Is the bond tent effective at reducing SoRR?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Is the bond tent effective at reducing SoRR?

Post by YRT70 »

After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?

I'm asking this as a general question but it also applies directly to my personal situation. I retired in 2019 with about 50/50 asset allocation. I want my portfolio to last for probably 40 years. I'm wondering if going to 40/60 now could reduce SoRR.
terran
Posts: 1884
Joined: Sat Jan 10, 2015 10:50 pm

Re: Is the bond tent effective at reducing SoRR?

Post by terran »

Here's something to consider: https://earlyretirementnow.com/2017/09/ ... lidepaths/. It could be argued that if anything you have too much in bonds and that over time you should start gliding to higher equities to combat inflation.
Topic Author
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Is the bond tent effective at reducing SoRR?

Post by YRT70 »

terran wrote: Sun Jun 14, 2020 8:34 am Here's something to consider: https://earlyretirementnow.com/2017/09/ ... lidepaths/. It could be argued that if anything you have too much in bonds and that over time you should start gliding to higher equities to combat inflation.
Thank you. Yes that's one of the articles I read. Going by that article one shouldn't be going lower than 60/40 for 60 year retirements. But I probably won't need 60 years so I'm left wondering what would work best in my situation.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

Keep in mind that addressing SORR in isolation does not necessarily benefit an overall retirement income plan.

So you are asking the wrong question. The right question is, whatever the SOR, does a bond tent reduce the risk of retirement failure in a retirement plan. You can read Pfau and others and see. I am pretty sure the effect is very small. It is very difficult to play off overall return against variability in return and make headway in retirement withdrawal.

Also the result could be contingent on conditions. I don't doubt that some retirees are in a situation where for a short initial time their withdrawals are very high intentionally. It would not be a surprise that providing for that high spending with a less volatile bucket would make sense.

Also don't forget that sequence of returns risk is a very special kind of effect which in real life is confounded with many other things, in particular what average return, of which the sequence is a part, you are going to get.
AlohaJoe
Posts: 5671
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Is the bond tent effective at reducing SoRR?

Post by AlohaJoe »

YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?
No, it isn't effective.

Check out the link to EarlyRetirementNow from YRT70. And that's among the most positive takes on it.

The safe withdrawal rate goes from 3.25% to 3.44%, assuming you picked exactly the right kind of bond tent to use. If you picked the wrong one then it might have gone down to 3.05%. That's with the "fail-safe". If you accept 1% failures then it the SWR goes from 3.43% to 3.58%. Again, assuming you managed to pick the exactly right kind of bond tent. If you picked the wrong one it goes down to 3.21%.

So

a) Hope you pick exactly the right bond tent!
b) Even if you do, the absolute improvement is small, an extra $150 a month on a $1,000,000 portfolio is the best possible outcome. Not nothing but...hardly a panacea.
Last edited by AlohaJoe on Sun Jun 14, 2020 8:50 am, edited 1 time in total.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

YRT70 wrote: Sun Jun 14, 2020 8:42 am
terran wrote: Sun Jun 14, 2020 8:34 am Here's something to consider: https://earlyretirementnow.com/2017/09/ ... lidepaths/. It could be argued that if anything you have too much in bonds and that over time you should start gliding to higher equities to combat inflation.
Thank you. Yes that's one of the articles I read. Going by that article one shouldn't be going lower than 60/40 for 60 year retirements. But I probably won't need 60 years so I'm left wondering what would work best in my situation.
It is very difficult to prove that you can improve the prospects of doing better than 60/40, give or take some ten percents. It is also the case that it is difficult to prove that other choices must be worse. Trying to find a "best" plan for a long retirement is tilting at windmills. It is probably better to be comfortable that what you do is likely going to be good enough or at least not be certainly wrong.

Forty years of retirement will present you with so much uncertainty that any plan is highly provisional.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

AlohaJoe wrote: Sun Jun 14, 2020 8:48 am
YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?
No, it isn't effective.

Check out the link to EarlyRetirementNow from YRT70. And that's among the most positive takes on it.

The safe withdrawal rate goes from 3.25% to 3.44%, assuming you picked exactly the right kind of bond tent to use. If you picked the wrong one then it might have gone down to 3.05%. That's with the "fail-safe". If you accept 1% failures then it the SWR goes from 3.43% to 3.58%. Again, assuming you managed to pick the exactly right kind of bond tent. If you picked the wrong one it goes down to 3.21%.

