The current annuity factor

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
nedsaid
Posts: 14713
Joined: Fri Nov 23, 2012 12:33 pm

Re: The current annuity factor

Post by nedsaid »

bobcat2 wrote: Mon Jun 07, 2021 5:55 pm Hi Nedsaid,

I can see why you wouldn't think much of the article if you read the Indian version. :)

Try this link from Harvard instead.
https://scholar.harvard.edu/files/manki ... _rates.pdf

Best,
BobK
Thanks so much, when I get time I will read the article thoroughly. I also like your comments comparing U.S. to Japan, I think our economy will have a lot of similarities to Japan in the future the main difference is that our population will grow through immigration. One reason that I want to increase my International exposure over time, both with Stocks and Bonds. I think Vanguard's recommendation of International investments being 40% of equities and 30% of Fixed Income is very rational. I don't think we will experience a 30 bear market here as Japan has experienced but remember unlikely does not mean impossible.
A fool and his money are good for business.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

bobcat2 wrote: Mon Jun 07, 2021 10:35 pm I am not surprised that many of the researches on SWRs are allocating bonds in a way that doesn't make sense for a retiree's portfolio from the POV of financial economics. There is absolutely no reason that the first tranche of withdrawals each year can't be duration matched to provide safe income. The tranche becomes smaller when interest rates are lower.

Retirement income layers
SS
db nominal pension
life annuity nominal or escalating
duration matched bond income
longevity annuity
withdrawals from equities

We are not the first to point out that this “start by spending x percent” strategy has no basis in economic theory.
Milevsky and Huang - Spending Retirement on Planet Vulcan

BobK :wink:
I think most people understand that trying to get a fixed amount of money (in real terms) from a variable portfolio is risky. I mean, it just is. How risky depends on how much you try to pull though.

I've skimmed the paper a couple of times and still don't really see how the authors have helped much in solving this problem:

1) You don't know how long you will live.
2) You don't know what equity and fixed income returns will be going forward
3) You don't know what late life expenses might hit.

I mean, the suggestion to get/create steady income stream(s) is pretty mainstream I thought (SS, pension, maybe buy an SPIA, etc). But you don't want to over-annutize.

For example, the paper says this:
The optimal trajectory of financial capital also declines with age. Moreover, for retirees with pre-existing pension income, spending down wealth by some advanced age, and thereafter living exclu-sively on pension income, is rational. The WDT can be at age 90—or even 80 if the pension income is sufficiently large. Greater (longevity) risk aver-sion, which is associated with lower consumption, induces greater financial capital at all ages. Planning to deplete wealth by some advanced age is neither wrong nor irrational.
Anyone who has seen a person in late life burn through well into six figures of medical expenses in a year or less* knows that you don't want all your money to be coming from an income stream. You want some resources left over. You'd think a Vulcan would understand that.

An upper-middle income person would be totally irrational to deplete their wealth by an advanced age, even if they had a pretty good pension. Someone with lesser means might do this, relying upon medicaid or family resources for things like LTC if needed.

* LTC insurance may buffer this, but will not cover some situations.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

vineviz wrote: Tue Jun 08, 2021 6:27 am
TN_Boy wrote: Mon Jun 07, 2021 9:39 pm The typical two or three fund portfolio boglehead recommendation does not, I believe use LT bonds. Please correct me if I'm wrong.
You aren't wrong, but the "typical" bond fund recommendation almost certainly is wrong or at least suboptimal.

This is not an argument against holding LT bonds incidentally; just a statement that I have missed portfolio recommendations to buy and hold LT bonds until they mature as part of the typical portfolio. The typical recommendation is something like the total bond fund.

