To SPIA or not to SPIA - a new study

[mathjak107]

how ironic that a joint venture between dr pfau and michael kitces came to the conclusion spending other sources of income first while allowing stocks to grow grew more money than conventionally drawing from stocks early on.


Understanding The True Impact Of Single Premium Immediate Annuities On Retirement Income Sustainability - Kitces | Nerd's Eye View


Mod note - this thread was hived off from the Ray Lucia banning thread.

 

[audreyh1]

how ironic that a joint venture between dr pfau and michael kitces came to the conclusion spending other sources of income first while allowing stocks to grow grew more money than conventionally drawing from stocks early on.


Understanding The True Impact Of Single Premium Immediate Annuities On Retirement Income Sustainability - Kitces | Nerd's Eye View

Wow - this is a big deal and deserves its own thread!

Pretty much throws out SPIA supplemental strategies. When to take SS can be optimized for longevity risk, but otherwise forget the SPIA.

I'll have to seriously consider NOT reducing my equity exposure as I age. Much more study required, of course, but this wakes me up!

Kitces captures some gems.

 

[mathjak107]

i think that was one of the most eye opening studies in a long time.

 

[NW-Bound]

I have always planned to maintain constant AA as I age, and not to cut down on equities. This is now affirmed as a reasonable thing to do.

 

[tulak]

Wow - this is a big deal and deserves its own thread!

Pretty much throws out SPIA supplemental strategies. When to take SS can be optimized for longevity risk, but otherwise forget the SPIA.

I'll have to seriously consider NOT reducing my equity exposure as I age. Much more study required, of course, but this wakes me up!

Kitces captures some gems.

+1 - this deserves it's own thread.

Great read. Mathjak, thanks for sharing the link.

 

[Katsmeow]

how ironic that a joint venture between dr pfau and michael kitces came to the conclusion spending other sources of income first while allowing stocks to grow grew more money than conventionally drawing from stocks early on.


Understanding The True Impact Of Single Premium Immediate Annuities On Retirement Income Sustainability - Kitces | Nerd's Eye View

This was really interesting. He mentions in the article another article finding that traditional cash bucket strategies to keep from drawing from the portfolio during bear markets don't work as well as more typical portfolios that don't have the cash buckets because putting so much money into a cash bucket is an overall drag on performance.

He says in the SPIA article:

n the other hand, it's notable that at least this SPIA form of bucket strategy was improving retirement outcomes, a notable difference from cash reserve "buffer zone" bucket strategies that have been shown to decrease income sustainability!

However, he doesn't explain why the SPIA form of bucket strategy is successful while the typical cash bucket strategy isn't. I guess it is because in the typical cash bucket strategy you rebalance and always keep the cash bucket forever while with the SPIA form of bucket strategy you don't replenish the cash bucket and you don't rebalance. However, I wish this was more explicitly explained.

 

[audreyh1]

This was really interesting. He mentions in the article another article finding that traditional cash bucket strategies to keep from drawing from the portfolio during bear markets don't work as well as more typical portfolios that don't have the cash buckets because putting so much money into a cash bucket is an overall drag on performance.

He says in the SPIA article:



However, he doesn't explain why the SPIA form of bucket strategy is successful while the typical cash bucket strategy isn't. I guess it is because in the typical cash bucket strategy you rebalance and always keep the cash bucket forever while with the SPIA form of bucket strategy you don't replenish the cash bucket and you don't rebalance. However, I wish this was more explicitly explained.

In one case they're using cash buckets, in another they are using bonds. And, yes, they aren't replenishing the bond allocation either, but letting it dwindle down as it is being drawn from - thus the AA keepings getting more aggressive with time.

 

[audreyh1]

I have always planned to maintain constant AA as I age, and not to cut down on equities. This is now affirmed as a reasonable thing to do.

Especially as you are already at 70% equities!

 

[audreyh1]

Thanks Mathjak for pulling this section out into a new thread where it can gain a wider audience!

 

[urn2bfree]

I started a thread which was an offshoot of that paper about cash buffers( which I only found by reading about the SPIA paper mentioned in this thread:

http://www.early-retirement.org/forums/f28/cash-buffer-yields-67618.html

I argue that the cash buffer bashing makes some assumptions about cash yields which could be unrealistically or irresponsibly low. With higher yielding cash vehicles I wonder if the buffer performs better.

 

[ClockWatcher]

Thank you for highlighting Kitces and Pfau’s latest research. Their study supports guidance published by Laurence J. Kotlikoff and Scott Burns in their 2008 book, Spend ‘Til the End. On page 248 they express,

Uncertainty about benefits is greatest in our late forties and fifties, when the benefits are close enough to smell and big enough to matter, but not yet available to touch. Once we hit our golden years and start collecting these benefits, it’s not likely that the government will snatch them away. This pattern of benefit uncertainty with age suggests reducing your stock holdings as you approach retirement but increasing them thereafter. [emphasis added]

They take this a step further by saying,

Health expenditure uncertainty is greatest when we reach older age, roughly age eighty. This uncertainty suggests a smaller allocation to stocks in our advanced years.

