a specific job their asset allocations and sizes are pretty much predetermined.The overall asset allocation is automatically what it needs to be to meet all one's requirements.  On what basis do you decide on the %'s in a single portfolio intended to do everything?

One downside is that it takes significant work to set up in the first place in designing and sizing the buckets to meet their objectives.

 - 

What is your current drawdown rate, ie what percentage of the total of all your buckets are you taking as income?  I suspect this is the metric that has the biggest effect on success, regardless of the nuances of how investments are categorised.  You mention "expenditure reducing strategies like Guyton Klinger" but these reducing strategies were devised to allow a higher initial drawdown rate.  If your initial drawdown rate is sufficiently low, there is no requirement for expenditure reducing strategies.

This is a difficult question as there are significant one-offs that confuse the issue. Hence the cash holdings are far larger than they need be for supporting ongoing expenditure. The cash bucket has been irrelevent for ongoing income for at least 4 years and probably much longer.

  So if we just look at the growth and income buckets the annual drawdown to cover ongoing expenditure is currently  entirely in the form of dividends/interest  and comprises 3% of the total invested assets.  So rather low compared with an SWR but not trivial.  It will increase in the future as inflation further devalues my fixed rate annuities,