So

a) Hope you pick exactly the right bond tent!
b) Even if you do, the absolute improvement is small, an extra $150 a month on a $1,000,000 portfolio is the best possible outcome. Not nothing but...hardly a panacea.
Thanks. I thought I was remembering something like that but did not want to try to quote numbers in my post above where the comment was simply that "it is hard to prove" that there is a benefit.
KlangFool
Posts: 19109
Joined: Sat Oct 11, 2008 12:35 pm

Re: Is the bond tent effective at reducing SoRR?

Post by KlangFool »

YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?

I'm asking this as a general question but it also applies directly to my personal situation. I retired in 2019 with about 50/50 asset allocation. I want my portfolio to last for probably 40 years. I'm wondering if going to 40/60 now could reduce SoRR.
YRT70,

we are missing one key information in answering your question. What is your portfolio size versus the annual expense? Aka, your withdrawal rate.

KlangFool
User avatar
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Is the bond tent effective at reducing SoRR?

Post by Ben Mathew »

YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?
The problem with minimizing sequence of returns risk is that it could come at the cost of minimizing consumption risk, which is what we care about. A simple way to minimize sequence of returns risk is to consume only bonds in the early part of retirement, and then consume stocks late. You've reduced sequence of return risk, but at the expense of increasing consumption risk in later years. Another example: you can eliminate stocks altogether and go to 100% bonds. You've eliminated sequence of return risk, but maybe gave up too much return--likely suboptimal for most people. The focus should remain on consumption risk, not sequence of return risk.

Any sensible withdrawal plan in retirement will have some sequence of return risk because you have to decide how much to consume today without knowing what your return tomorrow will be. If the return late in life is higher than expected, you would have consumed too little early on. There is no way around it.
YRT70 wrote: Sun Jun 14, 2020 8:28 am I'm asking this as a general question but it also applies directly to my personal situation. I retired in 2019 with about 50/50 asset allocation. I want my portfolio to last for probably 40 years. I'm wondering if going to 40/60 now could reduce SoRR.
Lifecycle investing (under CRRA utility assumptions) says to maintain a fixed allocation in retirement years (counting social security and pensions as bonds). This eliminates unnecessary sequence of return risk. Trying to reduce sequence of return risk further would be suboptimal.

In your situation, consider maintaining the 50/50 or 40/60 allocation throughout retirement. Whether you go with 50/50 or 40/60 will determine how risky your consumption will be. The right choice would depend on your risk aversion, not explicitly sequence of return considerations.
Last edited by Ben Mathew on Sun Jun 14, 2020 11:15 am, edited 1 time in total.
aristotelian
Posts: 8627
Joined: Wed Jan 11, 2017 8:05 pm

Re: Is the bond tent effective at reducing SoRR?

Post by aristotelian »

Bonds definitely reduce volatility. The problem in my mind with the "tent" is the rising equity allocation later in retirement.

If the market goes down early in retirement: I honestly would not have the risk tolerance to raise my equity allocation. "Overbalancing" the dip sounds good in principle but you would be taking tremendous risk at exactly the wrong time in your life. What happens if you overbalance after a 30% drop, then you increase your equity allocation and the market drops another 50%?

If the market goes up: I would actually get more conservative with less need to take risk.

In either case I would not increase my equity allocation. I also think a constant allocation behaviorally is going to be much easier to maintain without getting into market timing and behavioral risk.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

Ben Mathew wrote: Sun Jun 14, 2020 11:01 am

Lifecycle investing (under CRRA utility assumptions) says to maintain a fixed allocation in retirement years (counting social security and pensions as bonds). This eliminates unnecessary sequence of return risk. Trying to reduce sequence of return risk further would be suboptimal.

In your situation, consider maintaining the 50/50 or 40/60 allocation throughout retirement. Whether you go with 50/50 or 40/60 will determine how risky your consumption will be. The right choice would depend on your risk aversion, not explicitly sequence of return considerations.
Good explanation. It all comes back to the fact that targeting any particular risk as the gremlin du jour is not a very good way to do financial planning.

You only do that when for some reason a particular hazard is both likely and severe. Otherwise it is like stepping off a kerb in London while only looking left.
randomguy
Posts: 9208
Joined: Wed Sep 17, 2014 9:00 am

Re: Is the bond tent effective at reducing SoRR?