 

[audreyh1]

Thanks Mathjak for pulling this section out into a new thread where it can gain a wider audience!

Thanks to Moderator also.

 

[obgyn65]

My spreadsheet still includes SPIAs bought in 21 years, after age 70. Of course I reserve the right to change my mind by then if I am still alive

 

[Marc]

i asked about this idea over three years ago (see post 83)

http://www.early-retirement.org/forums/f28/immediate-annuities-43636-5.html

and I didn't get any positive responses. I still think a SPIA can substitute for a large bond allocation and also allow you to increase SWR on the non-SPIA portion of your portfolio.

As an additional point, I think postponing SS past 62 also allows you to increase SWR from 4% to 5% for those years from 62 to eventually taking out SS then lowering by same amount as SS received (or some adjustments to maintain same long term risk).

Marc

 

[Midpack]

Thanks mathjak. Isn't that interesting, Pfau (who I've read with interest over the past few years) has seemed more and more inclined to advocate for annuitization/"floor income" (without any commercial interest, so genuinely IMO) - and now Kitces names him as a co-author discouraging SPIAs except for extreme longevity cases. Once again:

• Statistics can be applied to anything, doesn't mean the results are meaningful, aka GIGO.
• You can find a convincing argument for almost anything you want to believe on the internet.
• Caveat emptor...

My plan remains, I plan/hope to avoid buying a SPIA (especially at today's historically high costs/low returns), but I have not ruled it out when we get to our 70-80's. Unfortunately (from a financial planning POV) we do expect to live into our 90's based on both family histories...

 

[Gearhead Jim]

I noticed that the paper talked about using Monte Carlo simulations, but (I was reading quickly) did not notice anything about using Firecalc or a similar historically-based method of testing.

If your portfolio was heavily in equities and you were 75 years old when the market took a deep and prolonged dump, seems like you might have lived the rest of your life in poverty.

Other sources of portfolio "stability" are not just bonds now, but also include CD's, foreign bonds, SPIAs, etc.

Discussion?

 

[Animorph]

I noticed that the paper talked about using Monte Carlo simulations, but (I was reading quickly) did not notice anything about using Firecalc or a similar historically-based method of testing.

If your portfolio was heavily in equities and you were 75 years old when the market took a deep and prolonged dump, seems like you might have lived the rest of your life in poverty.

Other sources of portfolio "stability" are not just bonds now, but also include CD's, foreign bonds, SPIAs, etc.

Discussion?

Hopefully you had 10 years of great equity gains before you hit 75. That may have reduced your WR. And at 75 you only have to plan for 20-22 years remaining and can sustain a higher WR. And you should be internationally diversified as well. So a U.S. market bear at 75 is not the end of the world.

 

[Running_Man]

Hopefully you had 10 years of great equity gains before you hit 75. That may have reduced your WR. And at 75 you only have to plan for 20-22 years remaining and can sustain a higher WR. And you should be internationally diversified as well. So a U.S. market bear at 75 is not the end of the world.

This just does not seem reasonable considering the individuals who will be making these decisions. Really a 75-90 % Stock allocation at age 75 when mental facilities are in decline and need for money for medical resources and potentially high cost assisted living will be at it's peak? Stock markets can and will decline 80-90 percent or more, I do not see how a retiree could possibly endure this and to not plan for this contiginecy is reckless to me. That this is being marketed to financial planners will lead to too many individuals who will not have the intestinal fortitude to be able to stick to the plan when the super bear market comes along (despite the fact that Helicopter Ben believes he can keep them away forever).

I have not looked into detail in the study other than to skim the surface, but superficially the logic appears flawed to me. I will have to try in the next couple of days to study the details quite a bit more. This to me is akin to a pension fund claiming that over the long term they will earn 8 percent per year and therefore aren't underfunded while depleting their base funds (see Illinois pension fund declining to a less than 45 percent funding level by next year).

 

[Midpack]

Not sure he said that exactly, but if for example one has a COLAd pension that takes care of all your income needs, you can have a very high equity allocation at an advanced age. Too bad we usually don't know the details that accompany many posts...we often don't know who's FIRE, who's SIRE, or where in between.

 

[RunningBum]

This just does not seem reasonable considering the individuals who will be making these decisions. Really a 75-90 % Stock allocation at age 75 when mental facilities are in decline and need for money for medical resources and potentially high cost assisted living will be at it's peak? Stock markets can and will decline 80-90 percent or more, I do not see how a retiree could possibly endure this and to not plan for this contiginecy is reckless to me.

Really? Stock markets will decline 80% or more? Really?

I probably won't be increasing my equities as I get older. I don't like the idea of taking on increasing risk as I get older. I may have to think about this more, but I expect by age 75 I'll feel quite certain that my plan is working and I won't need high returns and the risk that go along with them.