TN_Boy wrote: Mon Jun 07, 2021 9:39 pm My portfolio is not structured as a liability matching portfolio. It is handled more like the SWR studies are handled. Pull the needed amount and re balance (this is also what I would do with an RMD approach, or RMD + dividends, etc). There are other withdrawal methods; for example "Prime Harvesting" which pulls only from the bond sleeve (necessitating selling bonds in many cases) until the stock portion of the portfolio has grown a certain amount.
The rational approach to retirement spending will, generally, pull from all assets proportionately each time. But regardless, the withdrawal method should not drive the asset allocation nor the duration decision around the bonds.
TN_Boy wrote: Mon Jun 07, 2021 9:39 pm At current yields, you would suggest someone 10 years from retirement to load up on LT treasuries?
Clearly the answer is "yes", assuming that by "Treasuries" you include both Treasuries and TIPS and that by "load up" you mean "put most of their bond allocation in".
I know you have a thread advocating at least 20% in LT bonds. If my question is answered in that thread just point me back to it.

Are you saying that someone (for example) with a 70/30 portfolio 10 years from retirement should have all the fixed income in LT treasuries or LT TIPs? Split 50-50 between nominal and TIPs?

But I think you are saying the bonds are not really for liability matching -- you seem to agree that you pull and rebalance which may involve selling bonds prior to maturity.

Have any of the SWR studies used LT bonds only? The interest rate risk of having all your bonds with a long duration seems a bit iffy, especially if I might not hold them all to maturity.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

bobcat2 wrote: Mon Jun 07, 2021 6:46 pm
TN_Boy wrote: Mon Jun 07, 2021 6:17 pm But what if the economy is doing okay (growing) and inflation remains under control? But real interest rates are low. Have we seen this condition before?
Our economy looks a lot like the Japanese economy this century, particularly since 2007. The only significant difference is our population and labor force have been growing between 0.5% to 1% per year until 2020 and in Japan the population and labor force are not growing. In other words on a per capita or per worker basis, there is very little difference between Japanese and American economic growth rates. The Japanification of the US economy is a major research area of academic economists in recent years. We've had low inflation, low interest rates, and an economy that has trouble growing over 2% per year. This is Japan, but with population growing modestly.

Japanification
https://www.bloomberg.com/news/articles ... -quicktake

BTW this may seem surprising, but the unemployment rate in Japan in recent years has been lower than the US unemployment rate. The UR in Japan last year was about 2.3% and in 2019 it was also about 2.3%.

BobK
So I find the Japan/US comparison new and frankly surprising. Do you have any other articles with a similar theme? The Bloomberg article is behind a paywall for me. I'll also do some searching.

I always thought (was told?) the US economy was more flexible than Japan's economy: more women in the labor force, more immigration, better at letting failed enterprises go ahead and fail, more automation in the office environment, etc. Slower than desired growth has been a feature of many European economies in recent years as well, right?
qwerty3656
Posts: 3
Joined: Sat Aug 06, 2011 11:17 am

Re: The current annuity factor

Post by qwerty3656 »

How can living off dividends and interest be risky, but living off dividends and interest PLUS RMD's be a viable option?
User avatar
Topic Author
bobcat2
Posts: 5565
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: The current annuity factor

Post by bobcat2 »

TN_Boy wrote: Tue Jun 08, 2021 4:44 pmSo I find the Japan/US comparison new and frankly surprising. Do you have any other articles with a similar theme? The Bloomberg article is behind a paywall for me. I'll also do some searching.
Japanification and Secular Stagnation
Some of these links may also be behind paywalls but probably not all will be.

Why US economists are obsessed with 'Japanification' (just before the pandemic)
https://www.youtube.com/watch?v=zZMY1e7j134

Japanification & Secular Stagnation (just before the pandemic) Panel discussion at American Economic Assoc. Annual Meeting (rather wonky)
https://www.aeaweb.org/webcasts/2020/ja ... challenges

US economy grows at slowest annual pace since 2016 (just before the pandemic)
https://www.ft.com/content/3746c864-436 ... 7a29cd66fe

Japanification: investors fear malaise is spreading globally (August, 2019)
https://www.ft.com/content/314c626a-c77 ... 69401ba76f

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
Wrench
Posts: 301
Joined: Sun Apr 28, 2019 10:21 am

Re: The current annuity factor

Post by Wrench »

CheepSkate wrote: Tue Jun 08, 2021 10:48 am <SNIP>

I humbly suggest that long-term portfolio survival means adapting to changing circumstances and being sufficiently educated to handle the necessary instruments for those circumstances. The random person on the street won't successfully manage a portfolio, and the successful path for the current generation of retirees will become the generic advice for the next generation, as it always has.
On this we can agree - one must be open to new approaches and be willing/able to adapt to changing circumstances as much as one is able.

