Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Random Walker
Posts: 3443
Joined: Fri Feb 23, 2007 8:21 pm

Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Random Walker » Tue Jan 08, 2019 3:19 pm

https://www.advisorperspectives.com/art ... 6956764657

In this article Larry goes into significant depth on retirement planning with Monte Carlo Simulation. MCS results are usually described as a % likelihood of success and an average time to failure in the cases that don’t succeed. He reviews work by Javier Estrada which potentially improves on this approach. After this discussion, Larry reviews Estrada’s analysis of the bucket approach. Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave

Amadis_of_Gaul
Posts: 16
Joined: Tue Dec 18, 2018 5:57 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Amadis_of_Gaul » Tue Jan 08, 2019 4:16 pm

Thanks for sharing! That bit about D-RAS is fascinating. I'm shocked that across 21 countries and 115 years, the average optimal portfolio was 85/15. That's AWFULLY aggressive for a retirement portfolio!

I've seen similar numbers from the spreadsheet on earlyretirementnow.com, but ERN only looks at US market performance. I've dismissed the recommendations from that calculator because of the possibility that in the 30 years of my retirement, something worse than the Great Depression might happen. However, Estrada's 21 countries have to include some that got clobbered in World War II, experienced currency devaluation, and so forth. Even against the backdrop of all that, going long on equities STILL seems to be the best course of action.

I can't read Estrada's papers at work, but I will certainly look into them when I get home.

Anybody else think this is crazy?

User avatar
Horton
Posts: 173
Joined: Mon Jan 21, 2008 3:53 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Horton » Tue Jan 08, 2019 4:29 pm

Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
After this discussion, Larry reviews Estrada’s analysis of the bucket approach. Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.
Thanks for sharing.

The bucket strategy discussed in the article is short-term and vastly different than the two bucket Liability Matching Portfolio / Risk Portfolio (LMP/RP) approach that Bill Bernstein discusses.

It would be interesting to see the analysis applied to an LMP/RP or a rising equity glidepath strategy, which Pfau notes are more or less equivalent and:
have the potential to actually reduce both the probability of failure and the magnitude of failure
for client portfolios.

Source

User avatar
Artsdoctor
Posts: 3689
Joined: Thu Jun 28, 2012 3:09 pm
Location: Los Angeles, CA

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Artsdoctor » Tue Jan 08, 2019 4:32 pm

I'm not too surprised that the bucket approach did worse, although I'm surprised by how often it "failed," according to the models. In real life, I'm sure that retirees would adapt as their portfolios dwindled down, and they would adjust their spending accordingly. This would probably constitute a "failure" although the retiree probably wouldn't see it that way.

Interesting data.

Cody
Posts: 900
Joined: Sun Dec 02, 2007 9:19 am
Location: Stillwater, Mn

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Cody » Tue Jan 08, 2019 5:00 pm

It would seem that one could use Lifestrategy 80 (80% equity and 20 bonds) accross all of ones retirement accounts to make this model work (D-RAS). But is the rebalancing of LS80 (done automatically for you frequently) good enough to capture the benifit of D-RAS?

Cody

Tdubs
Posts: 255
Joined: Tue Apr 24, 2018 7:50 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Tdubs » Tue Jan 08, 2019 6:23 pm

I was under the impression from somewhere that the bucket approach has value when you have a high withdrawal rate and bad markets early on. These scenarios were all run at 4 percent SWR, right?
Last edited by Tdubs on Tue Jan 08, 2019 7:35 pm, edited 1 time in total.

are_cynic
Posts: 43
Joined: Wed Jul 25, 2018 8:14 am

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by are_cynic » Tue Jan 08, 2019 7:26 pm

Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
"Invert, always invert" ~Carl Jacobi

bhsince87
Posts: 1961
Joined: Thu Oct 03, 2013 1:08 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by bhsince87 » Tue Jan 08, 2019 7:27 pm

are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
That's my take from this study.
Retirement: When you reach a point where you have enough. Or when you've had enough.

User avatar
WoodSpinner
Posts: 550
Joined: Mon Feb 27, 2017 1:15 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by WoodSpinner » Tue Jan 08, 2019 7:42 pm

There are some interesting nuggets in this paper that a make it worth the read—especially for additional metrics to use when evaluating Monte Carlo Simulations. My current tool for these simulations is the Flexible Retirement Planner (FRP) but I don’t think it can answer some of the questions highlighted in this paper.

I think the D-RAS approach has merit but would like to couple this with an expense reduction approach at various bands.

So some additional key measures might include:
1. Number of Years to first failure (failure being not able to fund even after reducing expense by a factor (e.g. 10%)
2. Expense reduction Mean and Standard Deviation to meet a specific success rate (e.g. 90% success with a 11% reduction +/- 5%).
3. Number of years requiring Expense Reduction.
4. Number of years of failure.

Does anyone have any suggestions for a better Monte Carlo Analysis tool?


I terms of the bucket strategy analysis, I think Estrada’s approach is significantly flawed and Larry just seems to be adding to the confusion. My main concerns involve the use of only 2 buckets and a small set of rules for how to refill. Simplistic to say the least!

Would love to see a more detailed analysis with a wider variety of approaches.

For instance I am using a modified bucket approach that involves:
1. Bucket-1, 2 years of rollingexpected Portfolio Withdrawels, Invested in CDs or Short Term Treasury fund
2. Bucket-2, 8 years of Portfolio Withdrawels in Intermediate Bond Fund
3. Bucket-3, Equities (75% Total Stock Market, 25% International)

**Note: Portfolio Withdrawels Represent Expenses - Income (annuities, pensions etc.).

Bucket 1&2 represent a floor of reserves that can shift but to no less than 40% of the portfolio. This allows some flexibility to buy equities on the downturn. It also provides some protection for funding expenses in the short term.

Bucket 3 represent a hedge against inflation and the expected long term growth engine for the portfolio. This can shift between 40-60% of the portfolio.