 

[Running_Man]

OK in looking at the actual data at thirty years using the returns from the Evensky study for a 4% withdrawal for a 65 year old male the first scenario used by Michael and Wade, the new method has a 20.8 percent chance of depletion in 30 years, with a 50/50 starting stock bond portfolio increasing over time while the SPIA has a 7.6 percent chance of depletion. The 20.8 is a reduction from the 26.8 percent depletion rate with a straight 50/50 portfolio composition. Of course a depleted portfolio with an SPIA continues to pay the SPIA amount out versus zero for the alternative strategy.

The major advantage gained is primarily in larger portfolio value in death before age 85 the 20 year mark, which to me pretty much eliminates the actual need of the SPIA. As at the 20 year mark the depletion chances are about even at about 1.3 percent, versus the 3.6% for a straight up 50/50 portfolio.

 

[Gearhead Jim]

Hopefully you had 10 years of great equity gains before you hit 75. That may have reduced your WR. And at 75 you only have to plan for 20-22 years remaining and can sustain a higher WR. And you should be internationally diversified as well. So a U.S. market bear at 75 is not the end of the world.

This just does not seem reasonable considering the individuals who will be making these decisions. Really a 75-90 % Stock allocation at age 75 when mental facilities are in decline and need for money for medical resources and potentially high cost assisted living will be at it's peak? Stock markets can and will decline 80-90 percent or more, I do not see how a retiree could possibly endure this and to not plan for this contiginecy is reckless to me. That this is being marketed to financial planners will lead to too many individuals who will not have the intestinal fortitude to be able to stick to the plan when the super bear market comes along (despite the fact that Helicopter Ben believes he can keep them away forever).

I have not looked into detail in the study other than to skim the surface, but superficially the logic appears flawed to me. I will have to try in the next couple of days to study the details quite a bit more. This to me is akin to a pension fund claiming that over the long term they will earn 8 percent per year and therefore aren't underfunded while depleting their base funds (see Illinois pension fund declining to a less than 45 percent funding level by next year).

So can somebody use Firecalc or a similar historical calculator to see how the "new" system of increasing equity in the AA as you get older, would have worked for a 60 yr old who retired in 1929 and lived for 35 years?

 

[bmcgonig]

Thanks mathjak. Isn't that interesting, Pfau (who I've read with interest over the past few years) has seemed more and more inclined to advocate for annuitization/"floor income" (without any commercial interest, so genuinely IMO) - and now Kitces names him as a co-author discouraging SPIAs except for extreme longevity cases. Once again:

[*]Statistics can be applied to anything, doesn't mean the results are meaningful, aka GIGO.
[*]You can find a convincing argument for almost anything you want to believe on the internet.
[*]Caveat emptor...
My plan remains, I plan/hope to avoid buying a SPIA (especially at today's historically high costs/low returns), but I have not ruled it out when we get to our 70-80's. Unfortunately (from a financial planning POV) we do expect to live into our 90's based on both family histories...

Midpack

How do you intend to use an SPIA when you're 80 if you don't mind me asking? I.e., what will determine the decision and how much? And what are your thoughts on a longevity annuity that would start at 80/85. The latter is what I was considering as my mother is 95 and kicking. Also, I like that it almost solves the question of how many years should I plan for I.e., up till age 80/85. The big problem with them is that you have to estimate how much you would need by guessing the future inflation rate as they're not cola'd

One recent example I've seen is a 65 yr old man putting down 100000 would see 65000 per year when he hits 85. At 3% inflation that would be around half of that in today's dolars

 

[mathjak107]

what i found especially interesting is the fact putting the emotional issue a side retirement planning is really backwards.

allocations to stocks are cut and sometimes ala bill bernstein eliminated from retirement planning.

the way facts show us to plan is to increase allocations as we age.

we always hear about the fact if we lose money we can't replace it and how equities are risky as we age.

but the fact is no group of retirees ever failed even at 100% equities at a 3.50% swr as far as i know.

really so far that thinking about getting more and more conservative has been been much ado about nothing.

we don't know what the future will bring and i always think i will be the exception if i try it or we will be another japan .

but facts are fact and as wade and kitces say increasing equities by 1% a year may be the best way to go.

 

[Midpack]

How do you intend to use an SPIA when you're 80 if you don't mind me asking? I.e., what will determine the decision and how much? And what are your thoughts on a longevity annuity that would start at 80/85. The latter is what I was considering as my mother is 95 and kicking. Also, I like that it almost solves the question of how many years should I plan for I.e., up till age 80/85. The big problem with them is that you have to estimate how much you would need by guessing the future inflation rate as they're not cola'd.

I would never consider anything but a SPIA, no VA or deferred annuities for me (a separate debate).

When and if? I monitor my annuitization hurdle quarterly, here's a good summary of the strategy http://www.schulmerichandassoc.com/Modern_Portfolio_Decumulation.pdf. If markets are favorable in the next 20 years or so, we should never need a SPIA, but it is part of our "plan B."

And annuities are at an all time high WRT cost due to low interest rates/yields. I wouldn't buy one now even if I was 75-85 if I could avoid it. Costs (per income provided) can only go down in the future for practical purposes, though we all realize it may be years or decades before that happens. This aspect has also been debated repeatedly, so others may have a very different POV.