Why buy annuity

Cobbler_tone

Moonwolf said:

Cobbler_tone said:

They didn’t go out for dinner and fish & chips was our treat…well that costs a bomb nowadays!

I was born in the mid 60s.

As part of a divorced family we went out quite a lot with grandparents and the other parent.  This included Little Chef, Kardomah, Bernie Inn etc.  

Admittedly. holidays weren't abroad, usually we went to stay with family friends who lived on the coast.  My grandfather had a boss at work who had a cottage in Wales we used to go to off season, no idea what, if anything was paid.

We bought our first house 36 years ago and our mortgage rate was 15% which is fine when you have budgeted for it although it made us cautious and one of the reasons we didn't ramp-up is I always worried about rates jumping again.  Someone recently who's fixed rate at 2% ran out and had to take 6% must have found it difficult.

We used to 'dream' of going to the Bernie Inn for a fruit juice starter! 
My childhood memories are of camper van trips to Cornwall.

I think my highest mortgage rates were 7%(ish) in the late 90's. I know my first house (£40k) would now set you back £240k. It's mad thinking back that I once didn't up-size as the house was another £10k. That's all relative though and heard a similar story (to a much lower level) with my parents. If you knew then what you know now! 

I see the other changing attitude is around inheritance. There seems to be more talk (certainly on here) about gifting money to kids and worrying about leaving wealth beyond the grave. My parents and others of a certain age are more 'you get it when they're gone', dependant of whether it has been spent on care. 

zagfles

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

zagfles

OldScientist said:

Cobbler_tone said:

zagfles said:

Cobbler_tone said:

zagfles said:

Cobbler_tone said:

But £25k (for a couple) are protected from inflation to a certain extent, unless you mean if the state pension is frozen, which is a different conversation. Hence why those in later life are normally more concerned about savings rates as opposed to mortgage rates and inflation.

I’d say the families pulling in £40k a year, with a hefty mortgage and 3 kids, running a car etc are more vulnerable to large spikes in inflation. People in that category must be really feeling the pinch. 

The family on £40k would likely get pay increases in line with inflation

Not many working folk kept up with inflation recently. Hence the ‘cost of living crisis’, which I’m not sure was an actual thing judging by the queues at Costa.

Over the longer term they did. The "cost of living crisis" as you imply was mainly a soundbite for sensationalist journalists and politicians trying to score political points. 

It is always fun to compare previous generations to current and future ones…but it is impossible.

They didn’t go out for dinner and fish & chips was our treat…well that costs a bomb nowadays!
I think consumer choice is the key driver why so many people struggle these days. Their kids £100 trainers and £40 a month on iPhones, to go with the PS5 and latest tablet PC.

Thankfully most people have been priced out of smoking!

Way OT, but I suspect it is house prices and rents (with much less in the way of council housing as an alternative to the private sector) compared to salary that has had the largest impact on budgets over the last 50 or more years. For working people, the high inflation of the 70s was OK since wages largely kept up but it had a significant effect on those living off their assets (at least until the 80s). Over the last decade, the opposite has largely been the case, wages have not kept up with inflation, but asset values have.

Thanks to the removal of credit restrictions, a lot of major purchases in the 60s and 70s were made on the never never, so monthly outgoings are not a new thing! I'd suggest that a mobile is now pretty well essential and not the luxury that it once was (although to be fair, there are cheaper models than the iphone).

Yes, in fact wages were one of the causes of it with the wage-price sprial (ie prices rise, so wage demands rise, so prices rise etc). 

Like with mortgages, buying stuff on the "never never" wasn't nearly so bad then as now because the real value of the debt was being reduced, what cost a month's wage in 1970 was a week's wage in 1980. 

High inflation is generally OK for working people but bad for people with assets. 

MK62

zagfles said:

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

With respect, what has never been about what gives the highest lifetime income?

Cobbler_tone

After clocking the threads from the person who wouldn't use Vista print....
I think they may be the type who would 'cash in' a £20k a year DB pension for £400k in the back burner.   

Acromion

Cobbler_tone said:

…I see the other changing attitude is around inheritance. There seems to be more talk (certainly on here) about gifting money to kids and worrying about leaving wealth beyond the grave. My parents and others of a certain age are more 'you get it when they're gone', dependant of whether it has been spent on care. 

Since the imposition of IHT on pensions I’ve started to think about gifting, but at the back of my mind is the fact that , if you are fortunate enough to live long, funds steadily deplete in old age.

My parents were moderately prosperous, dad drove a Bentley at one stage, they stayed 5 star in London, took us on skiing holidays, paid for private education for my brother and me. My father died age 84, and my mother, 10 years later, also 84. Neither required “care” other than for a few terminal weeks. Yet their estate did not trouble the (then) IHT threshold of £250,000. The same was true of my wife’s family, comfortably off retired civil servants.

I suspect that for most of us excessive “gifting” is probably unnecessary and quite possibly counter productive. And that annuities, by removing a big chunk of the inheritance taxable estate, are seeing a new dawn - in appropriate circumstances etc etc.

LHW99

MK62 said:

zagfles said:

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

With respect, what has never been about what gives the highest lifetime income?

