Follow-up discussion on AA around one's retirement date

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Follow-up discussion on AA around one's retirement date

Post by David Jay »

[UPDATE] If you are just joining this thread, a careful review of the graphics in the post of May 12 @ 4:54 PM can really help clarify the sequence of returns subject.

This thread is a break-out from a discussion that was hijacking the "rule of thumb" thread by "vineviz", here: viewtopic.php?f=10&t=314214

I am willing to answer any questions and continue any conversations here on this new thread.

For background, "JustinR" posted a graph of a possible AA curve to which I - ahem - took exception (far too bluntly). The next morning I went back and edited my reply to provide an explanation of why I felt it was a less than stellar idea. JustinR then posted this revised graph, too which I responded "much better":

Image

At the bottom my reply I mentioned what I was doing, as follows:

I was 100% stocks until about 5 years before retirement (In retrospect I would recommend others to start earlier). Retired @62 with an AA of 45/55. I am letting my AA drift up, planning 60/40 at age 70. [note for clarification: ...and hold at 60/40.]

That started a lot of discussion, so I edited JustinR's graph with a visual showing what a bond tent would look like if applied to JustinR's plan.

Image

That prompted even more discussion (which, in my opinion derailed the thread). Let's continue the discussion here...
Last edited by David Jay on Fri May 15, 2020 8:36 am, edited 6 times in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

I want to make one point at the beginning, because the bond tent seems to create a lot of consternation. I think that is because it is somewhat counter-intuitive to those who have been thinking about retirement for many years and have always seen equity exposure drop in all the graphs they have seen. Then they see a graph where equity exposure is increasing, even for just 5 or 10 years, and it seems completely "wrong" (causes cognitive dissonance, to use big words):

The bond tent is not about increasing equity exposure over the retirement years. It is about decreasing equity exposure during the few years when a portfolio has it's greatest sequence-of-return and portfolio size risks.

Using the second graph above, note that in the first case (red line) the long term retirement goal is a 60/40 portfolio. Note that in the second case (the blue line), the long term retirement goal is also 60/40. A bond tent does not seek to be more aggressive in the out-years of retirement. A bond tent is a temporary deviation from the long term plan, strictly as a risk-reduction feature.
Last edited by David Jay on Tue May 12, 2020 9:58 am, edited 1 time in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around the retirement years

Post by Triple digit golfer »

Is the idea here simply to be invested more conservatively when you have the most money, which is just before retirement, at retirement, and just after retirement? Or am I off the mark?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Triple digit golfer wrote: Tue May 12, 2020 9:46 am Is the idea here simply to be invested more conservatively when you have the most money, which is just before retirement, at retirement, and just after retirement? Or am I off the mark?
You nailed it.

To stick with the graph example, an individual has selected a 60/40 portfolio for the remainder of their lives. This is true for both the red curve and the blue curve. But in the blue curve, the portfolio has taken a "temporary" drop in equity to minimize the risk associated with the years immediately around the retirement date.

Studies have shown that sequence-of-return risk is mostly in the first 10 years.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Doc
Posts: 9807
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Follow-up discussion on AA around one's retirement date

Post by Doc »

David Jay wrote: Tue May 12, 2020 9:38 am I want to make one point at the beginning, because the bond tent seems to create a lot of consternation. I think that is because it is somewhat counter-intuitive to those who have been thinking about retirement for many years and have always seen equity exposure drop in all the graphs they have seen. Then they see a graph where equity exposure is increasing, even for just 5 or 10 years, and it seems completely "wrong" (causes cognitive dissonance, to use big words):
I used a bond tent with a change from "retirement" to "after retirement" but not for the reasons you suggest. I made the change because our retirement income and expenses were such that we had "enough" and started to move our investment plan towards the grandchildren and less to ourselves.

I also recently changed the 50/50 +/- 5% target to 50/50 +10%/-5% for similar reasons.
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around one's retirement date

Post by rbaldini »

Correct me if I'm wrong, but sequence of returns risk always exists after retirement. It's not like it goes away after the first x years. It is always the case that losing a bunch now is worse than later. I suppose one difference is that you are x years older, and therefore your long term needs are reduced (in other words, you are closer to death), so maybe there isn't the same need to guard against that risk?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

rbaldini wrote: Tue May 12, 2020 10:40 am Correct me if I'm wrong, but sequence of returns risk always exists after retirement. It's not like it goes away after the first x years. It is always the case that losing a bunch now is worse than later. I suppose one difference is that you are x years older, and therefore your long term needs are reduced (in other words, you are closer to death), so maybe there isn't the same need to guard against that risk?
Sustainable withdrawal rates are strongly affected by the early years. Compounding effects occur in both positive and negative directions. A major downturn in first 5 years after retirement has a much greater effect on portfolio longevity than in - say - the 15th year of retirement.