In my case, most of my the majority of expected Portfolio Withdrawals occur in the next 10 years and will significantly reduce when SS begins at 70.

In practice this provides me an Asset Allocation that is somewhat dynamic and can adjust based on my needs and market forces within a pre-defined band. Currently I am 59 and in the 2nd year of retirementat. At 54%/46% and am hoping to continue an slow upward glide path to 60/40.

Wonder what this would look like in Estrada’s testing?

WoodSpinner

Topic Author
Random Walker
Posts: 3443
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Random Walker » Tue Jan 08, 2019 7:48 pm

bhsince87 wrote:
Tue Jan 08, 2019 7:27 pm
are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
That's my take from this study.
I wouldn’t call it a rebalancing premium. If take from safe asset class to rebalance into risky asset class after the risky asset class has gone down, you are simply increasing the risk and expected return of the portfolio. In this case, the investor is rebalancing to increase the level of portfolio risk back to his desired baseline.

Dave

User avatar
willthrill81
Posts: 7220
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by willthrill81 » Tue Jan 08, 2019 8:18 pm

Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.
It may be more than just mental accounting. In a deep bear market for stocks, retirees may hesitate to rebalance their 'safe' bonds into the stocks that have recently plummeted in value for fear of 'rebalancing into oblivion'. In a prior thread, I compared the effect of rebalancing or not for year 2000 retirees using the '4% rule' for withdrawals and found that rebalancing resulted in a large improvement in portfolio performance as compared to not. But I'd say that it's safe to say that many retirees hesitated to rebalance their portfolios in late 2008.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

betablocker
Posts: 464
Joined: Mon Jan 11, 2016 1:26 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by betablocker » Tue Jan 08, 2019 8:20 pm

This is similar to Betterment’s critique of emergency funds though emergency funds are obviously not as large as a five year bucket. Good points though and again if buckets help people remain logical then they might indeed be worth it.

international001
Posts: 561
Joined: Thu Feb 15, 2018 7:31 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by international001 » Tue Jan 08, 2019 8:47 pm

PErhaps what is needed is a comparable bucket rule. i.e. one that rebalances the overall portfolio (all buckets at the same time)

AlohaJoe
Posts: 4043
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by AlohaJoe » Tue Jan 08, 2019 9:00 pm

Amadis_of_Gaul wrote:
Tue Jan 08, 2019 4:16 pm
However, Estrada's 21 countries have to include some that got clobbered in World War II, experienced currency devaluation, and so forth.
[...]
I can't read Estrada's papers at work, but I will certainly look into them when I get home.
It doesn't sound crazy to me. All of things you listed are not great for stocks, true. But they are absolutely devastating for bonds. If you look at true "worst case scenarios" in the 20th century, virtually of them have the best result with 100% stocks. But we should be clear about why, so we don't learn the wrong lesson.

The worst scenarios are usually the result of WW1 or WW2 and something like widespread governmental collapse/dysfunction. Belgium was invaded. France was occupied. Japan had atomic bombs dropped on it. Germany lots millions of its citizens and was split into two countries. Spain had a civil war. Portugal was ruled by a dictatorship. Etc.

Given the government problems, is it any surprise that government bonds didn't do well? Private industry -- stocks -- are at least kinda sorta different from government.

AlohaJoe
Posts: 4043
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by AlohaJoe » Tue Jan 08, 2019 9:11 pm

WoodSpinner wrote:
Tue Jan 08, 2019 7:42 pm
Would love to see a more detailed analysis with a wider variety of approaches.
I don't understand the rules for your bucket strategy. When do you refill Bucket 1? When do you refill Bucket 2? How do you insure that equities remains at 40-60% with the refilling rules you've chosen?

Amadis_of_Gaul
Posts: 16
Joined: Tue Dec 18, 2018 5:57 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Amadis_of_Gaul » Tue Jan 08, 2019 9:46 pm

AlohaJoe wrote:
Tue Jan 08, 2019 9:00 pm

It doesn't sound crazy to me. All of things you listed are not great for stocks, true. But they are absolutely devastating for bonds. If you look at true "worst case scenarios" in the 20th century, virtually of them have the best result with 100% stocks. But we should be clear about why, so we don't learn the wrong lesson.

The worst scenarios are usually the result of WW1 or WW2 and something like widespread governmental collapse/dysfunction. Belgium was invaded. France was occupied. Japan had atomic bombs dropped on it. Germany lots millions of its citizens and was split into two countries. Spain had a civil war. Portugal was ruled by a dictatorship. Etc.

Given the government problems, is it any surprise that government bonds didn't do well? Private industry -- stocks -- are at least kinda sorta different from government.
I read the papers, and you're right. If you were retiring and trying to live off of savings in Germany between 1900 and 2014, you were doomed not to make it 30 years about half the time. There are risks that no asset allocation can guard against.

However, stocks did better in those countries than bonds did, either because of sovereign collapse or because they made enough money to give you a better chance of squeaking through.

The data suggest, then, that bonds are better at protecting against shallow risk, but stocks are better at protecting against deep risk. Frankly, I'm not that concerned about shallow risk. Given enough money, enough time, and enough diversification, an investor can come through whatever a "normal" market might throw at him just fine. Deep risk, though, to the extent that you can address it at all, you have to address it with stocks.

Amirite?

User avatar
nedsaid
Posts: 10933
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by nedsaid » Tue Jan 08, 2019 9:54 pm

willthrill81 wrote:
Tue Jan 08, 2019 8:18 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.
It may be more than just mental accounting. In a deep bear market for stocks, retirees may hesitate to rebalance their 'safe' bonds into the stocks that have recently plummeted in value for fear of 'rebalancing into oblivion'. In a prior thread, I compared the effect of rebalancing or not for year 2000 retirees using the '4% rule' for withdrawals and found that rebalancing resulted in a large improvement in portfolio performance as compared to not. But I'd say that it's safe to say that many retirees hesitated to rebalance their portfolios in late 2008.
I was 49 years old at the time and I was hesitant to rebalance my portfolio from bonds to stocks. I was scared. So I did the next best thing and put 100% of my new monies for investment into stocks for a year.
A fool and his money are good for business.