A high income when retired is nice, but what I aim for is a retirement income that a) isn't going to drop like a stone when I can't do anything about it. b) that ensure that I and the OH will have a reasonably comfortable income, whichever of us goes first and c) doesn't require too much thinking about / form filling as the marbles keep disappearing!

That may not in fact be the highest possible lifetime income - but it'll do.

westv

LHW99 said:

MK62 said:

zagfles said:

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

With respect, what has never been about what gives the highest lifetime income?

A high income when retired is nice, but what I aim for is a retirement income that a) isn't going to drop like a stone when I can't do anything about it. b) that ensure that I and the OH will have a reasonably comfortable income, whichever of us goes first and c) doesn't require too much thinking about / form filling as the marbles keep disappearing!

That may not in fact be the highest possible lifetime income - but it'll do.

I think that probably covers most of us on here.

MK62

westv said:

LHW99 said:

MK62 said:

zagfles said:

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

With respect, what has never been about what gives the highest lifetime income?

A high income when retired is nice, but what I aim for is a retirement income that a) isn't going to drop like a stone when I can't do anything about it. b) that ensure that I and the OH will have a reasonably comfortable income, whichever of us goes first and c) doesn't require too much thinking about / form filling as the marbles keep disappearing!

That may not in fact be the highest possible lifetime income - but it'll do.

I think that probably covers most of us on here.

Me included.........for me, it's not just about maximising income absolutely, rather maximising income commensurate with the level of the perceived risks I'm willing to take on.......that most likely will not be the absolute highest possible (which, of course, we'll only know in  XX years time).....I don't think anyone on here is willing to dial up the risk to bitcoin level though.
Equally it's not just about what is "safe" either.........though what exactly "safe" means here may be the bone of contention.

Cobbler_tone

LHW99 said:

MK62 said:

zagfles said:

OldScientist said:

zagfles said:

OldScientist said:

Finally, since I said I'd include another couple of historical cases, here are 1910 (fairly bad for stock/bond portfolio) and 1970 (actually OK for stock/bond portfolio - note the target WR of 5% instead of 3.5% - but horrible inflation). In each case, the highest annuity payout rate is for a single life level annuity aged 67 based on historic yields and modern mortality rates.





For the 1910 case there was deflation during the 1920s (that's why the income from the annuity goes back up - IIRC, the last occasion when we had some deflation in RPI was 2008). The large effect of inflation on the annuity income during the 1970s can also be seen and the annuity only marginally improves things even at the highest payout rate. However, in each of the example cases, provided the payout rate was high enough, the level annuity improved the outcome compared to just running drawdown. In some respects, this should not be too surprising since the retiree is swapping bonds held in their portfolio for a collapsing bond ladder held by the insurance company (with a small boost due to mortality rates).

I've looked at this in a more systematic way and found that provided the payout rate for a level annuity exceeded a threshold of somewhere between 5.5 and 6.0% it would have improved the outcomes for the worst historical cases in the UK.

I might have missed something but the 1970 graphs seem to show income plummetting in about 1992/3 with the pure drawdown but 1987 with the DD/8.5% annuity. Or are you adding a couple of decades of the continuing 0.5% or so annuity income? 

It is the 'providing the annuity payout was high enough' part of the highlighted sentence that is critical. With a payout of 10.5%, the portfolio expired about the same time as it did with no annuity - in other words the payout needed to be higher than that to improve the longevity of the full income (at the time, the annuity rates for 65yo males were about 14% because of the shorter life expectancies). While there is residual annuity income, in real terms, it is definitely in the 'slightly better than nothing' category!

It comes back to what I think we all agree on - it is not possible at the start of retirement to predict whether relying solely on drawdown or buying an annuity of either level or RPI type will, by the end of retirement have provided the most income. In other words, the choice is more about securing the properties of income profile (e.g., constant floor, front loading, etc.) required than maximising. There is also some element of suiting the required complexity.

Indeed, it's never been about what gives the highest lifetime income, which as you say is unpredictable, but about what is "safe" in the same way as SWRs. 

"Safe" obviously needs a different definition for flat annuities as you know for certain your income will fall in real terms, so a fairly simple definition may be that real income doesn't fall to less than 50% over 20 years, or 30% over 30. Actually probably better to do it by reference to the initial pot value, so maybe 2 or 3% of the pot. 

Be interesting to compare with drawdown in terms of "safety", but instead of using SWRs start with the sort of amount a flat annuity would pay, eg 7.5% of the pot, and reduce the draw in real terms annually by 4%, sort of replicating a flat annuity with average inflation, but then once the minimum is hit (eg 2 or 3% of pot), keep the real draw fixed. Does that succeed more often than a flat annuity? 

If anything front loaded drawdown should be safer generally than fixed as you're drawing more earlier, but there will be exceptions. 

With respect, what has never been about what gives the highest lifetime income?

A high income when retired is nice, but what I aim for is a retirement income that a) isn't going to drop like a stone when I can't do anything about it. b) that ensure that I and the OH will have a reasonably comfortable income, whichever of us goes first and c) doesn't require too much thinking about / form filling as the marbles keep disappearing!

That may not in fact be the highest possible lifetime income - but it'll do.

Amen brother/sister!