This is essentially the definition of “sequence of returns” risk. The sequence makes a dramatic difference. Let’s take a 30 year retirement, broken up into 5 year periods. Your portfolio will go through four “up” 5 year periods and two “down” 5 year periods. Now let’s say you have a 6% (inflation adjusted) withdrawal rate. If the first two 5 year periods are both “down”, you may end up eating dog food. If the first two 5 year periods are “up”, you will likely be just fine.
Last edited by David Jay on Tue May 12, 2020 11:06 am, edited 1 time in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around one's retirement date

Post by rbaldini »

David Jay wrote: Tue May 12, 2020 10:55 am
rbaldini wrote: Tue May 12, 2020 10:40 am Correct me if I'm wrong, but sequence of returns risk always exists after retirement. It's not like it goes away after the first x years. It is always the case that losing a bunch now is worse than later. I suppose one difference is that you are x years older, and therefore your long term needs are reduced (in other words, you are closer to death), so maybe there isn't the same need to guard against that risk?
Sustainable withdrawal rates are strongly affected by the early years. Compounding effects occur in both positive and negative directions. A major downturn in first 5 years after retirement has a much greater effect on portfolio longevity than in - say - the 15th year of retirement.
Let's say you are in the 10th year of retirement *right now*. Is it not the case that "a major downturn" right now is much worse that a major downturn in 10 more years? (Replace 10 with any number you want).

In other words, *at every moment in retirement*, it is always worse to lose a lot in the near term than in the far term. In other words, there is always the same sequence of return risk, because the constant withdrawal rate reduces your ability to make up the difference later. If this requires you to be conservative at year 1 of your retirement, should it not require the same every year after?
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around the retirement years

Post by Triple digit golfer »

David Jay wrote: Tue May 12, 2020 9:56 am
Triple digit golfer wrote: Tue May 12, 2020 9:46 am Is the idea here simply to be invested more conservatively when you have the most money, which is just before retirement, at retirement, and just after retirement? Or am I off the mark?
You nailed it.

To stick with the graph example, an individual has selected a 60/40 portfolio for the remainder of their lives. This is true for both the red curve and the blue curve. But in the blue curve, the portfolio has taken a "temporary" drop in equity to minimize the risk associated with the years immediately around the retirement date.

Studies have shown that sequence-of-return risk is mostly in the first 10 years.
Thanks for confirming. I like it because it is logical. When I have the most money, I have the most to lose and therefore should be invested more conservatively.
User avatar
Uncorrelated
Posts: 1050
Joined: Sun Oct 13, 2019 3:16 pm

Re: Follow-up discussion on AA around the retirement years

Post by Uncorrelated »

David Jay wrote: Tue May 12, 2020 9:56 am Studies have shown that sequence-of-return risk is mostly in the first 10 years.
Ehm. Excuse me, but it would be more accurate to say that blogs have shown that sequence of returns risk is mostly in the first 10 years. I don't recall any serious academic using the words sequence of returns risk. All "risk" that is described by sequence of returns risk is just a subset of ordinary investment risk.

My research has shown that glidepaths in retirement generally have a negligible effect on the safe withdrawal rate compared to a constant asset allocation. For example, here is (for an arbitrary chosen retirement duration) the survival chance chance for various asset allocations, with randomly selected returns (no autocorrelation).
Image
As you can see the survival chance for the optimally chosen glidepath is only barely better than the optimal chosen static asset allocation. Smarter strategies are able to do much better.

research by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.

There is little point in attempting to "beat" this asset allocation by inventing your own glide path.

I would describe the image in the opening post as market timing. The suggested asset allocation violates the principle of optimality, It's not obvious but this eventually results in the strategy suggesting different asset allocations to people that are in identical financial situations.
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

rbaldini wrote: Tue May 12, 2020 10:58 am
David Jay wrote: Tue May 12, 2020 10:55 am
rbaldini wrote: Tue May 12, 2020 10:40 am Correct me if I'm wrong, but sequence of returns risk always exists after retirement. It's not like it goes away after the first x years. It is always the case that losing a bunch now is worse than later. I suppose one difference is that you are x years older, and therefore your long term needs are reduced (in other words, you are closer to death), so maybe there isn't the same need to guard against that risk?
Sustainable withdrawal rates are strongly affected by the early years. Compounding effects occur in both positive and negative directions. A major downturn in first 5 years after retirement has a much greater effect on portfolio longevity than in - say - the 15th year of retirement.
Let's say you are in the 10th year of retirement *right now*. Is it not the case that "a major downturn" right now is much worse that a major downturn in 10 more years? (Replace 10 with any number you want).

In other words, *at every moment in retirement*, it is always worse to lose a lot in the near term than in the far term. In other words, there is always the same sequence of return risk, because the constant withdrawal rate reduces your ability to make up the difference later. If this requires you to be conservative at year 1 of your retirement, should it not require the same every year after?
What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around the retirement years

Post by rbaldini »

Uncorrelated wrote: Tue May 12, 2020 11:08 am research by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I am curious: how the optimal glidepath is defined? Feel free to use technical and mathematical terminology.
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around one's retirement date

Post by rbaldini »

David Jay wrote: Tue May 12, 2020 11:11 am What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
I understand that 2 million > 500 thousand. What I don't understand is why a person who is 5 years into retirement needs to care less about sequence of returns risk than someone 0 years in. The only reason I could see is that the former is closer to death, and therefore doesn't need quite as much money in the long run.
Juice3
Posts: 222
Joined: Sun Nov 05, 2017 7:40 am
Location: The Web
Contact:

Re: Follow-up discussion on AA around the retirement years

Post by Juice3 »

David Jay wrote: Tue May 12, 2020 11:18 am
Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
Here is (one of) the Kitces articles on Bond Tenting
https://www.kitces.com/blog/managing-po ... -red-zone/

I am not what "not serious academic" would be mean but I would characterize Kitces work is commercial in nature not academic but it obviously does have a strong learning bent thus a connection to the word academic.
User avatar
Uncorrelated
Posts: 1050
Joined: Sun Oct 13, 2019 3:16 pm

Re: Follow-up discussion on AA around the retirement years

Post by Uncorrelated »

David Jay wrote: Tue May 12, 2020 11:18 am
Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
I believe you are referring to this paper: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930

I think this paper is a joke. The argument boils down to "we tested 100 glidepaths and 10 constant asset allocations, in most cases the best asset allocation was a glidepath by a very tiny margin". Not surprising, you tested 10 times as many. The conclusion isn't necessarily wrong but the methodology does not give much confidence.

