Bucket drawdown strategy, what to put in medium term bucket?

MK62

Pat38493 said:

Linton said:

Why is "a better result"  one that maximises income or wealth at death and the psychological benefits almost an after-thought? 
 <snip>
But there is more to it than that.  In my view 2 years of income is too far too little and misses out on much potential benefit.  A major one is flexibility.  With a large low risk bucket (LRB) you can make major spending decisions such as an exotic/expensive holiday or a new car on the spot without replanning the rest of your life and reorganising your investments.
The final problem to solve is inflation.  With current interest rates the LRB should be able to generate enough income internally to largely match inflation but if necessary, rather than selling off growth assets directly for immediate consumption they may be better used to generate extra future stable income.  The net result being that your overall asset allocation will become less growth oriented over time. This is similar to the prime harvesting idea.

It's not - in fact my current plan involves something like a 2-3 year cash amount or low risk short term bonds, even though I've seen so much research that this is not the optimal long term strategy.  

Apart from anything else, we are in that rare time when cash returns are higher than inflation for the moment, and bonds are still sinking, so there is little to lose in the short term, but I will move to a more conventional approach at some point.

In fact, I am very likely on the verge of pulling all my TFC out of my largest pension fund to prepare for retirement.

I also am starting with a cash amount which is much larger than 2 years of long term rolling income need but this is mainly because I am bridging to a combined guaranteed income stream in 12 years from now which covers 90% or more of our current annual spend (and way more than 100% of essential expenses).  However I have planned big expenditures in the first 2 years including a new car and caravan and paying off at least 50% of the mortgage maybe more - hence I know that my cash outlays in the first 2 years will be a lot larger.

Raw SWR strategies and statistics are pretty useless for me as my withrawal plan ranges from negative (due to still working making large contributions this year) to 16% next year, but then ramps down to about 6.7% by the 3rd year, about 5% from year 9, but then down to 1.7% in year 14 and way less than 1% after that.  Therefore quoting 3.5% rates for 30 years doesn't really help me in any way.  My risk is between having way too much, or running out in about year 11! (even this wouldn't be catastrophic due to the guaranteed incomes mentioned above).

Also I am prepared to flex my spending if needed whereas I think you have posted in the past that you are not.

That research will also tell you that market "crashes" are relatively uncommon, and in the absence of a market crash, a cash buffer will likely cost you (in terms of lost opportunity......certainly in the era of near zero interest rates we've recently emerged from).........however, if a crash does come along.........

As to the issue of when to use the cash buffer and when to replenish it, it's hard to be definitive as are there are countless ways to implement one........but the basic portfolio drawdown decisions remain even without a cash buffer.......you still need to decide what to sell and when (though like the cash buffer, it's hard to be definitive as there are countless ways to implement a drawdown portfolio too). There are those who might decide to go down the natural yield/income path so no such decisions are  needed, but then you still need to decide what to do with excess income, and what to do in years where income falls short........no drawdown portfolio is decision free in the end

Linton

JohnWinder said:

'The bucket strategy I am proposing is addressing far more than SOR.  The requirement it aims to meet is to generate a steady inflation-matching longterm income with the option for one-off expenses'

But surely the bucket strategy is only about SOR risk if you're after steady growth for income and one-off big spending. Imagine stocks returned 2% above inflation every year without variation and without fail. To meet your needs you'd hold 100% stocks, because to do otherwise by holding another asset class that returned less than 2%/year, however steadily, would be to give you less returns for the same risk. You're proposing holding cash for one-off purchases, but the cash in my example barely keeps up with inflation; you're going to get rid of the cash with your purchase whereas you could have been holding stocks instead of cash, and then get rid of the stocks to make the purchase; all the while, the stocks earning more than the cash.

It's only because stocks and bond funds will have a variation in their returns, sometimes too large for comfort, that you need buckets, n'est pas?

1) About 50% of ongoing expenditure is covered by fixed rate annuities. Obviously the % is falling over time.

2) The missing income for ongoing expenditure comes from dividends and interest. This is more stable than capital value and involves minimal management. But it only partially matches inflation.