Post by randomguy »

aristotelian wrote: Sun Jun 14, 2020 11:13 am Bonds definitely reduce volatility. The problem in my mind with the "tent" is the rising equity allocation later in retirement.

If the market goes down early in retirement: I honestly would not have the risk tolerance to raise my equity allocation. "Overbalancing" the dip sounds good in principle but you would be taking tremendous risk at exactly the wrong time in your life. What happens if you overbalance after a 30% drop, then you increase your equity allocation and the market drops another 50%?

If the market goes up: I would actually get more conservative with less need to take risk.

In either case I would not increase my equity allocation. I also think a constant allocation behaviorally is going to be much easier to maintain without getting into market timing and behavioral risk.
With a constant allocation, how do you deal with the problem of high equity allocations early in retirement? If instead of holding 25/75 and climbing to 60/40, you have a constant 50/50 allocation you are going to be taking more risk on early in retirement at exactly the wrong time in your life.

And yes in real life, things can. happen to change your plan. If you retire in 1985 and at 15 years in you have made a killing, revisting your plan can make sense. Same thing for the 1966 person who is struggling.

60 year time frames like ERE uses probably doesn't apply to most of us and to some extent those eliminate some of the benefits of the tent (i.e. over 30 year time periods, a much larger chunk of your spending comes from savings than over a 60 where earnings dominate. Missing out on gains from those early years matter a lot more). But yes the benefits of the tent or rising equity glide path tend to be pretty minimal in every study I have seen. The big one is lower volatility early (in exchange for less money on average later) which is appealing to some people. After all not having money coming in is stressful.
aristotelian
Posts: 8627
Joined: Wed Jan 11, 2017 8:05 pm

Re: Is the bond tent effective at reducing SoRR?

Post by aristotelian »

randomguy wrote: Sun Jun 14, 2020 2:30 pm
aristotelian wrote: Sun Jun 14, 2020 11:13 am Bonds definitely reduce volatility. The problem in my mind with the "tent" is the rising equity allocation later in retirement.

If the market goes down early in retirement: I honestly would not have the risk tolerance to raise my equity allocation. "Overbalancing" the dip sounds good in principle but you would be taking tremendous risk at exactly the wrong time in your life. What happens if you overbalance after a 30% drop, then you increase your equity allocation and the market drops another 50%?

If the market goes up: I would actually get more conservative with less need to take risk.

In either case I would not increase my equity allocation. I also think a constant allocation behaviorally is going to be much easier to maintain without getting into market timing and behavioral risk.
With a constant allocation, how do you deal with the problem of high equity allocations early in retirement? If instead of holding 25/75 and climbing to 60/40, you have a constant 50/50 allocation you are going to be taking more risk on early in retirement at exactly the wrong time in your life.
A diversified portfolio and low enough withdrawal rate should weather the storm with a level allocation or bond tent. The question for me is what to do later, whether you keep it level or increase the equity allocation. I don't see the point of increasing the equity allocation with a good sequence of returns, and I will not have the risk tolerance to increase the equity allocation in the middle of a market crash.
heyyou
Posts: 3845
Joined: Tue Feb 20, 2007 4:58 pm

Re: Is the bond tent effective at reducing SoRR?

Post by heyyou »

Consider just watching your spending until your portfolio recovers, however long is necessary. As has been mentioned elsewhere, a stock crash is common in the early years of retirement because the stock run-up hit the magic number for a class of prospective retirees. This crash is not unique except for recent retirees, where it is the first one with no job income, such as yourself. The big picture is these stock crashes are built into the history that determined the size of the safe withdrawal rates.

Note that the worst years to retire were followed by a long period of high inflation with no real gains (starting 1966 to 1982), and that was using the SWR method that rose with the inflation. We know better now than to blindly keep boosting our spending due to the inflation, always ignoring our remaining portfolio balance relative to our current spending. To avoid SoRR, adapt your spending to your current real asset level, knowing you can spend and travel more when the stock market is higher in real terms. That is just living within your (future) means, during retirement.

Sample of one: Retired late in 2005, then stunned by the portfolio losses in 2008. We were so happy to just not have to go to our jobs, that being at home was special for us. We did day trips to many locations in our surrounding recreation area at low cost, since we were avoiding lodging expenses. For special meals, we would splurge on ingredients at the grocery store then eat at home, or go to the nice restaurant for lunch instead of dinner, then drink adult beverages afterwards, at home.
User avatar
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Is the bond tent effective at reducing SoRR?