User avatar
siamond
Posts: 4410
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by siamond » Tue Jan 08, 2019 10:51 pm

Amadis_of_Gaul wrote:
Tue Jan 08, 2019 9:46 pm
The data suggest, then, that bonds are better at protecting against shallow risk, but stocks are better at protecting against deep risk. Frankly, I'm not that concerned about shallow risk. Given enough money, enough time, and enough diversification, an investor can come through whatever a "normal" market might throw at him just fine. Deep risk, though, to the extent that you can address it at all, you have to address it with stocks.

Amirite?
Yes, indeed. When thinking of 'risk' for a retiree while going beyond the naive 'risk == volatility' mantra, you quickly end up analyzing effects over the entire retirement period, hence addressing deep(er) risks. Which are, imho, much more meaningful. Then yes, when running corresponding simulations (which I personally did at length), a 75/25 asset allocation during retirement suddenly start making a lot more sense than (flawed) common wisdom would dictate... And even more if you happen to be an early retiree like me.

This really should be combined with a variable withdrawal method though, not with the non-sensical constant withdrawal method that retirement researchers keep beating to death.

heyyou
Posts: 3254
Joined: Tue Feb 20, 2007 4:58 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by heyyou » Wed Jan 09, 2019 12:03 am

Is there a fixed percentage of portfolio method with bands to reduce the annual volatility?

User avatar
danielc
Posts: 533
Joined: Sun Dec 10, 2017 4:48 am
Location: Pennsylvania, USA
Contact:

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by danielc » Wed Jan 09, 2019 12:08 am

Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

In this article Larry goes into significant depth on retirement planning with Monte Carlo Simulation. MCS results are usually described as a % likelihood of success and an average time to failure in the cases that don’t succeed. He reviews work by Javier Estrada which potentially improves on this approach. After this discussion, Larry reviews Estrada’s analysis of the bucket approach. Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave

Is having an emergency fund not a type of "bucket approach". How about the liability-matching portfolio?

CurlyDave
Posts: 795
Joined: Thu Jul 28, 2016 11:37 am

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by CurlyDave » Wed Jan 09, 2019 3:48 am

siamond wrote:
Tue Jan 08, 2019 10:51 pm
...Yes, indeed. When thinking of 'risk' for a retiree while going beyond the naive 'risk == volatility' mantra, you quickly end up analyzing effects over the entire retirement period, hence addressing deep(er) risks. Which are, imho, much more meaningful. Then yes, when running corresponding simulations (which I personally did at length), a 75/25 asset allocation during retirement suddenly start making a lot more sense than (flawed) common wisdom would dictate... And even more if you happen to be an early retiree like me.

This really should be combined with a variable withdrawal method though, not with the non-sensical constant withdrawal method that retirement researchers keep beating to death.
+1

I have never been comfortable with conflating risk and volatility, even though this is the beginning of Nobel prizes in economics.

And, the concept of demanding a constant, inflation indexed income in retirement, when for the 45 years of prior adult life I made my outgo fit my income rather than the other way around, seems unrealistic.

To top it off, there are clearly non-economic risks in retirement which are seldom mentioned. Who has done a study explicitly considering my risk of dying before I run out of money? A 65 year old man in the US has a life expectancy of less than 20 years. A woman just slightly over 20. When my chances of dying greatly exceed my chances of running out of money I might be spending too much time thinking about the wrong risk...

User avatar
siamond
Posts: 4410
Joined: Mon May 28, 2012 5:50 am

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by siamond » Wed Jan 09, 2019 4:26 am

heyyou wrote:
Wed Jan 09, 2019 12:03 am
Is there a fixed percentage of portfolio method with bands to reduce the annual volatility?
You pretty much defined the core of the Guyton-Klinger method, an underappreciated and yet very reasonable approach to withdrawals.

Grt2bOutdoors
Posts: 19711
Joined: Thu Apr 05, 2007 8:20 pm
Location: New York

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Grt2bOutdoors » Wed Jan 09, 2019 7:20 am

CurlyDave wrote:
Wed Jan 09, 2019 3:48 am
siamond wrote:
Tue Jan 08, 2019 10:51 pm
...Yes, indeed. When thinking of 'risk' for a retiree while going beyond the naive 'risk == volatility' mantra, you quickly end up analyzing effects over the entire retirement period, hence addressing deep(er) risks. Which are, imho, much more meaningful. Then yes, when running corresponding simulations (which I personally did at length), a 75/25 asset allocation during retirement suddenly start making a lot more sense than (flawed) common wisdom would dictate... And even more if you happen to be an early retiree like me.

This really should be combined with a variable withdrawal method though, not with the non-sensical constant withdrawal method that retirement researchers keep beating to death.
+1

I have never been comfortable with conflating risk and volatility, even though this is the beginning of Nobel prizes in economics.

And, the concept of demanding a constant, inflation indexed income in retirement, when for the 45 years of prior adult life I made my outgo fit my income rather than the other way around, seems unrealistic.

To top it off, there are clearly non-economic risks in retirement which are seldom mentioned. Who has done a study explicitly considering my risk of dying before I run out of money? A 65 year old man in the US has a life expectancy of less than 20 years. A woman just slightly over 20. When my chances of dying greatly exceed my chances of running out of money I might be spending too much time thinking about the wrong risk...
+1. Am not yet retirement age, but I agree, my standard of living is based on my inflows.
"One should invest based on their need, ability and willingness to take risk - Larry Swedroe" Asking Portfolio Questions

AlohaJoe
Posts: 4043
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by AlohaJoe » Wed Jan 09, 2019 7:22 am

CurlyDave wrote:
Wed Jan 09, 2019 3:48 am
Who has done a study explicitly considering my risk of dying before I run out of money?
Lots and lots of people have done the study.