According to my reserarch, a glidepath during retirement offers only a very, very tiny improvement. In the context of fixed withdrawals, I don't think it's worth bothering with.

rbaldini wrote: Tue May 12, 2020 11:43 am
Uncorrelated wrote: Tue May 12, 2020 11:08 am research by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I am curious: how the optimal glidepath is defined? Feel free to use technical and mathematical terminology.
The optimal glidepath is the solution to Merton's portfolio problem with specific assumptions. These assumptions generally include stock/bond returns following some arbitrary distribution, an utility function that is generally a CRRA (constant relative risk aversion), and stochastic lifespan.

The research from Gordon Irlam is incredibly useful. For example, one look at this page should make it immediately obvious that the traditional concept of glidepaths (dependent only on time) does just not work. His methodology is described in detail in his papers: https://www.aacalc.com/about.
User avatar
Doc
Posts: 9807
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Follow-up discussion on AA around one's retirement date

Post by Doc »

rbaldini wrote: Tue May 12, 2020 11:47 am
David Jay wrote: Tue May 12, 2020 11:11 am What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
I understand that 2 million > 500 thousand. What I don't understand is why a person who is 5 years into retirement needs to care less about sequence of returns risk than someone 0 years in. The only reason I could see is that the former is closer to death, and therefore doesn't need quite as much money in the long run.
Exactly. If you $100K a year for the next 50 years until you intend to die you can put the first $5M in T-bills and assume absolutely no sequence of return risk. If your portofolio at the time is $8M you can take that extra $3M and put it in the most risky asset you can imagine and still not have any reason to worry except for possibly harm to your ego.

The problem with the whole idea of this thread is that "one size does not fit all".
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.
User avatar
Doc
Posts: 9807
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Follow-up discussion on AA around one's retirement date

Post by Doc »

Here's a good way to completely eliminate the sequence of returns risk. Use the first $699 to buy a Ruger 357 Blackhawk and use it to end your retirement completely if the rest of your money goes down the drain. :twisted:
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around the retirement years

Post by rbaldini »

Uncorrelated wrote: Tue May 12, 2020 12:23 pm
rbaldini wrote: Tue May 12, 2020 11:43 am
Uncorrelated wrote: Tue May 12, 2020 11:08 am research by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I am curious: how the optimal glidepath is defined? Feel free to use technical and mathematical terminology.
The optimal glidepath is the solution to Merton's portfolio problem with specific assumptions. These assumptions generally include stock/bond returns following some arbitrary distribution, an utility function that is generally a CRRA (constant relative risk aversion), and stochastic lifespan.

The research from Gordon Irlam is incredibly useful. For example, one look at this page should make it immediately obvious that the traditional concept of glidepaths (dependent only on time) does just not work. His methodology is described in detail in his papers: https://www.aacalc.com/about.
Many thanks for sharing. I was thinking of building my own stochastic simulations to determine an optimal asset allocation glide-path, based on the goal of "maximize expected return subject to the constraint of achieving a net worth of at least $x with probability >= y at retirement." Or something similar. I have seen models that use different objectives and constraints (e.g. minimize variance while achieving a given expected return) that I don't particularly like, so whenever someone says "optimal" without qualification, I get suspicious. In any case, no need to reinvent the wheel; I'll read your linked stuff first.
marcopolo
Posts: 3484
Joined: Sat Dec 03, 2016 10:22 am

Re: Follow-up discussion on AA around one's retirement date

Post by marcopolo »

rbaldini wrote: Tue May 12, 2020 11:47 am
David Jay wrote: Tue May 12, 2020 11:11 am What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
I understand that 2 million > 500 thousand. What I don't understand is why a person who is 5 years into retirement needs to care less about sequence of returns risk than someone 0 years in. The only reason I could see is that the former is closer to death, and therefore doesn't need quite as much money in the long run.

The reason it has less impact later is that during the intervening good years the portfolio has built up a larger excess (i.e., that starting 4% WR is now only 2.5% of the portfolio), so one can more easily tolerate a loss without impacting portfolio survivability.
Once in a while you get shown the light, in the strangest of places if you look at it right.
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around one's retirement date

Post by rbaldini »

marcopolo wrote: Tue May 12, 2020 1:09 pm
rbaldini wrote: Tue May 12, 2020 11:47 am
David Jay wrote: Tue May 12, 2020 11:11 am What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
I understand that 2 million > 500 thousand. What I don't understand is why a person who is 5 years into retirement needs to care less about sequence of returns risk than someone 0 years in. The only reason I could see is that the former is closer to death, and therefore doesn't need quite as much money in the long run.

