Pension Funds and De-Risking

DT2001

If time in the market is correct then having cash must be detrimental?

If you have a cash buffer when do you refill? Is that then timing the market?

Psychologically IMO cash is good.

I think the emphasis should be on flexibility of expenditure rather than asset allocation. If your greatest worry is sequence of return work out how to reduce that through an annuity or bond ladder. Lower returns but greater peace of mind.

zagfles

k_man said:

zagfles said:

Prism said:

k6chris said:

dunstonh said:

95% of market falls recover within 3 years (actually much less than that).   So, a cash buffer avoids sales of units in the vast majority of negative periods.   You cannot cover all eventualities but you can take reasonable steps.

Genuine question, how and when do you replace that cash buffer? Thanks.

I did a simulation of a cash buffer with the 2000 crash and it was almost impossible to replace it at any time within the first 15 years. In fact the cash buffer with a high allocation of equities didn't really work at all as you ended up holding onto all of those equites to the very bottom in 2003, at which point the buffer was gone and then you started selling equites as they began to rise (just when it would have been ideal to hold onto them). Then 2008 hit and there was no cash buffer and this was the real crash.

All that time I imagine someone would have been constantly questioning their choices on what to use that year - cash or equities. Who knew that the best approach would have been to sell equities in 2000, 2001 and 2002 because it was in fact going to go even lower.

Of course someone with a 'proper' retirement allocation of bonds and equities wouldn likely have never needed the cash buffer as they would have sold some bonds.

Yes I did similar myself, downloaded some historical SORs and using a "cash buffer" made hardly any difference to outcomes, however the criteria was tweaked. Maybe someone here who favours it/uses it can point us at an analysis of using a cash buffer strategy and how it would have fared in a wide variety of historical and international markets, together with rules and critera for using it/refloating it?

Thanks to both you and @Prism for doing simulations/ analysis.
In your simulations, did a cash buffer make outcomes significantly worse?

As above, mine made hardly any difference.

I am trying to understand why the cash buffer isn't just considered part of the non equity part of the portfolio, that just happens to be outside of the pension wrapper.

It should be. The amount you draw from your pension would usually be determined by tax, for instance you might want to draw £12570 a year to use up your personal allowance, but your overall strategy should take into account all your assets and determine whether you spend from cash, bonds or equities.

If eg you have all equities in the pension and all cash outside, and you want to spend cash rather than equities but want to draw from the pension to use up the PA, you could cash in some equities in the pension to release drawdown cash, and at the same time buy the same equities outside the pension, eg in an ISA. So you're spending cash and maintaining equities at the same level, while still drawing your PA from the pension.

zagfles

coastline said:

The last two decades highlight the problems despite decent market returns. 2000 and 2008 onwards have resulted in 50% falls before recovery. Like other posters I looked at this a while ago and bookmarked a few.
From here you can set up various portfolios including a MSCI World Index fund and FTSE World Index fund.

 The backtesting tool for European index investors · Backtest (curvo.eu)

Three asset allocations set here 100% equity , 60/40 , and 40/60.It's US based but gives you an idea what can happen. From the OP's point of view it's clear 100% equity has created some scary moments. 
First two links are simple rule of thumb 4% annual SWR and other no withdrawals. Date is 2000-2022 highlighting the 3 year market crash in 2000-2003. At the right hand side of the heading Portfolio Analysis Results there's an arrow with a Link . That's how you create an address if anyone wants to post more data to the forum thread or bookmark their work. Change the various tabs to view drawdown etc.  Worth noting you can adjust inflation under the portfolio growth chart.

Backtest Portfolio Asset Class Allocation (portfoliovisualizer.com)

Backtest Portfolio Asset Class Allocation (portfoliovisualizer.com)

Similar set up for 2007-2022.