Here is one recent example called "Rich, Broke, Dead". Here's an simulation showing some retiring early (at age 55) but using 4% withdrawal (what many people would consider reckless for an early retiree). It has a 79% chance of success -- or 21% chance of failure. Yet the thin red line shows the actual chances of both being alive and broke. You actually never have more than a 2.3% chance of "real failure" (that is, being broke and also alive to experience being broke).

Image

Or let's take a scenario where someone retires normally but decides 4% withdrawals aren't enough and they want to withdraw 5%. That's only a 69% chance of success -- 31% chance of failure! Again, we see that in reality there is never more than a 6.3% chance of "real failure".

Image

(The simulation isn't perfect because it doesn't simulate couples with joint mortality which seems to me like by far the most common case...)

Previous discussion of Rich, Broke, Dead on Bogleheads: viewtopic.php?t=260050

Amadis_of_Gaul
Posts: 16
Joined: Tue Dec 18, 2018 5:57 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Amadis_of_Gaul » Wed Jan 09, 2019 7:44 am

CurlyDave wrote:
Wed Jan 09, 2019 3:48 am

+1

I have never been comfortable with conflating risk and volatility, even though this is the beginning of Nobel prizes in economics.

And, the concept of demanding a constant, inflation indexed income in retirement, when for the 45 years of prior adult life I made my outgo fit my income rather than the other way around, seems unrealistic.

To top it off, there are clearly non-economic risks in retirement which are seldom mentioned. Who has done a study explicitly considering my risk of dying before I run out of money? A 65 year old man in the US has a life expectancy of less than 20 years. A woman just slightly over 20. When my chances of dying greatly exceed my chances of running out of money I might be spending too much time thinking about the wrong risk...
The thing is that volatility doesn't really measure the risk that I care about. I am not overly concerned about whether the stock market goes down as long as it reverts to the mean, which, in the absence of deep risk, it will. I am concerned about whether I or my widow will run out of money/face straitened circumstances in retirement. That might happen because of a) poor planning (not enough money, bad asset allocation, etc.), or b) deep risk. The hammer might never drop, but how can I position myself in case it does?

I think the safe-withdrawal-rate literature is the equivalent of approximating a sphere in physics. Yes, this doesn't really happen, but it's a useful way of assessing cashflows in retirement. For my part, even though I intend to work as long as I can, I find it easier to think about retirement in terms of having enough savings than in terms of a desired retirement age. The latter is unknowable, but I can at least approximate the former using my current spending.

Which takes me to your third point. In retirement, (barring the supernatural), death is not a risk but a certainty. Frankly, I _hope_ to die before I run out of money! It's easy to "plan" so that the opposite happens. There are lots of folks who end up eating Alpo in their golden years without having planned very much at all.

Instead, the great personal risk that I see is having my earning years cut short by early incapacity or death, such that I or my widow might outlive the money I have saved. Though I don't want to retire at 60 or even earlier, I'm planning so that I can if I have to.

User avatar
willthrill81
Posts: 7220
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by willthrill81 » Wed Jan 09, 2019 11:17 am

nedsaid wrote:
Tue Jan 08, 2019 9:54 pm
willthrill81 wrote:
Tue Jan 08, 2019 8:18 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.
It may be more than just mental accounting. In a deep bear market for stocks, retirees may hesitate to rebalance their 'safe' bonds into the stocks that have recently plummeted in value for fear of 'rebalancing into oblivion'. In a prior thread, I compared the effect of rebalancing or not for year 2000 retirees using the '4% rule' for withdrawals and found that rebalancing resulted in a large improvement in portfolio performance as compared to not. But I'd say that it's safe to say that many retirees hesitated to rebalance their portfolios in late 2008.
I was 49 years old at the time and I was hesitant to rebalance my portfolio from bonds to stocks. I was scared. So I did the next best thing and put 100% of my new monies for investment into stocks for a year.
Thanks for sharing. I very much doubt that you were alone in your fear. That's why I believe that the bucket approach may be preferable to some. If it enables them to avoid panic selling, then it's probably better than a fixed AA.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Call_Me_Op
Posts: 7168
Joined: Mon Sep 07, 2009 2:57 pm
Location: Milky Way

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Call_Me_Op » Wed Jan 09, 2019 11:43 am

Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

In this article Larry goes into significant depth on retirement planning with Monte Carlo Simulation. MCS results are usually described as a % likelihood of success and an average time to failure in the cases that don’t succeed. He reviews work by Javier Estrada which potentially improves on this approach. After this discussion, Larry reviews Estrada’s analysis of the bucket approach. Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
I do not think that the bucket approach prohibits rebalancing into stocks. This is the assumption that makes the strategy tend to under-perform, but the assumption is not necessarily correct. In other words, if you define the bucket approach to fail - it will.

A much better approach is to allow rebalancing into the long-term bucket unless doing so drops the SAFE assets below a pre-defined floor.

As I have maintained, there is really no fundamental difference between bucketing and the standard total return approach.
Best regards, -Op | | "In the middle of difficulty lies opportunity." Einstein

User avatar
WoodSpinner
Posts: 550
Joined: Mon Feb 27, 2017 1:15 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by WoodSpinner » Wed Jan 09, 2019 12:15 pm

AlohaJoe wrote:
Tue Jan 08, 2019 9:11 pm
WoodSpinner wrote:
Tue Jan 08, 2019 7:42 pm
Would love to see a more detailed analysis with a wider variety of approaches.
I don't understand the rules for your bucket strategy. When do you refill Bucket 1? When do you refill Bucket 2? How do you insure that equities remains at 40-60% with the refilling rules you've chosen?

Buckets are refilled via a yearly rebalancing excercise and more often as needed using a +/-20% guardrail.