Post by Ben Mathew »

randomguy wrote: Sun Jun 14, 2020 2:30 pm With a constant allocation, how do you deal with the problem of high equity allocations early in retirement? If instead of holding 25/75 and climbing to 60/40, you have a constant 50/50 allocation you are going to be taking more risk on early in retirement at exactly the wrong time in your life.
For the lifecycle strategy with CRRA utility, this is not a problem because the withdrawal is percentage based. i.e It says something like: At age 83, you should withdraw 6.3% of your portfolio. When age 83 rolls around, if your portfolio turns out to be $100,000, you withdraw $6,300. If your portfolio turns out to be $1,000,000, you withdraw $63,000. (VPW for example works like this.)

With CRRA utility + fixed allocation + percentage withdrawal, you get a neat result: Your portfolio at the start of retirement, say at age 65, can be separated mentally into different buckets, one for each year of consumption. i.e. You can put $30,000 into a bucket for age 66, $28,000 into a bucket for age 67, and so on. Whatever the buckets grow into, that's what you get to withdraw when you get to that age. Of course, you are free to move money from one bucket to another. But you won't want to, because all buckets are growing or shrinking by the same percentage each year (the portfolio return). Since you have CRRA utility, you won't want to move money from one bucket to another.

I find this bucket view helpful in understanding how the strategy works. It's easy to see what's going on once you've separated your portfolio into different buckets for different ages. Your bucket for age 83, for example, sits untouched from age 65 to age 83. No money comes in and no money leaves. It grows and shrinks with each year's returns. Each year it is rebalanced back to the same fixed allocation, along with all the other buckets. So the risk being applied to this bucket remains constant every year. That means the order of returns from age 65 to age 82 will not affect consumption at age 83. So there is no sequence of return risk over past returns.

Is there sequence of return risk overall? Yes. High return early is better because it positively affects more buckets. i.e. A high return at age 65 positively impacts consumption for ages 66 to 100. But a high return at age 99 positively impacts consumption for only age 100. So you would prefer high returns to come early and poor returns late. There is no reasonable way to get around this without breaking other things. But you do have a limited version of sequence of return independence--consumption at any given age is not affected by the sequence of returns of previous years.
randomguy
Posts: 9208
Joined: Wed Sep 17, 2014 9:00 am

Re: Is the bond tent effective at reducing SoRR?

Post by randomguy »

Ben Mathew wrote: Sun Jun 14, 2020 5:50 pm
randomguy wrote: Sun Jun 14, 2020 2:30 pm With a constant allocation, how do you deal with the problem of high equity allocations early in retirement? If instead of holding 25/75 and climbing to 60/40, you have a constant 50/50 allocation you are going to be taking more risk on early in retirement at exactly the wrong time in your life.
For the lifecycle strategy with CRRA utility, this is not a problem because the withdrawal is percentage based. i.e It says something like: At age 83, you should withdraw 6.3% of your portfolio. When age 83 rolls around, if your portfolio turns out to be $100,000, you withdraw $6,300. If your portfolio turns out to be $1,000,000, you withdraw $63,000. (VPW for example works like this.)

With CRRA utility + fixed allocation + percentage withdrawal, you get a neat result: Your portfolio at the start of retirement, say at age 65, can be separated mentally into different buckets, one for each year of consumption. i.e. You can put $30,000 into a bucket for age 66, $28,000 into a bucket for age 67, and so on. Whatever the buckets grow into, that's what you get to withdraw when you get to that age. Of course, you are free to move money from one bucket to another. But you won't want to, because all buckets are growing or shrinking by the same percentage each year (the portfolio return). Since you have CRRA utility, you won't want to move money from one bucket to another.

I find this bucket view helpful in understanding how the strategy works. It's easy to see what's going on once you've separated your portfolio into different buckets for different ages. Your bucket for age 83, for example, sits untouched from age 65 to age 83. No money comes in and no money leaves. It grows and shrinks with each year's returns. Each year it is rebalanced back to the same fixed allocation, along with all the other buckets. So the risk being applied to this bucket remains constant every year. That means the order of returns from age 65 to age 82 will not affect consumption at age 83. So there is no sequence of return risk over past returns.