Backtest Portfolio Asset Class Allocation (portfoliovisualizer.com)

Backtest Portfolio Asset Class Allocation (portfoliovisualizer.com)

I might have missed something but I can't see an option of a variable asset allocation, which effectively having/using a cash buffer is? Eg "if equities drop use cash buffer, if equities rise, refloat cash buffer"

k_man

zagfles said:

k_man said:

zagfles said:

Prism said:

k6chris said:

dunstonh said:

95% of market falls recover within 3 years (actually much less than that).   So, a cash buffer avoids sales of units in the vast majority of negative periods.   You cannot cover all eventualities but you can take reasonable steps.

Genuine question, how and when do you replace that cash buffer? Thanks.

I did a simulation of a cash buffer with the 2000 crash and it was almost impossible to replace it at any time within the first 15 years. In fact the cash buffer with a high allocation of equities didn't really work at all as you ended up holding onto all of those equites to the very bottom in 2003, at which point the buffer was gone and then you started selling equites as they began to rise (just when it would have been ideal to hold onto them). Then 2008 hit and there was no cash buffer and this was the real crash.

All that time I imagine someone would have been constantly questioning their choices on what to use that year - cash or equities. Who knew that the best approach would have been to sell equities in 2000, 2001 and 2002 because it was in fact going to go even lower.

Of course someone with a 'proper' retirement allocation of bonds and equities wouldn likely have never needed the cash buffer as they would have sold some bonds.

Yes I did similar myself, downloaded some historical SORs and using a "cash buffer" made hardly any difference to outcomes, however the criteria was tweaked. Maybe someone here who favours it/uses it can point us at an analysis of using a cash buffer strategy and how it would have fared in a wide variety of historical and international markets, together with rules and critera for using it/refloating it?

Thanks to both you and @Prism for doing simulations/ analysis.
In your simulations, did a cash buffer make outcomes significantly worse?

As above, mine made hardly any difference.

I am trying to understand why the cash buffer isn't just considered part of the non equity part of the portfolio, that just happens to be outside of the pension wrapper.

It should be. The amount you draw from your pension would usually be determined by tax, for instance you might want to draw £12570 a year to use up your personal allowance, but your overall strategy should take into account all your assets and determine whether you spend from cash, bonds or equities.

If eg you have all equities in the pension and all cash outside, and you want to spend cash rather than equities but want to draw from the pension to use up the PA, you could cash in some equities in the pension to release drawdown cash, and at the same time buy the same equities outside the pension, eg in an ISA. So you're spending cash and maintaining equities at the same level, while still drawing your PA from the pension.

So...
The cash buffer could just be the 'pot' you spend from, rather than or before selling bonds (when equities have risen less that 20%) if using Prime Harvesting?


And as such

DT2001 said:

If time in the market is correct then having cash must be detrimental?

If you have a cash buffer when do you refill? Is that then timing the market?

Psychologically IMO cash is good.

I think the emphasis should be on flexibility of expenditure rather than asset allocation. If your greatest worry is sequence of return work out how to reduce that through an annuity or bond ladder. Lower returns but greater peace of mind.

Refill cash when equities have reached their trigger rise level?

zagfles

bostonerimus said:

zagfles said:

Prism said:

k6chris said:

dunstonh said:

95% of market falls recover within 3 years (actually much less than that).   So, a cash buffer avoids sales of units in the vast majority of negative periods.   You cannot cover all eventualities but you can take reasonable steps.

Genuine question, how and when do you replace that cash buffer? Thanks.

I did a simulation of a cash buffer with the 2000 crash and it was almost impossible to replace it at any time within the first 15 years. In fact the cash buffer with a high allocation of equities didn't really work at all as you ended up holding onto all of those equites to the very bottom in 2003, at which point the buffer was gone and then you started selling equites as they began to rise (just when it would have been ideal to hold onto them). Then 2008 hit and there was no cash buffer and this was the real crash.

All that time I imagine someone would have been constantly questioning their choices on what to use that year - cash or equities. Who knew that the best approach would have been to sell equities in 2000, 2001 and 2002 because it was in fact going to go even lower.

Of course someone with a 'proper' retirement allocation of bonds and equities wouldn likely have never needed the cash buffer as they would have sold some bonds.