For me the key is that my AA can shift within a band of 40/60 to 60/40. I attempt to keep the rolling 10 years of Portfolio Withdrawals invested in low risk Short Term/Intermediate Term bonds. Current Years expenses can come out of any of the buckets depending on the current valuations.

During Bear Markets the preference will be to spend down the fixed income side and use any excess capacity of the Fixed Income allocation band to buy equities.

During Bull Markets, the preference will be to spend down from the Equity side and use any excess capacity to buy Fixed Income.

I am definitely new at all of this and realize that some of this may be a behavioral crutch—on the other hand it seems like a plan that I can work during good and bad times with some concrete metrics that can drive decisions for change.

Back to the referenced article — my point is there are many variations on the bucket approach and the authors analysis utilizes a very simplistic implementation method and uses that to disparage the entire methodology.

WoodSpinner

User avatar
Abe
Posts: 1830
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Abe » Wed Jan 09, 2019 12:24 pm

Call_Me_Op wrote:
Wed Jan 09, 2019 11:43 am
I do not think that the bucket approach prohibits rebalancing into stocks. This is the assumption that makes the strategy tend to under-perform, but the assumption is not necessarily correct. In other words, if you define the bucket approach to fail - it will.

A much better approach is to allow rebalancing into the long-term bucket unless doing so drops the SAFE assets below a pre-defined floor.

As I have maintained, there is really no fundamental difference between bucketing and the standard total return approach.
I have been reading the post regarding the bucket strategy, and I pretty much concluded the same thing. I really don't see much difference between bucketing and total return approach.
Slow and steady wins the race.

kolea
Posts: 1320
Joined: Fri Jul 11, 2014 5:30 pm
Location: Maui and Columbia River Gorge

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by kolea » Wed Jan 09, 2019 1:36 pm

are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
That is one way of looking at what happened in the Estrada study that Larry refers to. But Estrada had a particular strategy he modeled for how his buckets were managed, so it is hard to say whether the difference was solely due to not rebalancing, or was also due to the way the buckets were filled and drained. In other words there were multiple things going on. But it certainly is true that the Estrada bucket of bonds was allowed to grow much larger than the equities bucket.
Kolea (pron. ko-lay-uh). Golden plover.

jwhitaker
Posts: 46
Joined: Thu Jun 16, 2016 12:57 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by jwhitaker » Wed Jan 09, 2019 2:09 pm

I worry about survivorship bias in the selection of countries used in the data. Yes war affected Japan and many of the European countries but they all came back to stability and growth within the period. I know this flaw is present in most of these analyses, and it is better than using US only which I am also guilty of. But of course your historical analysis will say you should be 85%+ in stocks using countries whose economies you know went up. I'd love to see something that throws in some more Asian, Middle Eastern and African countries. I will give them that Japan is not at its historic peak recently at least. I guess if I live in a country that goes to complete heck, I have bigger problems than my bond allocation anyway.

aristotelian
Posts: 5083
Joined: Wed Jan 11, 2017 8:05 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by aristotelian » Wed Jan 09, 2019 2:12 pm

Call_Me_Op wrote:
Wed Jan 09, 2019 11:43 am
I do not think that the bucket approach prohibits rebalancing into stocks. This is the assumption that makes the strategy tend to under-perform, but the assumption is not necessarily correct. In other words, if you define the bucket approach to fail - it will.

A much better approach is to allow rebalancing into the long-term bucket unless doing so drops the SAFE assets below a pre-defined floor.

As I have maintained, there is really no fundamental difference between bucketing and the standard total return approach.
Agreed. Yes, if you definite it narrowly as keeping a large portion of the portfolio in cash earning less than inflation, it's pretty obvious that it is going to underperform. With a high withdrawal rate, you are increasing the chance of failure. However, it can work with a low withdrawal rate, or some simple steps, such as a CD ladder, to get some return on the "buckets".

bhsince87
Posts: 1961
Joined: Thu Oct 03, 2013 1:08 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by bhsince87 » Wed Jan 09, 2019 2:26 pm

aristotelian wrote:
Wed Jan 09, 2019 2:12 pm
Call_Me_Op wrote:
Wed Jan 09, 2019 11:43 am
I do not think that the bucket approach prohibits rebalancing into stocks. This is the assumption that makes the strategy tend to under-perform, but the assumption is not necessarily correct. In other words, if you define the bucket approach to fail - it will.

A much better approach is to allow rebalancing into the long-term bucket unless doing so drops the SAFE assets below a pre-defined floor.

As I have maintained, there is really no fundamental difference between bucketing and the standard total return approach.
Agreed. Yes, if you definite it narrowly as keeping a large portion of the portfolio in cash earning less than inflation, it's pretty obvious that it is going to underperform. With a high withdrawal rate, you are increasing the chance of failure. However, it can work with a low withdrawal rate, or some simple steps, such as a CD ladder, to get some return on the "buckets".
And yet another perspective is, who cares if the approach under-performs?

At that point in life, preserving capital and keeping income/returns at least matched to inflation is a much more likely goal than maximizing returns.
Retirement: When you reach a point where you have enough. Or when you've had enough.

User avatar
Abe
Posts: 1830
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Abe » Wed Jan 09, 2019 3:28 pm

bhsince87 wrote:
Wed Jan 09, 2019 2:26 pm

And yet another perspective is, who cares if the approach under-performs?

At that point in life, preserving capital and keeping income/returns at least matched to inflation is a much more likely goal than maximizing returns.
Amen to that.
Slow and steady wins the race.

User avatar
BlueEars
Posts: 3642
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by BlueEars » Wed Jan 09, 2019 4:03 pm

Apparently in that article the Treasury bills bucket was very deep i.e. 5 years spending at 4% implies 20% of the portfolio in Treasury bills. Then it's not surprising that this would lag a "static" 60/40 rebalanced portfolio.