Is there sequence of return risk overall? Yes. High return early is better because it positively affects more buckets. i.e. A high return at age 65 positively impacts consumption for ages 66 to 100. But a high return at age 99 positively impacts consumption for only age 100. So you would prefer high returns to come early and poor returns late. There is no reasonable way to get around this without breaking other things. But you do have a limited version of sequence of return independence--consumption at any given age is not affected by the sequence of returns of previous years.
So basically after all the preceding mental gymnastics SOR risk is still there and they way. you chose to deal with it is by varying income with portfolio value? Great but now you have to worry about sequence of income issues. Getting 65k when your 95 doesn't help you when your 65 with a 10k property tax bill and formula is only giving you 6.3k. And no pensions, SS, annuities, looking in your couch cushions and so on aren't a solution.
Topic Author
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Is the bond tent effective at reducing SoRR?

Post by YRT70 »

KlangFool wrote: Sun Jun 14, 2020 8:59 am
YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?

I'm asking this as a general question but it also applies directly to my personal situation. I retired in 2019 with about 50/50 asset allocation. I want my portfolio to last for probably 40 years. I'm wondering if going to 40/60 now could reduce SoRR.
YRT70,

we are missing one key information in answering your question. What is your portfolio size versus the annual expense? Aka, your withdrawal rate.

KlangFool
Portfolio size is about 1.2M. My current spending strategy, given current market turmoil, is ALAP: As little as possible. I'll probably end up below 3%. Maybe around 2.5%, I'm not sure.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

YRT70 wrote: Sun Jun 14, 2020 9:12 pm
Portfolio size is about 1.2M. My current spending strategy, given current market turmoil, is ALAP: As little as possible. I'll probably end up below 3%. Maybe around 2.5%, I'm not sure.
At withdrawal rates that low I would not be making a demon of sequence of returns risk. A greater risk is that your estimate of and control over spending is not what you hope it is. Probably an even bigger risk than that is unexpected events of a rare but possibly serious nature, whatever those might be.

PS The market is always in turmoil, periodically more rather than less. At the moment the actual turmoil is not very severe. Events may change, of course.
Topic Author
YRT70
Posts: 801
Joined: Sat Apr 27, 2019 8:51 am

Re: Is the bond tent effective at reducing SoRR?

Post by YRT70 »

AlohaJoe wrote: Sun Jun 14, 2020 8:48 am
YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?
No, it isn't effective.

Check out the link to EarlyRetirementNow from YRT70. And that's among the most positive takes on it.

The safe withdrawal rate goes from 3.25% to 3.44%, assuming you picked exactly the right kind of bond tent to use. If you picked the wrong one then it might have gone down to 3.05%. That's with the "fail-safe". If you accept 1% failures then it the SWR goes from 3.43% to 3.58%. Again, assuming you managed to pick the exactly right kind of bond tent. If you picked the wrong one it goes down to 3.21%.

So

a) Hope you pick exactly the right bond tent!
b) Even if you do, the absolute improvement is small, an extra $150 a month on a $1,000,000 portfolio is the best possible outcome. Not nothing but...hardly a panacea.
Thanks. That's one of the articles I read previously. Going by that article a riding equity glide path starting at 60% was effective for 60 year retirements. My retirement is probably shorter so I'm wondering if there could be a benefit to starting with equity lower.

And some of the quotes from ERN sound very positive indeed:
Some more details on the mechanics of the glidepath and why it’s so successful in smoothing out Sequence of Return Risk.
And then I've been reading Kitces and Pfau too who also seem to believe in a benefit but they're also ambiguous about it.
dbr
Posts: 34232
Joined: Sun Mar 04, 2007 9:50 am

Re: Is the bond tent effective at reducing SoRR?

Post by dbr »

YRT70 wrote: Sun Jun 14, 2020 9:25 pm

Thanks. That's one of the articles I read previously. Going by that article a riding equity glide path starting at 60% was effective for 60 year retirements. My retirement is probably shorter so I'm wondering if there could be a benefit to starting with equity lower.
When you think about questions like this you have to be more detailed and explicit about what sorts of benefits you have in mind. Some people are worrying unnecessarily about not having enough money in retirement. Some people are not going to have enough money in retirement and the problem can't be fixed except by planning on spending less. Messing with asset allocation does not have a strong effect on how much you can spend in retirement but it has a big effect on how much might or might not be left over when you die. Asset allocation also has a big effect on how large are the ups and downs you see in your portfolio even when they don't matter for the outcome. (That's right -- big ups and downs don't necessarily mean bad news for the retiree nor does a very steady portfolio ensure good news for the retiree.)
User avatar
Ben Mathew
Posts: 1409
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Is the bond tent effective at reducing SoRR?