Yes I did similar myself, downloaded some historical SORs and using a "cash buffer" made hardly any difference to outcomes, however the criteria was tweaked. Maybe someone here who favours it/uses it can point us at an analysis of using a cash buffer strategy and how it would have fared in a wide variety of historical and international markets, together with rules and critera for using it/refloating it?

Here is such a study, make of it what you will.

https://www.financialplanningassociation.org/article/journal/SEP13-benefits-cash-reserve-strategy-retirement-distribution-planning

Interesting but it's US based and the main advantage of using the cash buffer seems to be transaction costs and taxes, something unlikely to be make a significant difference in the UK using funds in pensions and cash unwrapped/in ISAs. In fact I would guess taxes are likely be be higher in the UK using a cash buffer as the best interest rates for cash are unwrapped. When these are ignored, they come to the same conclusion I did, hardly any difference.

zagfles

DT2001 said:

If time in the market is correct then having cash must be detrimental?

If you have a cash buffer when do you refill? Is that then timing the market?

Yes. Any variable asset allocation strategy is timing the market, but there's a difference between short term and long term timing. The cash buffer strategy is effectively short term timing (unless the buffer is very large), whereas strategies like Prime Harvesting are long term. When you think about it "time in the market" is long term market timing, ie making the assumption that over the long term equities go up.  If you can successfully make short term market timing calls, then you should be doing it from your luxury yacht!

DT2001 said:

Psychologically IMO cash is good.

I think the emphasis should be on flexibility of expenditure rather than asset allocation. If your greatest worry is sequence of return work out how to reduce that through an annuity or bond ladder. Lower returns but greater peace of mind.

Yes, flexibility of expenditure can work together with flexibility of asset allocation. I think people will naturally do that anyway - if equities have plummetted then everyone with equities in their portfolio will suffer, and they'd probably be less likely to book that luxury cruise...

zagfles

k_man said:

zagfles said:

k_man said:

zagfles said:

Prism said:

k6chris said:

dunstonh said:

95% of market falls recover within 3 years (actually much less than that).   So, a cash buffer avoids sales of units in the vast majority of negative periods.   You cannot cover all eventualities but you can take reasonable steps.

Genuine question, how and when do you replace that cash buffer? Thanks.

I did a simulation of a cash buffer with the 2000 crash and it was almost impossible to replace it at any time within the first 15 years. In fact the cash buffer with a high allocation of equities didn't really work at all as you ended up holding onto all of those equites to the very bottom in 2003, at which point the buffer was gone and then you started selling equites as they began to rise (just when it would have been ideal to hold onto them). Then 2008 hit and there was no cash buffer and this was the real crash.

All that time I imagine someone would have been constantly questioning their choices on what to use that year - cash or equities. Who knew that the best approach would have been to sell equities in 2000, 2001 and 2002 because it was in fact going to go even lower.

Of course someone with a 'proper' retirement allocation of bonds and equities wouldn likely have never needed the cash buffer as they would have sold some bonds.

Yes I did similar myself, downloaded some historical SORs and using a "cash buffer" made hardly any difference to outcomes, however the criteria was tweaked. Maybe someone here who favours it/uses it can point us at an analysis of using a cash buffer strategy and how it would have fared in a wide variety of historical and international markets, together with rules and critera for using it/refloating it?

Thanks to both you and @Prism for doing simulations/ analysis.
In your simulations, did a cash buffer make outcomes significantly worse?

As above, mine made hardly any difference.

I am trying to understand why the cash buffer isn't just considered part of the non equity part of the portfolio, that just happens to be outside of the pension wrapper.

It should be. The amount you draw from your pension would usually be determined by tax, for instance you might want to draw £12570 a year to use up your personal allowance, but your overall strategy should take into account all your assets and determine whether you spend from cash, bonds or equities.

If eg you have all equities in the pension and all cash outside, and you want to spend cash rather than equities but want to draw from the pension to use up the PA, you could cash in some equities in the pension to release drawdown cash, and at the same time buy the same equities outside the pension, eg in an ISA. So you're spending cash and maintaining equities at the same level, while still drawing your PA from the pension.