FWIW, I hold a shallow bucket currently with 7% of portfolio in short term investment grade (ST IG bonds). This is constantly replenished from rising equities or from the intermediate term bonds. As needed I sell ST IG to move to our checking account. I never thought of this as a bucket strategy though. The ST IG bonds are part of our AA which is generally held at 60/40 i.e. the ST IG bonds are a part of the fixed income as opposed to a separate account (or bucket). From VPW simulations I determined the size of the ST IG bond holdings to supplement spending when we have a series of down years such as the late 1960's into the 1970's.

international001
Posts: 561
Joined: Thu Feb 15, 2018 7:31 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by international001 » Wed Jan 09, 2019 5:51 pm

I tend to agree. Given that the asset allocation is the same with buckets, what buckets really allow you is to smooth out your withdrawal rate, so it depends less (on the short term) of markets ups and downs; but if markets go up a lot or down a lot on the long term, your withdrawals move in the same direction.

gtwhitegold
Posts: 293
Joined: Fri Sep 21, 2012 1:55 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by gtwhitegold » Wed Jan 09, 2019 6:03 pm

My question is tangentially related to this topic. What is the safe allocation to cash for some approaching or in retirement? I was thinking 2 years, but it realistically could be more or less than that.

User avatar
willthrill81
Posts: 7220
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by willthrill81 » Wed Jan 09, 2019 6:06 pm

gtwhitegold wrote:
Wed Jan 09, 2019 6:03 pm
My question is tangentially related to this topic. What is the safe allocation to cash for some approaching or in retirement? I was thinking 2 years, but it realistically could be more or less than that.
It's totally dependent on your preferences. If by 'cash' you mean a MM fund like VMMXX currently paying 1.95%, then you could easily allocate your entire portfolio to just stocks and cash. But many retirees don't have a specific cash allocation at all, simply using a mixture of stocks and bonds, which has historically been more 'efficient' than using just stocks and cash.

For instance, from 1972-2018, a portfolio with 50% U.S. total stock market and 50% intermediate-term Treasuries had an annual return of 8.95%, a standard deviation (measure of volatility) of 8.39%, and a maximum drawdown of -22%. In order to achieve that same return with just stocks and cash (in this case, 3 month Treasury bills), you need a 75% allocation to stocks and 25% to cash, but the standard deviation was was significantly higher at 11.48%, and the maximum drawdown was -39%. At least during this 47 year period, intermediate-term Treasuries were much better to hold than cash.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

MathWizard
Posts: 3175
Joined: Tue Jul 26, 2011 1:35 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by MathWizard » Wed Jan 09, 2019 6:24 pm

gtwhitegold wrote:
Wed Jan 09, 2019 6:03 pm
My question is tangentially related to this topic. What is the safe allocation to cash for some approaching or in retirement? I was thinking 2 years, but it realistically could be more or less than that.
You have to live on something, and most people don't want to be redeeming stocks/bonds on a monthly basis.

In the old days, stock transactions were expensive, so dividend funds filled the gap.

Now, with index funds, his is not a problem, so one can use a total return approach, and use a bucket approach,
fill buckets at the beginning (or end of a year) then you only need to do this once a year.

Before age 70, I will be doing something similar to get some ROTH conversions in before RMDs kick in and I take SS.
After that, I plan to use a SPIA and SS for a spending floor, and to keep enough cash for liquidity, much like I use my
first tier EF now.

So I will only need about 3 to 6 months expenses in cash depending on any upcoming cash needs, not 2 years worth.

User avatar
Sandtrap
Posts: 5668
Joined: Sat Nov 26, 2016 6:32 pm
Location: Hawaii😀 Northern AZ.😳 Retired.

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Sandtrap » Wed Jan 09, 2019 6:52 pm

WoodSpinner wrote:
Tue Jan 08, 2019 7:42 pm
There are some interesting nuggets in this paper that a make it worth the read—especially for additional metrics to use when evaluating Monte Carlo Simulations. My current tool for these simulations is the Flexible Retirement Planner (FRP) but I don’t think it can answer some of the questions highlighted in this paper.

I think the D-RAS approach has merit but would like to couple this with an expense reduction approach at various bands.

So some additional key measures might include:
1. Number of Years to first failure (failure being not able to fund even after reducing expense by a factor (e.g. 10%)
2. Expense reduction Mean and Standard Deviation to meet a specific success rate (e.g. 90% success with a 11% reduction +/- 5%).
3. Number of years requiring Expense Reduction.
4. Number of years of failure.

Does anyone have any suggestions for a better Monte Carlo Analysis tool?


I terms of the bucket strategy analysis, I think Estrada’s approach is significantly flawed and Larry just seems to be adding to the confusion. My main concerns involve the use of only 2 buckets and a small set of rules for how to refill. Simplistic to say the least!

Would love to see a more detailed analysis with a wider variety of approaches.

For instance I am using a modified bucket approach that involves:
1. Bucket-1, 2 years of rollingexpected Portfolio Withdrawels, Invested in CDs or Short Term Treasury fund
2. Bucket-2, 8 years of Portfolio Withdrawels in Intermediate Bond Fund
3. Bucket-3, Equities (75% Total Stock Market, 25% International)

**Note: Portfolio Withdrawels Represent Expenses - Income (annuities, pensions etc.).

Bucket 1&2 represent a floor of reserves that can shift but to no less than 40% of the portfolio. This allows some flexibility to buy equities on the downturn. It also provides some protection for funding expenses in the short term.

Bucket 3 represent a hedge against inflation and the expected long term growth engine for the portfolio. This can shift between 40-60% of the portfolio.

In my case, most of my the majority of expected Portfolio Withdrawals occur in the next 10 years and will significantly reduce when SS begins at 70.

In practice this provides me an Asset Allocation that is somewhat dynamic and can adjust based on my needs and market forces within a pre-defined band. Currently I am 59 and in the 2nd year of retirementat. At 54%/46% and am hoping to continue an slow upward glide path to 60/40.