Relative to the use of options the way you describe, if it works for you, go for it. It's not for me, and I have been trading options for about 15 years in a wide range of different strategies. If a knowledgeable options trader like me is not comfortable with the approach, I would be very surprised if an average investor who likely doesn't even know what an option is, will be able to handle it. But, let's compare notes in 20 years and see which of us has been more successful with our different approaches! :D I would not be surprised (and would hope!) that we both have been equally successful.

Wrench
User avatar
vineviz
Posts: 10255
Joined: Tue May 15, 2018 1:55 pm

Re: The current annuity factor

Post by vineviz »

TN_Boy wrote: Tue Jun 08, 2021 4:31 pm Are you saying that someone (for example) with a 70/30 portfolio 10 years from retirement should have all the fixed income in LT treasuries or LT TIPs? Split 50-50 between nominal and TIPs?
Any combination of LT nominal Treasuries and LT TIPS would strike me as entirely defensible.
TN_Boy wrote: Tue Jun 08, 2021 4:31 pm But I think you are saying the bonds are not really for liability matching -- you seem to agree that you pull and rebalance which may involve selling bonds prior to maturity.
There are multiple ways to implement liability matching and duration matching is one of them. Rebalancing between stocks and bonds reduces the certainty of your income, but it preserves the interest rate risk reduction benefits of the duration match. I'm not dogmatic about whether rebalancing is required or not: there are many reasonable ways to implement things.
TN_Boy wrote: Tue Jun 08, 2021 4:31 pmHave any of the SWR studies used LT bonds only?
Not to my knowledge, but the historical data on long-term Treasuries is pretty iffy. 30-year bonds have not been consistently issued for instance, Treasury bonds used to be callable, Fed maintained a cap on long-term rates at just 2.5% for several decades, etc.

TN_Boy wrote: Tue Jun 08, 2021 4:31 pm The interest rate risk of having all your bonds with a long duration seems a bit iffy, especially if I might not hold them all to maturity.
Keeping the duration of the bonds matched to your actual investment horizon reduces, not increases, your interest rate risk. A 50-year old investor has a lot more interest rate risk with total bond market funds than with long-term Treasury funds.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

vineviz wrote: Tue Jun 08, 2021 8:03 pm
TN_Boy wrote: Tue Jun 08, 2021 4:31 pm Are you saying that someone (for example) with a 70/30 portfolio 10 years from retirement should have all the fixed income in LT treasuries or LT TIPs? Split 50-50 between nominal and TIPs?
Any combination of LT nominal Treasuries and LT TIPS would strike me as entirely defensible.
TN_Boy wrote: Tue Jun 08, 2021 4:31 pm But I think you are saying the bonds are not really for liability matching -- you seem to agree that you pull and rebalance which may involve selling bonds prior to maturity.
There are multiple ways to implement liability matching and duration matching is one of them. Rebalancing between stocks and bonds reduces the certainty of your income, but it preserves the interest rate risk reduction benefits of the duration match. I'm not dogmatic about whether rebalancing is required or not: there are many reasonable ways to implement things.
TN_Boy wrote: Tue Jun 08, 2021 4:31 pmHave any of the SWR studies used LT bonds only?
Not to my knowledge, but the historical data on long-term Treasuries is pretty iffy. 30-year bonds have not been consistently issued for instance, Treasury bonds used to be callable, Fed maintained a cap on long-term rates at just 2.5% for several decades, etc.