So...
The cash buffer could just be the 'pot' you spend from, rather than or before selling bonds (when equities have risen less that 20%) if using Prime Harvesting?


And as such

DT2001 said:

If time in the market is correct then having cash must be detrimental?

If you have a cash buffer when do you refill? Is that then timing the market?

Psychologically IMO cash is good.

I think the emphasis should be on flexibility of expenditure rather than asset allocation. If your greatest worry is sequence of return work out how to reduce that through an annuity or bond ladder. Lower returns but greater peace of mind.

Refill cash when equities have reached their trigger rise level?

Yes so then you'd be using PH, whereas some people are talking about having a "cash buffer" of eg 2 or 3 year's worth of cash which they'd use when they consider markets are "low" to avoid selling equities, and refloat it when they rise. But seemingly using "finger in air" judgement calls rather than a defined strategy like PH.

k_man

Thanks that makes it clearer.

Do you know (and apologies if this is in the linked book, but we are on a forum) why 20% is chosen for PH?

Is it a sweet spot, or just a good enough round number?

bostonerimus

zagfles said:

bostonerimus said:

zagfles said:

Prism said:

k6chris said:

dunstonh said:

95% of market falls recover within 3 years (actually much less than that).   So, a cash buffer avoids sales of units in the vast majority of negative periods.   You cannot cover all eventualities but you can take reasonable steps.

Genuine question, how and when do you replace that cash buffer? Thanks.

I did a simulation of a cash buffer with the 2000 crash and it was almost impossible to replace it at any time within the first 15 years. In fact the cash buffer with a high allocation of equities didn't really work at all as you ended up holding onto all of those equites to the very bottom in 2003, at which point the buffer was gone and then you started selling equites as they began to rise (just when it would have been ideal to hold onto them). Then 2008 hit and there was no cash buffer and this was the real crash.

All that time I imagine someone would have been constantly questioning their choices on what to use that year - cash or equities. Who knew that the best approach would have been to sell equities in 2000, 2001 and 2002 because it was in fact going to go even lower.

Of course someone with a 'proper' retirement allocation of bonds and equities wouldn likely have never needed the cash buffer as they would have sold some bonds.

Yes I did similar myself, downloaded some historical SORs and using a "cash buffer" made hardly any difference to outcomes, however the criteria was tweaked. Maybe someone here who favours it/uses it can point us at an analysis of using a cash buffer strategy and how it would have fared in a wide variety of historical and international markets, together with rules and critera for using it/refloating it?

Here is such a study, make of it what you will.

https://www.financialplanningassociation.org/article/journal/SEP13-benefits-cash-reserve-strategy-retirement-distribution-planning

Interesting but it's US based and the main advantage of using the cash buffer seems to be transaction costs and taxes, something unlikely to be make a significant difference in the UK using funds in pensions and cash unwrapped/in ISAs. In fact I would guess taxes are likely be be higher in the UK using a cash buffer as the best interest rates for cash are unwrapped. When these are ignored, they come to the same conclusion I did, hardly any difference.

I think you are not reading the results correctly. There is a marginal benefit (~1% more successful outcomes) to not having a cash buffer in a no tax and no transaction cost scenario. The cash buffer wins out with transaction costs and taxation. In the UK there will be tax on pension withdrawals and many people will have transaction costs, but once you get into the details of taxation comparisons do become difficult. The utility and practicality of cash in the bank makes it a good thing for everyone to have. When you are working it's good practice to have an emergency cash buffer of maybe a year's spending and I don't see any reason to have less than that when retired and I would tilt towards more.

The danger in retirement is that people are pushed towards more risk and more aggressive investment strategies in order to compensate for a small pension pot relative to their perceived income needs. This is a big downside to the liberalization of retirement finances in the UK.

zagfles

k_man said:

Thanks that makes it clearer.

Do you know (and apologies if this is in the linked book, but we are on a forum) why 20% is chosen for PH?

Is it a sweet spot, or just a good enough round number?

It's covered in the linked book extract. It's well worth a read, as it compares other strategies too, and variants of PH.