Post by Ben Mathew »

randomguy wrote: Sun Jun 14, 2020 8:41 pm
Ben Mathew wrote: Sun Jun 14, 2020 5:50 pm
randomguy wrote: Sun Jun 14, 2020 2:30 pm With a constant allocation, how do you deal with the problem of high equity allocations early in retirement? If instead of holding 25/75 and climbing to 60/40, you have a constant 50/50 allocation you are going to be taking more risk on early in retirement at exactly the wrong time in your life.
For the lifecycle strategy with CRRA utility, this is not a problem because the withdrawal is percentage based. i.e It says something like: At age 83, you should withdraw 6.3% of your portfolio. When age 83 rolls around, if your portfolio turns out to be $100,000, you withdraw $6,300. If your portfolio turns out to be $1,000,000, you withdraw $63,000. (VPW for example works like this.)

With CRRA utility + fixed allocation + percentage withdrawal, you get a neat result: Your portfolio at the start of retirement, say at age 65, can be separated mentally into different buckets, one for each year of consumption. i.e. You can put $30,000 into a bucket for age 66, $28,000 into a bucket for age 67, and so on. Whatever the buckets grow into, that's what you get to withdraw when you get to that age. Of course, you are free to move money from one bucket to another. But you won't want to, because all buckets are growing or shrinking by the same percentage each year (the portfolio return). Since you have CRRA utility, you won't want to move money from one bucket to another.

I find this bucket view helpful in understanding how the strategy works. It's easy to see what's going on once you've separated your portfolio into different buckets for different ages. Your bucket for age 83, for example, sits untouched from age 65 to age 83. No money comes in and no money leaves. It grows and shrinks with each year's returns. Each year it is rebalanced back to the same fixed allocation, along with all the other buckets. So the risk being applied to this bucket remains constant every year. That means the order of returns from age 65 to age 82 will not affect consumption at age 83. So there is no sequence of return risk over past returns.

Is there sequence of return risk overall? Yes. High return early is better because it positively affects more buckets. i.e. A high return at age 65 positively impacts consumption for ages 66 to 100. But a high return at age 99 positively impacts consumption for only age 100. So you would prefer high returns to come early and poor returns late. There is no reasonable way to get around this without breaking other things. But you do have a limited version of sequence of return independence--consumption at any given age is not affected by the sequence of returns of previous years.
So basically after all the preceding mental gymnastics SOR risk is still there
Yes. The mental gymnastics is to show that SORR on consumption in any given year has been eliminated over past returns. What remains cannot be eliminated in a natural way, because that would involve making today's consumption dependent on tomorrow's return. The ideal solution with no SORR would be to let the portfolio run without withdrawals till age 100, and then decide what age 66 consumption should be. Then there would be no SORR. That's not possible. So we have to be content to eliminate SORR partially (over prior years, not over future years)
randomguy wrote: Sun Jun 14, 2020 8:41 pm and they way. you chose to deal with it is by varying income with portfolio value?
Not sure what you mean by this. Do you mean that the withdrawal strategy is a percentage of portfolio? If so, yes. This is not unusual, though.
randomguy wrote: Sun Jun 14, 2020 8:41 pm Great but now you have to worry about sequence of income issues.
The lifecycle strategy (fixed allocations and percentage withdrawals) comes directly from utility maximization of a person with CRRA utility. It doesn't try to minimize SORR. It just tries to maximize utility. What I was trying to show was that the optimal solution that falls out of that happens to eliminate unnecessary SORR. What's left is unavoidable SORR.
randomguy wrote: Sun Jun 14, 2020 8:41 pm Getting 65k when your 95 doesn't help you when your 65 with a 10k property tax bill and formula is only giving you 6.3k. And no pensions, SS, annuities, looking in your couch cushions and so on aren't a solution.
You can easily extend this model to include a liability matched portfolio to fund specific costs in retirement. That's basically extra buckets with 100% bond allocation towards certain non-negotiable goals (could be the property tax in your example). This is often done. You could also have other buckets with different AA to fund other goals that you would like to have a different sensitivity to wealth fluctuaions. E.g. You could say: if markets do really well, at age 70 I'll take the extended family on a big vacation. That bucket could be 100% stock. Think of it as your luxury bucket. Similarly you could have a charity bucket that is aggressively invested. The possibilities are endless. Different categories of buckets with different sensitivities. I think it's a great model that makes sense on a lot of levels and is flexible enough to work for a lot of different people in a lot of different situations. I think a lot more people should be using this model.
curmudgeon
Posts: 2132
Joined: Thu Jun 20, 2013 11:00 pm

Re: Is the bond tent effective at reducing SoRR?