TN_Boy wrote: Tue Jun 08, 2021 4:31 pm The interest rate risk of having all your bonds with a long duration seems a bit iffy, especially if I might not hold them all to maturity.
Keeping the duration of the bonds matched to your actual investment horizon reduces, not increases, your interest rate risk. A 50-year old investor has a lot more interest rate risk with total bond market funds than with long-term Treasury funds.
I should note for others watching this thread that vineviz is going over his take on the merits of LT bonds during retirement in this thread; more details there:

viewtopic.php?p=6009355#p6009355

I understand your answer about why SWR studies haven't used LT bonds, but I confess I would feel more comfortable if there was at least historical SWR studies (or even monte carlo simulations) showing that LT bonds really did, across a variety of interest rate environments, perform better than shorter bonds.

Obviously many of us now are thinking ... well, if we are in a period of rising interest rates do I REALLY want my bonds to 20 years and up?
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

bobcat2 wrote: Tue Jun 08, 2021 7:18 pm
TN_Boy wrote: Tue Jun 08, 2021 4:44 pmSo I find the Japan/US comparison new and frankly surprising. Do you have any other articles with a similar theme? The Bloomberg article is behind a paywall for me. I'll also do some searching.
Japanification and Secular Stagnation
Some of these links may also be behind paywalls but probably not all will be.

Why US economists are obsessed with 'Japanification' (just before the pandemic)
https://www.youtube.com/watch?v=zZMY1e7j134

Japanification & Secular Stagnation (just before the pandemic) Panel discussion at American Economic Assoc. Annual Meeting (rather wonky)
https://www.aeaweb.org/webcasts/2020/ja ... challenges

US economy grows at slowest annual pace since 2016 (just before the pandemic)
https://www.ft.com/content/3746c864-436 ... 7a29cd66fe

Japanification: investors fear malaise is spreading globally (August, 2019)
https://www.ft.com/content/314c626a-c77 ... 69401ba76f

BobK
Bob,

Thanks; I can get to the videos but not the FT articles. I'll go over those this weekend.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

qwerty3656 wrote: Tue Jun 08, 2021 5:31 pm How can living off dividends and interest be risky, but living off dividends and interest PLUS RMD's be a viable option?
I did not think people viewed living off dividends and interest as particularly risky, at least not in terms of portfolio survival. But the withdrawal rate tends to be low and variable. The former is pretty annoying unless you have a whole lot of money. The latter is also annoying, again, unless you have a whole lot of money.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

One question I forgot to ask Bob, what do you think about MYGAs, given the low interest rate environment we are facing.
Leesbro63
Posts: 7140
Joined: Mon Nov 08, 2010 4:36 pm

Re: The current annuity factor

Post by Leesbro63 »

One post referenced the being in the top 3% in regards to wealth...where is that dollarwise?
User avatar
vineviz
Posts: 10255
Joined: Tue May 15, 2018 1:55 pm

Re: The current annuity factor

Post by vineviz »

TN_Boy wrote: Wed Jun 09, 2021 7:44 pm
qwerty3656 wrote: Tue Jun 08, 2021 5:31 pm How can living off dividends and interest be risky, but living off dividends and interest PLUS RMD's be a viable option?
I did not think people viewed living off dividends and interest as particularly risky, at least not in terms of portfolio survival. But the withdrawal rate tends to be low and variable. The former is pretty annoying unless you have a whole lot of money. The latter is also annoying, again, unless you have a whole lot of money.
Stock dividends definitely vary, especially in real terms, but interest income from bonds is only variable if you want it to be. People tend to perceive it to be variable because investors often try to use short- and intermediate-term bonds ( or worse, cash) to fund their withdrawals over long-term periods such as retirement.

When the duration of your bonds matches your investment horizon the variability in withdrawal rate would be greatly reduced. When the cash flows are also matched (eg with a non-rolling bond ladder) the variability is completely eliminated.

Of course annuities provide the same stability of income.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
vineviz
Posts: 10255
Joined: Tue May 15, 2018 1:55 pm

Re: The current annuity factor

Post by vineviz »

TN_Boy wrote: Wed Jun 09, 2021 7:18 pm I understand your answer about why SWR studies haven't used LT bonds, but I confess I would feel more comfortable if there was at least historical SWR studies (or even monte carlo simulations) showing that LT bonds really did, across a variety of interest rate environments, perform better than shorter bonds.