3) Any excess income is reinvested in PBs, WP funds, cash and in the future perhaps by a gilt ladder. I would never use bond funds for this purpose because of their capital volatility.

4)  Growth investments (100% equity) are used to ensure inflation matching by buying extra dividend and interest income funds as required. Also the PBs and low risk funds are replenished from the growth portfolio. This movement of wealth only happens occasionally. So the growth funds are generally left to grow. A crash every few years can be ignored.

5) Surely your 2% above and below inflation argument applies equally to a traditional 60/40 portfolio. What I am doing is putting a 60/40 portfolio together in a different way using the stability of income investing to remove the need for volatile growth equity investments to provide for steady ongoing expenditure.  It seems foolish to use volatile returns to cover one’s need for steady income when more appropriate alternatives are available.

tigerspill

Linton said:

I do not agree with the waterfall approach whereby wealth flows from high to medium to low risk investments.  If that is what you are doing, I believe 2 buckets is enough.  Even with that I still think that simplistic operation of a waterfall model is not the best use of ones resources.

But I do run 3 buckets....

One of my 3  buckets is designed to generate ongoing high dividend/interest so the capital value is largely irrelevent.  It certainly can be regarded as significantly lower risk than the long term bucket although it is intended for the indefinite long term with very occasional top ups to ensure inflation matching.

The low risk bucket is a mixture of cash, PBs and low risk investments. At the moment the low risk investments are Wealth Protection funds but now that interest rates are higher bonds could well be more appropriate.  Note - actual bonds, not bond funds. It is intended that all expenditure is taken from the low risk bucket so there is no switching. All cash income is paid into the low risk bucket  - my current account being considered as just a part of the low risk bucket.

The growth bucket is 100 % equity with a 10+ year outlook and the aim of ensuring that the other portfolios increase with inflation over the long term.

Hi Linton,

Could I be so cheeky as to ask what funds you are using for the first (income based) investments?  Thanks

Linton

tigerspill said:

Linton said:

I do not agree with the waterfall approach whereby wealth flows from high to medium to low risk investments.  If that is what you are doing, I believe 2 buckets is enough.  Even with that I still think that simplistic operation of a waterfall model is not the best use of ones resources.

But I do run 3 buckets....

One of my 3  buckets is designed to generate ongoing high dividend/interest so the capital value is largely irrelevent.  It certainly can be regarded as significantly lower risk than the long term bucket although it is intended for the indefinite long term with very occasional top ups to ensure inflation matching.

The low risk bucket is a mixture of cash, PBs and low risk investments. At the moment the low risk investments are Wealth Protection funds but now that interest rates are higher bonds could well be more appropriate.  Note - actual bonds, not bond funds. It is intended that all expenditure is taken from the low risk bucket so there is no switching. All cash income is paid into the low risk bucket  - my current account being considered as just a part of the low risk bucket.

The growth bucket is 100 % equity with a 10+ year outlook and the aim of ensuring that the other portfolios increase with inflation over the long term.

Hi Linton,

Could I be so cheeky as to ask what funds you are using for the first (income based) investments?  Thanks

AEW UK Reit

CT Emerging Markets Bond Ins Inc

European Assets Trust

GCP Infrastructure

Janus Henderson Asian Dividend Income I Inc

MAN GLG UK Income Prof D Inc

Premier Global Infrastructure B Inc

Schroder High Yield Opp Z Inc

Schroder US equity income maximiser Z Dis

Sequoia economic infrastructure

Pat38493

Linton said:

tigerspill said:

Linton said:

I do not agree with the waterfall approach whereby wealth flows from high to medium to low risk investments.  If that is what you are doing, I believe 2 buckets is enough.  Even with that I still think that simplistic operation of a waterfall model is not the best use of ones resources.

But I do run 3 buckets....

One of my 3  buckets is designed to generate ongoing high dividend/interest so the capital value is largely irrelevent.  It certainly can be regarded as significantly lower risk than the long term bucket although it is intended for the indefinite long term with very occasional top ups to ensure inflation matching.