The reason it has less impact later is that during the intervening good years the portfolio has built up a larger excess (i.e., that starting 4% WR is now only 2.5% of the portfolio), so one can more easily tolerate a loss without impacting portfolio survivability.
This assumes the portfolio grows.
marcopolo
Posts: 3484
Joined: Sat Dec 03, 2016 10:22 am

Re: Follow-up discussion on AA around one's retirement date

Post by marcopolo »

rbaldini wrote: Tue May 12, 2020 1:32 pm
marcopolo wrote: Tue May 12, 2020 1:09 pm
rbaldini wrote: Tue May 12, 2020 11:47 am
David Jay wrote: Tue May 12, 2020 11:11 am What you say is logically true in isolation, but if you have 2 million at year 10 due to good sequence of returns in those first 10 years, it is very different from having only 500K due to a bad sequence of returns in those first 10 years.
I understand that 2 million > 500 thousand. What I don't understand is why a person who is 5 years into retirement needs to care less about sequence of returns risk than someone 0 years in. The only reason I could see is that the former is closer to death, and therefore doesn't need quite as much money in the long run.

The reason it has less impact later is that during the intervening good years the portfolio has built up a larger excess (i.e., that starting 4% WR is now only 2.5% of the portfolio), so one can more easily tolerate a loss without impacting portfolio survivability.
This assumes the portfolio grows.
That is the whole point of the SORR discussion!
If it doesn't grow, you got hit with Sequence risk early.
Once in a while you get shown the light, in the strangest of places if you look at it right.
rbaldini
Posts: 1436
Joined: Mon Mar 23, 2015 3:20 pm

Re: Follow-up discussion on AA around one's retirement date

Post by rbaldini »

marcopolo wrote: Tue May 12, 2020 1:38 pm
rbaldini wrote: Tue May 12, 2020 1:32 pm
marcopolo wrote: Tue May 12, 2020 1:09 pm The reason it has less impact later is that during the intervening good years the portfolio has built up a larger excess (i.e., that starting 4% WR is now only 2.5% of the portfolio), so one can more easily tolerate a loss without impacting portfolio survivability.
This assumes the portfolio grows.
That is the whole point of the SORR discussion!
If it doesn't grow, you got hit with Sequence risk early.
Not so. You might have gotten the best possible sequence of the observed returns, but still have lost money.
marcopolo
Posts: 3484
Joined: Sat Dec 03, 2016 10:22 am

Re: Follow-up discussion on AA around one's retirement date

Post by marcopolo »

rbaldini wrote: Tue May 12, 2020 1:45 pm
marcopolo wrote: Tue May 12, 2020 1:38 pm
rbaldini wrote: Tue May 12, 2020 1:32 pm
marcopolo wrote: Tue May 12, 2020 1:09 pm The reason it has less impact later is that during the intervening good years the portfolio has built up a larger excess (i.e., that starting 4% WR is now only 2.5% of the portfolio), so one can more easily tolerate a loss without impacting portfolio survivability.
This assumes the portfolio grows.
That is the whole point of the SORR discussion!
If it doesn't grow, you got hit with Sequence risk early.
Not so. You might have gotten the best possible sequence of the observed returns, but still have lost money.
Well, sure that is possible, then you may want to consider Doc's advice above.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

Note that this is only a 2.6% withdrawal rate, showing the dramatic impact of sequence of returns. In case #2 we have depleted a portfolio in 30 years with a 2.6% withdrawal rate! (I started with a $1M portfolio when I built the spreadsheet, thinking of a 5% withdrawal rate. Case 2 ended up $3.1 million in the hole. I increased the starting portfolio number until case 2 became a positive number.)
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Uncorrelated wrote: Tue May 12, 2020 12:23 pm
David Jay wrote: Tue May 12, 2020 11:18 am
Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
I believe you are referring to this paper: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930

I think this paper is a joke.
Sorry it didn't meet your expectations.

May I suggest the following:

1. Forbes is familiar with Sequence of Returns: https://www.forbes.com/sites/jamiehopki ... 3dee4227eb

2. Retirement Income Journal is familiar with Sequence of Returns: https://retirementincomejournal.com/art ... ence-risk/

3. The AAII Journal is familiar with Sequence of Returns: https://www.aaii.com/journal/article/th ... our-wealth?

4. U.S. News is familiar with Sequence of Returns: https://money.usnews.com/investing/inve ... sk-matters

5. Fidelity is familiar with Sequence of Returns: https://www.fidelity.com/bin-public/060 ... cation.pdf

6. Allianz is familiar with Sequence of Returns: https://www.allianzlife.com/get-answers ... money-last

7. Vanguard is familiar with Sequence of Returns: https://institutional.vanguard.com/VGAp ... eOfReturns

8. Blackrock is familiar with Sequence of Returns: https://www.blackrock.com/pt/literature ... -va-us.pdf

9. Jackson National is familiar with Sequence of Returns: https://d1xhgr640tdb4k.cloudfront.net/5 ... eturns.pdf

10. American Century is familiar with Sequence of Returns: https://ipro.americancentury.com/conten ... -risk.html

I'm kind of surprised that you're unfamiliar with the concept... (but I truly did appreciate your passive-aggressive "blogs have shown" snark).
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
KEotSK66
Posts: 364
Joined: Wed Mar 11, 2020 7:03 pm

*

Post by KEotSK66 »

i agree with rbaldini, sor can be a problem outside that 10-yr window...something i said yesterday on the other thread

the 100% stock portfolio leading up to retirement screams sor risk

i like uncorrelated's input too, re a static AA. my impression is a ~60/40 would be a good fit going forward, i hinted at this yesterday too
"i just got fluctuated out of $1,500", jerry
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: *

Post by David Jay »

KEotSK66 wrote: Tue May 12, 2020 4:44 pmthe 100% stock portfolio leading up to retirement screams sor risk
May I suggest the Jackson National link above. Little if any sequence-of-return issue in accumulation, it is the withdrawals that cause the problem.