Obviously many of us now are thinking ... well, if we are in a period of rising interest rates do I REALLY want my bonds to 20 years and up?
I think a lot will depend on what you think "perform better" means.

Using long-term bonds certainly will not always produce a higher SWR even though they usually will due to their higher yields and stronger diversification effects.

I'd say the definitive way in which long-term bonds do "perform better" is that they reduce the uncertainty about the SWR. When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.

And since we really have no idea whether we are in a "period of rising interest rates" or not, it seems to me the rational approach is one that doesn't depend on knowing something we can't possibly know: the future.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

vineviz wrote: Thu Jun 10, 2021 6:04 am
TN_Boy wrote: Wed Jun 09, 2021 7:44 pm
qwerty3656 wrote: Tue Jun 08, 2021 5:31 pm How can living off dividends and interest be risky, but living off dividends and interest PLUS RMD's be a viable option?
I did not think people viewed living off dividends and interest as particularly risky, at least not in terms of portfolio survival. But the withdrawal rate tends to be low and variable. The former is pretty annoying unless you have a whole lot of money. The latter is also annoying, again, unless you have a whole lot of money.
Stock dividends definitely vary, especially in real terms, but interest income from bonds is only variable if you want it to be. People tend to perceive it to be variable because investors often try to use short- and intermediate-term bonds ( or worse, cash) to fund their withdrawals over long-term periods such as retirement.

When the duration of your bonds matches your investment horizon the variability in withdrawal rate would be greatly reduced. When the cash flows are also matched (eg with a non-rolling bond ladder) the variability is completely eliminated.

Of course annuities provide the same stability of income.
I can't speak for qwerty3656, but the usual context of "live off dividends and interest" typically comes here up in SWR threads, and rarely is anybody duration matching; they own stocks and bonds and are asking/recommending about living off whatever dividends and interests are thrown off.
User avatar
Topic Author
bobcat2
Posts: 5565
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: The current annuity factor

Post by bobcat2 »

Question:
I confess I would feel more comfortable if there was at least historical SWR studies (or even monte carlo simulations) showing that LT bonds really did, across a variety of interest rate environments, "perform better" than shorter bonds.


Answer:
vineviz wrote: Thu Jun 10, 2021 6:45 amI think a lot will depend on what you think "perform better" means.

Using long-term bonds certainly will not always produce a higher SWR even though they usually will due to their higher yields and stronger diversification effects.

I'd say the definitive way in which long-term bonds do "perform better" is that they reduce the uncertainty about the SWR. When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.

And since we really have no idea whether we are in a "period of rising interest rates" or not, it seems to me the rational approach is one that doesn't depend on knowing something we can't possibly know: the future.
:thumbsup :thumbsup


When bonds are duration matched you are using a combination of ST & LT bonds or bond funds that match the duration of your spending. If you are retired and planning on spending an equal amount each year, and thus an equal withdrawal from bonds each year (in real terms), the duration of the withdrawals is half the investment horizon. So at age 65 with a 30 year investment horizon the duration is 15. At age 71 with a 24 year horizon the duration is 12.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
Wrench
Posts: 301
Joined: Sun Apr 28, 2019 10:21 am

Re: The current annuity factor

Post by Wrench »

Leesbro63 wrote: Wed Jun 09, 2021 8:09 pm One post referenced the being in the top 3% in regards to wealth...where is that dollarwise?
It depends on age. Here's a site you can see for yourself:
https://www.nerdwallet.com/article/fina ... rth-by-age
garlandwhizzer
Posts: 3192
Joined: Fri Aug 06, 2010 3:42 pm

Re: The current annuity factor

Post by garlandwhizzer »

vineviz wrote:

When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.
This is true in nominal terms but not in real terms. Who knows what the discount between real and nominal returns from cumulative inflation will be in 20 years? Answer: no one. Unless you are heavily exposed to TIPS, you really can't count on guaranteed nominal income to meet your real income needs especially in the distant future. The same vulnerability to cumulative inflation is a problem for all nominal income producing assets including annuities especially over a long time frame. There's a reason why financial companies no longer offer inflation adjusted annuities. It's impossible to predict cumulative inflation over a long time frame. All we can come up with is guesses. It is important for investors who are preparing for decades of retirement not to confuse the illusion of safety (guaranteed income producing assets in nominal terms) with the real thing (TIPS). If you're heavily invested in LTT I believe it wise to add adequate inflation protection with TIPS (short and long term protection) and/or diversified global equity (long term protection only).