The low risk bucket is a mixture of cash, PBs and low risk investments. At the moment the low risk investments are Wealth Protection funds but now that interest rates are higher bonds could well be more appropriate.  Note - actual bonds, not bond funds. It is intended that all expenditure is taken from the low risk bucket so there is no switching. All cash income is paid into the low risk bucket  - my current account being considered as just a part of the low risk bucket.

The growth bucket is 100 % equity with a 10+ year outlook and the aim of ensuring that the other portfolios increase with inflation over the long term.

Hi Linton,

Could I be so cheeky as to ask what funds you are using for the first (income based) investments?  Thanks

AEW UK Reit

CT Emerging Markets Bond Ins Inc

European Assets Trust

GCP Infrastructure

Janus Henderson Asian Dividend Income I Inc

MAN GLG UK Income Prof D Inc

Premier Global Infrastructure B Inc

Schroder High Yield Opp Z Inc

Schroder US equity income maximiser Z Dis

Sequoia economic infrastructure

Out of interest if I look at funds like that on sites like Morningstar, how can I tell which portion of the historic performance was thrown off as income versus capital growth?

Linton

Pat38493 said:

Linton said:

tigerspill said:

Linton said:

I do not agree with the waterfall approach whereby wealth flows from high to medium to low risk investments.  If that is what you are doing, I believe 2 buckets is enough.  Even with that I still think that simplistic operation of a waterfall model is not the best use of ones resources.

But I do run 3 buckets....

One of my 3  buckets is designed to generate ongoing high dividend/interest so the capital value is largely irrelevent.  It certainly can be regarded as significantly lower risk than the long term bucket although it is intended for the indefinite long term with very occasional top ups to ensure inflation matching.

The low risk bucket is a mixture of cash, PBs and low risk investments. At the moment the low risk investments are Wealth Protection funds but now that interest rates are higher bonds could well be more appropriate.  Note - actual bonds, not bond funds. It is intended that all expenditure is taken from the low risk bucket so there is no switching. All cash income is paid into the low risk bucket  - my current account being considered as just a part of the low risk bucket.

The growth bucket is 100 % equity with a 10+ year outlook and the aim of ensuring that the other portfolios increase with inflation over the long term.

Hi Linton,

Could I be so cheeky as to ask what funds you are using for the first (income based) investments?  Thanks

AEW UK Reit

CT Emerging Markets Bond Ins Inc

European Assets Trust

GCP Infrastructure

Janus Henderson Asian Dividend Income I Inc

MAN GLG UK Income Prof D Inc

Premier Global Infrastructure B Inc

Schroder High Yield Opp Z Inc

Schroder US equity income maximiser Z Dis

Sequoia economic infrastructure

Out of interest if I look at funds like that on sites like Morningstar, how can I tell which portion of the historic performance was thrown off as income versus capital growth?

Unfortunately it sems impossible to obtain historic data further back than 1-2 years  on income distributions from funds.  However you could get some idea, I guess, from comparing the performance with and without reinvestment from Trustnet/Charting.  But I havent worked out the maths of doing that since you have to remove the compounding effect.

In practice I find that the income distributions in £ terms are pretty consistent independent of share price and so that is what I monitor. 

OldScientist

Others in the thread have mentioned McClung. The first three chapters of his book are freely available at http://livingoffyourmoney.com/wp-content/uploads/2016/05/LivingOffYourOwnMoney_eBook_FirstThreeChapters.pdf with Chapter 3 presenting backtesting results for a variety of rebalancing strategies (what McClung calls income harvesting strategies) for the US, UK, and Japan. In amongst the strategies are various bucket approaches (including the three bucket strategy that is the topic here). It is interesting to note that the 3 bucket strategy was generally a poor performer compared to annual rebalancing (e.g., see Figure 18 in the book extract), even with McClung's 'generous simulation' of returns for the contents of bucket 2*, except the MSWR-100 for Japan (Figure 31, but see footnote on page 74), although the MSWR-90 for Japan is not so good (Figure 32).

* FWIW, I think the problem he had with the US corporate bond returns and why he had to simulate 'generous returns' is that the SBBI return series is for relatively long-term corporates (maturities around 20 years until 1969 and greater than 10 years after that) which had volatile returns and, therefore, didn't really meet the description for 'moderately safe' investments.