[edit] Those graphics provide such a clear illustration, I will post them here:

In accumulation:

Image

In decumulation:

Image
Last edited by David Jay on Tue May 12, 2020 5:41 pm, edited 1 time in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
KEotSK66
Posts: 364
Joined: Wed Mar 11, 2020 7:03 pm

dj

Post by KEotSK66 »

it is the withdrawals that cause the problem.

if you take a 50% haircut before you shift from 100/0 to 45/55 you'll see that statement is not complete
"i just got fluctuated out of $1,500", jerry
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: dj

Post by David Jay »

KEotSK66 wrote: Tue May 12, 2020 5:13 pm it is the withdrawals that cause the problem.

if you take a 50% haircut before you shift from 100/0 to 45/55 you'll see that statement is not complete
I was still working and had not announced a retirement date when I was at 100%. A 50% drop would have meant purchasing more stock in my 401K at bargain prices.

By the time I gave my boss a "defensive briefing", I was ramping down my equity percentage, probably somewhere between 70% and 60% stock. I didn't get all the way down to 45% stock until about 3 months before retirement.

In the original thread and again in the OP on this thread I stated that in retrospect, I should have started lowering my equity earlier. But in this thread and in the original thread you have pointed to my 100% equity position. How many times would you like me to state that in retrospect I should have started earlier?
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Uncorrelated
Posts: 1050
Joined: Sun Oct 13, 2019 3:16 pm

Re: Follow-up discussion on AA around the retirement years

Post by Uncorrelated »

David Jay wrote: Tue May 12, 2020 4:20 pm
Uncorrelated wrote: Tue May 12, 2020 12:23 pm
David Jay wrote: Tue May 12, 2020 11:18 am
Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
I believe you are referring to this paper: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930

I think this paper is a joke.
Sorry it didn't meet your expectations.

I'm kind of surprised that you're unfamiliar with the concept...
I'm familiar with the concept. I just think that the whole concept is poorly defined and overlaps greatly with other risks. Every time the term is mentioned it is followed by some analysis violating basic optimization principles. (For example, both wade pfau's paper and kitces blog violated the principle of optimality). This definitely doesn't help to give the term credibility.

Minimizing sequence of returns risk is not a proper goal. It's like minimizing inflation risk by purchasing real estate. Sure you reduced inflation risk, but was that really the best you can do? In practice you don't want to minimize risk X, you want to maximize some utility function, this requires a trade-off between many different kinds of unique risk. Over-analyzing a specific outcome or risk usually results in suboptimal solutions.

I highly recommend reading the research of Gordon Irlam. Conventional glidepaths that are often touted as a solution for "sequence of returns risk" are broken beyond repair. That was acceptable back in 1980 when there was insufficient computing power available to come up with a better solution, but isn't something that I consider to be even remotely acceptable today.
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Uncorrelated wrote: Tue May 12, 2020 5:54 pm
David Jay wrote: Tue May 12, 2020 4:20 pm
Uncorrelated wrote: Tue May 12, 2020 12:23 pm
David Jay wrote: Tue May 12, 2020 11:18 am
Uncorrelated wrote: Tue May 12, 2020 11:08 amI don't recall any serious academic using the words sequence of returns risk.
I would refer you to the Pfau and Kitces paper. Don’t know if that meets your definition of “serious”. I will provide a link later, I am on my way out the door right now.
I believe you are referring to this paper: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930

I think this paper is a joke.
Sorry it didn't meet your expectations.

I'm kind of surprised that you're unfamiliar with the concept...
I'm familiar with the concept. I just think that the whole concept is poorly defined and overlaps greatly with other risks. Every time the term is mentioned it is followed by some analysis violating basic optimization principles. (For example, both wade pfau's paper and kitces blog violated the principle of optimality). This definitely doesn't help to give the term credibility.

Minimizing sequence of returns risk is not a proper goal. It's like minimizing inflation risk by purchasing real estate. Sure you reduced inflation risk, but was that really the best you can do? In practice you don't want to minimize risk X, you want to maximize some utility function, this requires a trade-off between many different kinds of unique risk. Over-analyzing a specific outcome or risk usually results in suboptimal solutions.

I highly recommend reading the research of Gordon Irlam. Conventional glidepaths that are often touted as a solution for "sequence of returns risk" are broken beyond repair. That was acceptable back in 1980 when there was insufficient computing power available to come up with a better solution, but isn't something that I consider to be even remotely acceptable today.
Okay, thanks for the reply.

You may want to highlight the Irlam document in the original thread on "Rule of Thumb", the OP (vineviz) is pretty thorough and may like to see it.