Garland Whizzer
User avatar
vineviz
Posts: 10255
Joined: Tue May 15, 2018 1:55 pm

Re: The current annuity factor

Post by vineviz »

garlandwhizzer wrote: Thu Jun 10, 2021 12:59 pm
vineviz wrote:

When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.
This is true in nominal terms but not in real terms.
It's true in real terms if you use bonds that pay in real terms. Like TIPS.

And it's true in nominal terms if you use bonds that pay in nominal terms.

In other words it's always true, you just need use the bonds that match your inflation sensitivity as well as your (real) interest rate sensitivity.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Topic Author
bobcat2
Posts: 5565
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: The current annuity factor

Post by bobcat2 »

Leesbro63 wrote: Wed Jun 09, 2021 8:09 pm One post referenced the being in the top 3% in regards to wealth...where is that dollarwise?
I posted that and it was just a ball park figure of about $5,000,000. It would seem that if at least $4 million of that is financial wealth then you should be able to make ends meet in retirement even when real interest rates are zero.

Here are estimates for 2020 from the Federal Reserve's most recent Survey of Consumer Finances (SCF).

The least wealthy household in the top 3% had net worth of $4.64 million. To get in the top 1% you would have had net worth of at least $11.1 million. Median net worth for all US households was $121,400. The bottom 10% of households all had negative net worth.

Things aren't as bad as they seem at retirement. None of the above household net worth numbers include the present value at retirement of Social Security, db pensions, or Medicare. For a typical two adult household at retirement having Social Security and Medicare at age 65 is worth about $1 million.

Link for net worth numbers from SCF.
https://dqydj.com/average-median-top-ne ... rcentiles/

Link for net worth of Social Security and Medicare
https://www.urban.org/sites/default/fil ... 2020_0.pdf

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

bobcat2 wrote: Thu Jun 10, 2021 8:54 am Question:
I confess I would feel more comfortable if there was at least historical SWR studies (or even monte carlo simulations) showing that LT bonds really did, across a variety of interest rate environments, "perform better" than shorter bonds.


Answer:
vineviz wrote: Thu Jun 10, 2021 6:45 amI think a lot will depend on what you think "perform better" means.

Using long-term bonds certainly will not always produce a higher SWR even though they usually will due to their higher yields and stronger diversification effects.

I'd say the definitive way in which long-term bonds do "perform better" is that they reduce the uncertainty about the SWR. When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.

And since we really have no idea whether we are in a "period of rising interest rates" or not, it seems to me the rational approach is one that doesn't depend on knowing something we can't possibly know: the future.
:thumbsup :thumbsup


When bonds are duration matched you are using a combination of ST & LT bonds or bond funds that match the duration of your spending. If you are retired and planning on spending an equal amount each year, and thus an equal withdrawal from bonds each year (in real terms), the duration of the withdrawals is half the investment horizon. So at age 65 with a 30 year investment horizon the duration is 15. At age 71 with a 24 year horizon the duration is 12.

BobK
But for better or worse (you might well say worse :-) I am not using bonds as liability matching instruments, and my portfolio withdrawal would not necessarily be a fixed amount from bonds. I.e. if the stock market crashes, all my withdrawals from the portfolio will be from the bond side. To let the stocks recover. Which would quite possibly require selling bonds. Thus the actual amount of time the bond is held is unknown.