Wonder what this would look like in Estrada’s testing?

WoodSpinner
Outstanding!
Very well presented.
Mahalo,
j :happy

MnD
Posts: 3883
Joined: Mon Jan 14, 2008 12:41 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by MnD » Wed Jan 09, 2019 6:54 pm

siamond wrote:
Tue Jan 08, 2019 10:51 pm
Amadis_of_Gaul wrote:
Tue Jan 08, 2019 9:46 pm
The data suggest, then, that bonds are better at protecting against shallow risk, but stocks are better at protecting against deep risk. Frankly, I'm not that concerned about shallow risk. Given enough money, enough time, and enough diversification, an investor can come through whatever a "normal" market might throw at him just fine. Deep risk, though, to the extent that you can address it at all, you have to address it with stocks.

Amirite?
Yes, indeed. When thinking of 'risk' for a retiree while going beyond the naive 'risk == volatility' mantra, you quickly end up analyzing effects over the entire retirement period, hence addressing deep(er) risks. Which are, imho, much more meaningful. Then yes, when running corresponding simulations (which I personally did at length), a 75/25 asset allocation during retirement suddenly start making a lot more sense than (flawed) common wisdom would dictate... And even more if you happen to be an early retiree like me.

This really should be combined with a variable withdrawal method though, not with the non-sensical constant withdrawal method that retirement researchers keep beating to death.
1
Entered retirement days before the December dump - 70/30, no buckets or other mental accounting games, proportional withdrawals, rebalance if and when rebalancing bands are exceeded. So basically no change from when accumulating. SWR is 5% of annual portfolio with a 3% inflation-adjusted floor, so if and only if sequences are terrible, I'll have a ultra-conservative SWR. Using December 31 balance for annual portfolio balance based SWR so the recent market kerfuffle has already auto-adjusted things just 30 days in! :beer

I can't decide if it's shallow risk that the ultra-low % and low equity SWR advocates are afraid of _or_ if it just fear spending anything more than the blend of the stock dividends and bond interest rates. It does seem uncanny that the prevailing ultra-low SWR's I read here seem to track a blend of the yield on TBM and TSM.
Last edited by MnD on Wed Jan 09, 2019 7:13 pm, edited 1 time in total.

gtwhitegold
Posts: 293
Joined: Fri Sep 21, 2012 1:55 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by gtwhitegold » Wed Jan 09, 2019 6:56 pm

MathWizard wrote:
Wed Jan 09, 2019 6:24 pm
gtwhitegold wrote:
Wed Jan 09, 2019 6:03 pm
My question is tangentially related to this topic. What is the safe allocation to cash for some approaching or in retirement? I was thinking 2 years, but it realistically could be more or less than that.
You have to live on something, and most people don't want to be redeeming stocks/bonds on a monthly basis.

In the old days, stock transactions were expensive, so dividend funds filled the gap.

Now, with index funds, his is not a problem, so one can use a total return approach, and use a bucket approach,
fill buckets at the beginning (or end of a year) then you only need to do this once a year.

Before age 70, I will be doing something similar to get some ROTH conversions in before RMDs kick in and I take SS.
After that, I plan to use a SPIA and SS for a spending floor, and to keep enough cash for liquidity, much like I use my
first tier EF now.

So I will only need about 3 to 6 months expenses in cash depending on any upcoming cash needs, not 2 years worth.
I'm planning on doing Roth conversions after the Military if I can get an overseas contracting job. I'm also planning on early retirement, so I'll still try to save a good deal of money before full retirement. I'm also planning on using the TSP G Fund as a cash fund.

Starfish
Posts: 602
Joined: Wed Aug 15, 2018 6:33 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Starfish » Wed Jan 09, 2019 6:58 pm

are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
Of course rebalancing generates a return, but is they are ashamed of it because is market timing and active management. It's against the religion.

User avatar
willthrill81
Posts: 7220
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by willthrill81 » Wed Jan 09, 2019 7:07 pm

Starfish wrote:
Wed Jan 09, 2019 6:58 pm
are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
Of course rebalancing generates a return, but is they are ashamed of it because is market timing and active management. It's against the religion.
Not necessarily. From 1972-2018, a 50/50 portfolio comprised of TSM and ITT had a .03% lower CAGR if it was rebalanced annually vs. not rebalanced. But rebalancing reduced the standard deviation from 9.92% to 8.39%.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Starfish
Posts: 602
Joined: Wed Aug 15, 2018 6:33 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Starfish » Wed Jan 09, 2019 7:15 pm

willthrill81 wrote:
Wed Jan 09, 2019 7:07 pm
Starfish wrote:
Wed Jan 09, 2019 6:58 pm
are_cynic wrote:
Tue Jan 08, 2019 7:26 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
https://www.advisorperspectives.com/art ... 6956764657

...Compared to static asset allocation strategies, the bucket approach underperforms mainly because the investor does not rebalance from fixed income into equities after equities have done poorly. The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.

Dave
Most (all?) threads on rebalancing seem to conclude that rebalancing is for maintaining a desired risk level, not generating return. But now the bucket strategy underperforms because of not rebalancing? So there is a rebalancing premium?
Of course rebalancing generates a return, but is they are ashamed of it because is market timing and active management. It's against the religion.
Not necessarily. From 1972-2018, a 50/50 portfolio comprised of TSM and ITT had a .03% lower CAGR if it was rebalanced annually vs. not rebalanced. But rebalancing reduced the standard deviation from 9.92% to 8.39%.

Good point, is more like improved Sharpe ratio. Which is in a way similar.
Depending on the statics re balancing can create directly greater return, but not always.

AlohaJoe
Posts: 4043
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by AlohaJoe » Wed Jan 09, 2019 10:43 pm

WoodSpinner wrote:
Wed Jan 09, 2019 12:15 pm
AlohaJoe wrote:
Tue Jan 08, 2019 9:11 pm
WoodSpinner wrote:
Tue Jan 08, 2019 7:42 pm
Would love to see a more detailed analysis with a wider variety of approaches.
I don't understand the rules for your bucket strategy. When do you refill Bucket 1? When do you refill Bucket 2? How do you insure that equities remains at 40-60% with the refilling rules you've chosen?