[edit] Or you could do the community a favor by starting a brand new thread on Irlam's rules of thumb and see how it works as a discussion-starter.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Uncorrelated wrote: Tue May 12, 2020 11:08 amresearch by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I have had a chance to "play" with this formula this evening. It is interesting, the formula suggests that I should have been over 100% stocks throughout accumulation, as my net worth has never been greater than my future SS income (i.e. future income / current net worth > 1). I don't think I am a unique case, I think a lot of moderate income accumulators who apply this formula will end up with very high equity allocations.

I'm one of those individuals who can "go Spock" and over-ride the emotions created by market swings. I went through the flash crash, the dot com bubble and the financial crisis holding 100% equity. But the vast majority of accumulators are not able to do that.

So what does that say about the real-world applicability of this rule of thumb? It appears one can just discard it and operate on the basis of behavioral finance. That would likely be to select the highest AA that one can hold without selling out equity positions in a downturn. Which I have advocated for a long time - at least since 2008 after watching my sister sell all of her stock holdings at the very bottom of the financial crisis.

I will look at the decumulation formula tomorrow.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: dj

Post by David Jay »

David Jay wrote: Tue May 12, 2020 5:17 pmIn the original thread and again in the OP on this thread I stated that in retrospect, I should have started lowering my equity earlier.
I may have to re-think this in light of the Merton's Portfolio Model... :wink:
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around the retirement years

Post by Triple digit golfer »

David Jay wrote: Tue May 12, 2020 7:17 pm
Uncorrelated wrote: Tue May 12, 2020 11:08 amresearch by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I have had a chance to "play" with this formula this evening. It is interesting, the formula suggests that I should have been over 100% stocks throughout accumulation, as my net worth has never been greater than my future SS income (i.e. future income / current net worth > 1). I don't think I am a unique case, I think a lot of moderate income accumulators who apply this formula will end up with very high equity allocations.

I'm one of those individuals who can "go Spock" and over-ride the emotions created by market swings. I went through the flash crash, the dot com bubble and the financial crisis holding 100% equity. But the vast majority of accumulators are not able to do that.

So what does that say about the real-world applicability of this rule of thumb? It appears one can just discard it and operate on the basis of behavioral finance. That would likely be to select the highest AA that one can hold without selling out equity positions in a downturn. Which I have advocated for a long time - at least since 2008 after watching my sister sell all of her stock holdings at the very bottom of the financial crisis.

I will look at the decumulation formula tomorrow.
What am I doing wrong here? Let's do a made up example

Future income say 100k a year for 20 years. 2 million.
Current net worth 800k.

2 million divided by 800k equals 2.5.

Add 1 equals 3.5.

50% of 3.5 equals 1.75% stock allocation? Huh?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Triple digit golfer wrote: Tue May 12, 2020 8:15 pm
David Jay wrote: Tue May 12, 2020 7:17 pm
Uncorrelated wrote: Tue May 12, 2020 11:08 amresearch by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I have had a chance to "play" with this formula this evening. It is interesting, the formula suggests that I should have been over 100% stocks throughout accumulation, as my net worth has never been greater than my future SS income (i.e. future income / current net worth > 1). I don't think I am a unique case, I think a lot of moderate income accumulators who apply this formula will end up with very high equity allocations.

I'm one of those individuals who can "go Spock" and over-ride the emotions created by market swings. I went through the flash crash, the dot com bubble and the financial crisis holding 100% equity. But the vast majority of accumulators are not able to do that.

So what does that say about the real-world applicability of this rule of thumb? It appears one can just discard it and operate on the basis of behavioral finance. That would likely be to select the highest AA that one can hold without selling out equity positions in a downturn. Which I have advocated for a long time - at least since 2008 after watching my sister sell all of her stock holdings at the very bottom of the financial crisis.

I will look at the decumulation formula tomorrow.
What am I doing wrong here? Let's do a made up example

Future income say 100k a year for 20 years. 2 million.
Current net worth 800k.

2 million divided by 800k equals 2.5.

Add 1 equals 3.5.

50% of 3.5 equals 1.75% stock allocation? Huh?
Nothing. The formula recommends 100% stock for many people in accumulation. That is why I said that most people will make the AA decision based on behavioral issues, rather than the “optimization”.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around the retirement years

Post by Triple digit golfer »

David Jay wrote: Tue May 12, 2020 8:46 pm
Triple digit golfer wrote: Tue May 12, 2020 8:15 pm
David Jay wrote: Tue May 12, 2020 7:17 pm
Uncorrelated wrote: Tue May 12, 2020 11:08 amresearch by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I have had a chance to "play" with this formula this evening. It is interesting, the formula suggests that I should have been over 100% stocks throughout accumulation, as my net worth has never been greater than my future SS income (i.e. future income / current net worth > 1). I don't think I am a unique case, I think a lot of moderate income accumulators who apply this formula will end up with very high equity allocations.

I'm one of those individuals who can "go Spock" and over-ride the emotions created by market swings. I went through the flash crash, the dot com bubble and the financial crisis holding 100% equity. But the vast majority of accumulators are not able to do that.

So what does that say about the real-world applicability of this rule of thumb? It appears one can just discard it and operate on the basis of behavioral finance. That would likely be to select the highest AA that one can hold without selling out equity positions in a downturn. Which I have advocated for a long time - at least since 2008 after watching my sister sell all of her stock holdings at the very bottom of the financial crisis.