Post by curmudgeon »

I personally don't really consider a bond tent as all that meaningful to someone with a "middle of the road" asset allocation like the OP. It's just not going to move the needle that much differently than just withdrawing from whichever is overweight (or rebalancing).

I DO think it can be useful for someone who chooses to use a very high equity allocation by default. The bond tent in that case serves as a bit of "SORR insurance" for the early years of retirement. That's how I'm treating it, anyway, though as you intertwine it with various other moving parts (ACA subsidy, SS deferral, Roth conversions, IRMAA) it's not completely straightforward.
aristotelian
Posts: 8627
Joined: Wed Jan 11, 2017 8:05 pm

Re: Is the bond tent effective at reducing SoRR?

Post by aristotelian »

Ben Mathew wrote: Sun Jun 14, 2020 9:59 pm You can easily extend this model to include a liability matched portfolio to fund specific costs in retirement. That's basically extra buckets with 100% bond allocation towards certain non-negotiable goals (could be the property tax in your example). This is often done. You could also have other buckets with different AA to fund other goals that you would like to have a different sensitivity to wealth fluctuaions. E.g. You could say: if markets do really well, at age 70 I'll take the extended family on a big vacation. That bucket could be 100% stock. Think of it as your luxury bucket. Similarly you could have a charity bucket that is aggressively invested. The possibilities are endless. Different categories of buckets with different sensitivities. I think it's a great model that makes sense on a lot of levels and is flexible enough to work for a lot of different people in a lot of different situations. I think a lot more people should be using this model.
One specific take on liability matching I have found allows increased withdrawal rate is to set aside a bucket of bonds to cover the equivalent of Social Security income before claiming. It amounts to an increasing equity glide path as you spend down the liability matching bucket (unless you count Social Security as part of your fixed income). It doesn't really move the needle if you retire way early, but could reduce your FI/retirement number by 20% if you retire at 55.


I did some back of the napkin math and found a potential way to get to your FIRE number a little faster. Say you are seeking $60K inflation adjusted spending. Following the 4% rule, you would need $1.5M. However, say you are in your late 40's/early 50's, expecting say $20K social security and reasonably confident that it will be there. You can create a liability-matching portfolio to provide that chunk of income just for the bridge years before SS kicks in. In this case, say your are 45, planning to claim at 65, that would require $400K. Set that aside in bonds or TIPS and plan to spend it down, $20K per year. Your remaining portfolio needs to cover $40K, which you can do with $1M. That means your total FI number is now $1.4M ($1M + $400K), which reduces your FI number by $100K.

The difference is even greater the closer your are to being able to claim SS. Say you are 55, the liability matching side of your portfolio only needs to be $200K, making your FI number only $1.2M.

I know some folks do not like to count Social Security, and I do not wish to rehash that debate. However, for those who do believe that it is reasonable to assume Social Security will be there, this seems to me to be a pretty reasonable way to reduce your FI number without taking too much additional risk.
User avatar
Third Son
Posts: 291
Joined: Tue Dec 03, 2019 7:05 pm
Location: North of Nowhere

Re: Is the bond tent effective at reducing SoRR?

Post by Third Son »

YRT70 wrote: Sun Jun 14, 2020 8:28 am After reading multiple threads and articles this is still not clear to me: Is the bond tent strategy effective at reducing sequence of returns risk?

I'm asking this as a general question but it also applies directly to my personal situation. I retired in 2019 with about 50/50 asset allocation. I want my portfolio to last for probably 40 years. I'm wondering if going to 40/60 now could reduce SoRR.
I think the current situation (Cov-19) is actually dulling any SoRR for this year. The fact that travel is restricted, shops and restaurants have limited occupancy and staying at home is probably recommended for any one at risk helps keep costs down. I am retired and am 40/60 with enough cash to last a few years. If you control spending and weather any market downturns, I believe SoR can be controlled effectively. That is, if you have enough saved to get through a downturn.

I do think that once the pandemic is behind us, I will increase my equity allocation 5-10% but am in no hurry to do so.
"A part of all you earn is yours to keep" | | -The Richest Man in Babylon
Post Reply