Buckets are refilled via a yearly rebalancing excercise and more often as needed using a +/-20% guardrail.

For me the key is that my AA can shift within a band of 40/60 to 60/40. I attempt to keep the rolling 10 years of Portfolio Withdrawals invested in low risk Short Term/Intermediate Term bonds. Current Years expenses can come out of any of the buckets depending on the current valuations.

During Bear Markets the preference will be to spend down the fixed income side and use any excess capacity of the Fixed Income allocation band to buy equities.

During Bull Markets, the preference will be to spend down from the Equity side and use any excess capacity to buy Fixed Income.
Here are the results of the WoodSpinner strategy:

Image

It does worse on nearly every single metric: worse income in nearly every scenario, from 1st percentile (worst case) to 90th percentile (best case), lowest minimum income, lowest mean & median income.

Looking at specific stressful retirements:

It didn't help in the Great Depression

Image

It left you worse off in 1966 & 1969

Image
Image

And during "normal" retirements, like say 1975, you are substantially worse off

Image

Do you have any data or analysis showing what scenarios the strategy is useful in?

Scandinavian
Posts: 24
Joined: Sat May 07, 2011 4:20 am

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Scandinavian » Thu Jan 10, 2019 5:49 am

siamond wrote:
Tue Jan 08, 2019 10:51 pm
Amadis_of_Gaul wrote:
Tue Jan 08, 2019 9:46 pm
The data suggest, then, that bonds are better at protecting against shallow risk, but stocks are better at protecting against deep risk. Frankly, I'm not that concerned about shallow risk. Given enough money, enough time, and enough diversification, an investor can come through whatever a "normal" market might throw at him just fine. Deep risk, though, to the extent that you can address it at all, you have to address it with stocks.

Amirite?
Yes, indeed. When thinking of 'risk' for a retiree while going beyond the naive 'risk == volatility' mantra, you quickly end up analyzing effects over the entire retirement period, hence addressing deep(er) risks. Which are, imho, much more meaningful. Then yes, when running corresponding simulations (which I personally did at length), a 75/25 asset allocation during retirement suddenly start making a lot more sense than (flawed) common wisdom would dictate... And even more if you happen to be an early retiree like me.

This really should be combined with a variable withdrawal method though, not with the non-sensical constant withdrawal method that retirement researchers keep beating to death.
Siamond, I recall reading some of your analysis of a high equity allocation over longer periods. I specifically remember a graph showing how when equities has outperformed, they have done so by a lot. When they have underperformed, they have done so by very little. If you could point me to the thread where I read this, I would be grateful!

User avatar
HomerJ
Posts: 12133
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by HomerJ » Thu Jan 10, 2019 9:43 am

CurlyDave wrote:
Wed Jan 09, 2019 3:48 am
And, the concept of demanding a constant, inflation indexed income in retirement, when for the 45 years of prior adult life I made my outgo fit my income rather than the other way around, seems unrealistic.
Maybe I'm different from most people here, but my income was pretty steady throughout my adult life or increasing.

So retirement where my income fluctuates year to year will indeed be very different for me.
To top it off, there are clearly non-economic risks in retirement which are seldom mentioned. Who has done a study explicitly considering my risk of dying before I run out of money? A 65 year old man in the US has a life expectancy of less than 20 years. A woman just slightly over 20. When my chances of dying greatly exceed my chances of running out of money I might be spending too much time thinking about the wrong risk...
Very good point indeed.
The J stands for Jay

User avatar
nedsaid
Posts: 10933
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by nedsaid » Thu Jan 10, 2019 11:23 am

willthrill81 wrote:
Wed Jan 09, 2019 11:17 am
nedsaid wrote:
Tue Jan 08, 2019 9:54 pm
willthrill81 wrote:
Tue Jan 08, 2019 8:18 pm
Random Walker wrote:
Tue Jan 08, 2019 3:19 pm
The bucket approach is appealing psychologically for mental accounting, but is not financially optimal.
It may be more than just mental accounting. In a deep bear market for stocks, retirees may hesitate to rebalance their 'safe' bonds into the stocks that have recently plummeted in value for fear of 'rebalancing into oblivion'. In a prior thread, I compared the effect of rebalancing or not for year 2000 retirees using the '4% rule' for withdrawals and found that rebalancing resulted in a large improvement in portfolio performance as compared to not. But I'd say that it's safe to say that many retirees hesitated to rebalance their portfolios in late 2008.
I was 49 years old at the time and I was hesitant to rebalance my portfolio from bonds to stocks. I was scared. So I did the next best thing and put 100% of my new monies for investment into stocks for a year.
Thanks for sharing. I very much doubt that you were alone in your fear. That's why I believe that the bucket approach may be preferable to some. If it enables them to avoid panic selling, then it's probably better than a fixed AA.
There is also the Taylor Larimore withdrawal method. In good years, splurge a bit and in bad years cut back. Sometimes I think we way overthink this stuff.
A fool and his money are good for business.

Always passive
Posts: 320
Joined: Fri Apr 14, 2017 4:25 am
Location: Israel

Re: Larry Swedroe: Does The Bucket Approach Destroy Wealth?

Post by Always passive » Thu Jan 10, 2019 11:40 am

I think that the strategy that makes the most sense to me is the one offered by Zvi Bodie some years back. Build a TIPS ladder to cover the necessary retirement expenses with the rest going to equities.
I do not do that exactly, but keep a 10 year ladder.
It may not be the most efficient system if performance is the target, but it gives the retiree peace of mind.
BTW, I do not think that performance should be the goal for retirees; rather, making sure that the retiree dies before running out of money.

Post Reply