I will look at the decumulation formula tomorrow.
What am I doing wrong here? Let's do a made up example

Future income say 100k a year for 20 years. 2 million.
Current net worth 800k.

2 million divided by 800k equals 2.5.

Add 1 equals 3.5.

50% of 3.5 equals 1.75% stock allocation? Huh?
Nothing. The formula recommends 100% stock for many people in accumulation. That is why I said that most people will make the AA decision based on behavioral issues, rather than the “optimization”.
Is the 1.75% I came up with actually 175%?
palanzo
Posts: 1386
Joined: Thu Oct 10, 2019 4:28 pm

Re: Follow-up discussion on AA around one's retirement date

Post by palanzo »

David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
What is the AA for this portfolio? Are there 5 years of living expenses in cash?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around the retirement years

Post by David Jay »

Triple digit golfer wrote: Tue May 12, 2020 8:51 pm
David Jay wrote: Tue May 12, 2020 8:46 pm
Triple digit golfer wrote: Tue May 12, 2020 8:15 pm
David Jay wrote: Tue May 12, 2020 7:17 pm
Uncorrelated wrote: Tue May 12, 2020 11:08 amresearch by Gordon Irman has shown that the optimal glidepath for accumulation can be approximated with very simple rules. The asset allocation (stock allocation = 50% * ( 1 + future income / current net worth)) brings you within 5% of the optimal possible asset allocation.
I have had a chance to "play" with this formula this evening. It is interesting, the formula suggests that I should have been over 100% stocks throughout accumulation, as my net worth has never been greater than my future SS income (i.e. future income / current net worth > 1). I don't think I am a unique case, I think a lot of moderate income accumulators who apply this formula will end up with very high equity allocations.

I'm one of those individuals who can "go Spock" and over-ride the emotions created by market swings. I went through the flash crash, the dot com bubble and the financial crisis holding 100% equity. But the vast majority of accumulators are not able to do that.

So what does that say about the real-world applicability of this rule of thumb? It appears one can just discard it and operate on the basis of behavioral finance. That would likely be to select the highest AA that one can hold without selling out equity positions in a downturn. Which I have advocated for a long time - at least since 2008 after watching my sister sell all of her stock holdings at the very bottom of the financial crisis.

I will look at the decumulation formula tomorrow.
What am I doing wrong here? Let's do a made up example

Future income say 100k a year for 20 years. 2 million.
Current net worth 800k.

2 million divided by 800k equals 2.5.

Add 1 equals 3.5.

50% of 3.5 equals 1.75% stock allocation? Huh?
Nothing. The formula recommends 100% stock for many people in accumulation. That is why I said that most people will make the AA decision based on behavioral issues, rather than the “optimization”.
Is the 1.75% I came up with actually 175%?
Yes. I missed that (when you are familiar with the formula it’s easy to see what you expect to see), I automatically saw 175%.

The calculation is not 50% of 3.5, it is 50% times 3.5.
Last edited by David Jay on Tue May 12, 2020 9:05 pm, edited 1 time in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
Kookaburra
Posts: 344
Joined: Thu Apr 02, 2020 11:14 pm

Re: Follow-up discussion on AA around one's retirement date

Post by Kookaburra »

David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
Practically speaking, is SOR risk really an issue at a withdrawal rate of 3% if one’s AA is, say, static 50/50 going into retirement? In your example, 5 consecutive years with a portfolio down 20% would mean the market dropped 50% for each of 5 consecutive years.
Last edited by Kookaburra on Tue May 12, 2020 9:06 pm, edited 1 time in total.
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around one's retirement date

Post by Triple digit golfer »

palanzo wrote: Tue May 12, 2020 8:55 pm
David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
What is the AA for this portfolio? Are there 5 years of living expenses in cash?
Also want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

Kookaburra wrote: Tue May 12, 2020 9:05 pm
David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
Practically speaking, is SOR risk really an issue at a withdrawal rate of 3% if one’s AA is, say, static 50/50 going into retirement? In your example, 5 consecutive years with a portfolio down 20% would mean the market dropped 50% for each of 5 consecutive years.
It was a simple example to illustrate that SOR is real when making with withdrawals in the presence of return volatility, it was not meant to illustrate any particular portfolio.

If you have a bunch of extra money or if you hold a very low percentage of equities (low equities == low portfolio volatility) then SOR becomes a non-issue.

I don’t have a bunch of extra money so I chose the “low equity” route.
Last edited by David Jay on Mon Jun 08, 2020 5:10 pm, edited 2 times in total.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
palanzo
Posts: 1386
Joined: Thu Oct 10, 2019 4:28 pm

Re: Follow-up discussion on AA around one's retirement date

Post by palanzo »

David Jay wrote: Tue May 12, 2020 10:04 pm
Kookaburra wrote: Tue May 12, 2020 9:05 pm
David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
Practically speaking, is SOR risk really an issue at a withdrawal rate of 3% if one’s AA is, say, static 50/50 going into retirement? In your example, 5 consecutive years with a portfolio down 20% would mean the market dropped 50% for each of 5 consecutive years.
It was a simple example to illustrate that SOR is real when making with withdrawals in the presence of return volatility, it was not meant to illustrate any particular portfolio.

If you have a bunch of extra money or if you hold a low percentage of equities then SOR becomes a non-issue.