This is the way all the SWR studies are done, and while I understand a fixed withdrawal* from a variable return portfolio has risks, that is how I will do things. Within that very common context/method, I still wish that we had studies showing how LT bonds fared against the more common choice of intermediate term bonds across a variety of market scenarios.

And it doesn't matter how often I'm told LT bonds are better, I'd still love to see studies done. It appears such studies are not easy. But I would really not be happy holding for example 50% of my portfolio in LT bonds and watching what a sharp jump in interest rates would do. Especially if I might be selling some of those bonds due to the stock market also crashing .....

As garlandwhizzer points, you only know how much money you'll have from a bond if you are using TIPs.

* people can use withdrawal methods other than a 4% + inflation adjustment; there are variable withdrawal methods that attempt to
lower the risk of pulling money from a bunch of stocks and bonds.
TN_Boy
Posts: 2251
Joined: Sat Jan 17, 2009 12:51 pm

Re: The current annuity factor

Post by TN_Boy »

bobcat2 wrote: Thu Jun 10, 2021 1:24 pm
Leesbro63 wrote: Wed Jun 09, 2021 8:09 pm One post referenced the being in the top 3% in regards to wealth...where is that dollarwise?
I posted that and it was just a ball park figure of about $5,000,000. It would seem that if at least $4 million of that is financial wealth then you should be able to make ends meet in retirement even when real interest rates are zero.

Here are estimates for 2020 from the Federal Reserve's most recent Survey of Consumer Finances (SCF).

The least wealthy household in the top 3% had net worth of $4.64 million. To get in the top 1% you would have had net worth of at least $11.1 million. Median net worth for all US households was $121,400. The bottom 10% of households all had negative net worth.

Things aren't as bad as they seem at retirement. None of the above household net worth numbers include the present value at retirement of Social Security, db pensions, or Medicare. For a typical two adult household at retirement having Social Security and Medicare at age 65 is worth about $1 million.

Link for net worth numbers from SCF.
https://dqydj.com/average-median-top-ne ... rcentiles/

Link for net worth of Social Security and Medicare
https://www.urban.org/sites/default/fil ... 2020_0.pdf

BobK
Good numbers. I think many of us are not grateful enough for how powerful SS is for most of the US population.
Wrench
Posts: 301
Joined: Sun Apr 28, 2019 10:21 am

Re: The current annuity factor

Post by Wrench »

garlandwhizzer wrote: Thu Jun 10, 2021 12:59 pm
vineviz wrote:

When the duration of your bond holdings is matched to your investment time horizon you are immunized against needing to CARE about whether interest rates rise or fall, because your return from those bonds is locked in no matter what happens.
This is true in nominal terms but not in real terms. Who knows what the discount between real and nominal returns from cumulative inflation will be in 20 years? Answer: no one. Unless you are heavily exposed to TIPS, you really can't count on guaranteed nominal income to meet your real income needs especially in the distant future. The same vulnerability to cumulative inflation is a problem for all nominal income producing assets including annuities especially over a long time frame. There's a reason why financial companies no longer offer inflation adjusted annuities. It's impossible to predict cumulative inflation over a long time frame. All we can come up with is guesses. It is important for investors who are preparing for decades of retirement not to confuse the illusion of safety (guaranteed income producing assets in nominal terms) with the real thing (TIPS). If you're heavily invested in LTT I believe it wise to add adequate inflation protection with TIPS (short and long term protection) and/or diversified global equity (long term protection only).

Garland Whizzer
This post by Bobcat2 is the best explanation for "why financial companies no longer offer inflation adjusted annuities". It is likely not for the reason you suggest, but essentially because: "Inflation-adjusted life annuities went away because real interest rates remained very low for a long period of time and inflation was also very low. Both of those realities make inflation-adjusted life annuities expensive and unattractive to consumers."
viewtopic.php?p=6000631#p6000631

This makes sense because there are plenty of investment vehicles available to insurance companies to cover the inflation risk, including inflation protected securities (e.g. TIPS) and long term inflation swaps. But, the cost of those for a low volume market made the payouts of inflation adjusted annuities unattractive to consumers used to low inflation environments.

Wrench
Post Reply