I don’t have a bunch of extra money so I chose the “low equity” route.
5 consecutive years with a portfolio down 20% would mean the market dropped 50% for each of 5 consecutive years.
Anything is possible of course but this is a pretty extreme case. Has this ever happened?
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

Triple digit golfer wrote: Tue May 12, 2020 9:06 pmAlso want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
I don’t think where you withdraw the funds matters for SOR.

It comes down to the varying percentage of portfolio spending that the withdrawal represents. 50,000 is 2.6% of 1.9M, but it is only 2.4% of 2.05M (10% gain) and it is nearly 3.3% of 1.52M (20% loss). So one is spending a much higher percentage of their portfolio after a loss.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
palanzo
Posts: 1386
Joined: Thu Oct 10, 2019 4:28 pm

Re: Follow-up discussion on AA around one's retirement date

Post by palanzo »

David Jay wrote: Tue May 12, 2020 10:17 pm
Triple digit golfer wrote: Tue May 12, 2020 9:06 pmAlso want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
I don’t think where you withdraw the funds matters for SOR.

It comes down to the varying percentage of portfolio spending that the withdrawal represents. 50,000 is 2.6% of 1.9M, but it is only 2.4% of 2.05M (10% gain) and it is nearly 3.3% of 1.52M (20% loss). So one is spending a much higher percentage of their portfolio after a loss.
If one follows Humble Dollar advice and keeps 5 years of spending in cash then a 5 year SOR is less of an issue.
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

palanzo wrote: Tue May 12, 2020 8:55 pm
David Jay wrote: Tue May 12, 2020 3:12 pm I didn't expect to be defending the concept of Sequence of Return Risk, it is well established in the financial planning community and seems self evident to this retired engineer. But let's put numbers to it:

Example (using real dollars):
$1.9M portfolio
$50K annual withdrawal for living expenses (for replication, expenses are withdrawn after portfolio return each year)
30 years
25 up years, 10% annual gain each year
5 down years, 20% loss each year

Position the down years at the end of the sequence (25 up years, then 5 down years), remaining portfolio after year 30: $ 4,966,203

Position the down years at the beginning of the sequence (5 down years, then 25 up years), remaining portfolio after year 30: $ 7,150

The sequence of the down years makes a difference, if the down years occur earlier in the sequence, the final value is lower.
What is the AA for this portfolio? Are there 5 years of living expenses in cash?
AA and cash are immaterial in the example. (10% up year could represent a 10% gain in equities with a 100/0 portfolio or it could be a 19% gain in equities with a 50/50 portfolio).
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
User avatar
Topic Author
David Jay
Posts: 9391
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Follow-up discussion on AA around one's retirement date

Post by David Jay »

palanzo wrote: Tue May 12, 2020 10:21 pm
David Jay wrote: Tue May 12, 2020 10:17 pm
Triple digit golfer wrote: Tue May 12, 2020 9:06 pmAlso want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
I don’t think where you withdraw the funds matters for SOR.

It comes down to the varying percentage of portfolio spending that the withdrawal represents. 50,000 is 2.6% of 1.9M, but it is only 2.4% of 2.05M (10% gain) and it is nearly 3.3% of 1.52M (20% loss). So one is spending a much higher percentage of their portfolio after a loss.
If one follows Humble Dollar advice and keeps 5 years of spending in cash then a 5 year SOR is less of an issue.
No it isn’t. See my explanation at 10:17PM.

SOR is not about “selling stocks low”, it is about the percentage of the portfolio that is withdrawn for expenses. If you don’t withdraw any money then there is no SOR risk, the commutative property of multiplication applies.
Prediction is very difficult, especially about the future - Niels Bohr | To get the "risk premium", you really do have to take the risk - nisiprius
Triple digit golfer
Posts: 5560
Joined: Mon May 18, 2009 5:57 pm

Re: Follow-up discussion on AA around one's retirement date

Post by Triple digit golfer »

David Jay wrote: Tue May 12, 2020 10:17 pm
Triple digit golfer wrote: Tue May 12, 2020 9:06 pmAlso want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
I don’t think where you withdraw the funds matters for SOR.

It comes down to the varying percentage of portfolio spending that the withdrawal represents. 50,000 is 2.6% of 1.9M, but it is only 2.4% of 2.05M (10% gain) and it is nearly 3.3% of 1.52M (20% loss). So one is spending a much higher percentage of their portfolio after a loss.
That makes sense. Have to view portfolio as a whole.
palanzo
Posts: 1386
Joined: Thu Oct 10, 2019 4:28 pm

Re: Follow-up discussion on AA around one's retirement date

Post by palanzo »

Triple digit golfer wrote: Tue May 12, 2020 10:28 pm
David Jay wrote: Tue May 12, 2020 10:17 pm
Triple digit golfer wrote: Tue May 12, 2020 9:06 pmAlso want to ask, doesn't how money is withdrawn affect the outcome? If one has a 60/40 portfolio and stocks decline and it goes to 55/45, wouldnt the first money be withdrawn from bonds until back to 60/40?
I don’t think where you withdraw the funds matters for SOR.

It comes down to the varying percentage of portfolio spending that the withdrawal represents. 50,000 is 2.6% of 1.9M, but it is only 2.4% of 2.05M (10% gain) and it is nearly 3.3% of 1.52M (20% loss). So one is spending a much higher percentage of their portfolio after a loss.
That makes sense. Have to view portfolio as a whole.
Yes, true.
Post Reply