A question about glide paths.

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

A question about glide paths.

Post by Jaymover »

Hi

I am 52 year old DIY investor who takes an interest, and has chosen a moderately aggressive glide path to follow with my investments as I get older and move toward retirement (maybe 2035).

The question I have is; after a major correction and a subsequent gradual recovery, say a 50 percent fall and a 7 year recovery period, should I keep following the glide path during each rebalance or should I somehow "suspend" the glide path until the market fully recovers and only then glide more aggressively toward my target retirement allocation?

For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
User avatar
dogagility
Posts: 3237
Joined: Fri Feb 24, 2017 5:41 am

Re: A question about glide paths.

Post by dogagility »

Jaymover wrote: Wed Jun 09, 2021 4:49 am Hi
The question I have is; after a major correction and a subsequent gradual recovery, say a 50 percent fall and a 7 year recovery period, should I keep following the glide path during each rebalance or should I somehow "suspend" the glide path until the market fully recovers and only then glide more aggressively toward my target retirement allocation?

For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
I plan to manage sequence risk by having flexibility. Reducing portfolio withdrawals akin to what's baked into the VPW method, altering my retirement date, and/or finding employment if it really hits the fan. I'm not a fan of the cost of a bond tent insurance to my portfolio.

In the history of the US market, declines of 40-50% were near or at the bottom. It may behoove an investor to not further decrease equity exposure at such lows... at least in the past.
Make sure you check out my list of certifications. The list is short, and there aren't any. - Eric 0. from SMA
David Althaus
Posts: 418
Joined: Wed Feb 14, 2018 7:05 pm

Re: A question about glide paths.

Post by David Althaus »

Once you retire you will find that each day is the first day of retirement. That means sequence of return risk is always with you and (unfortunately) there is no perfect answer. At 73 we have seven years living expenses in fixed assets. This makes us comfortable that we can withstand (historically) every bear market event except a Great Depression. In that case it's unlikely all our planning and thinking will matter much anyway. More years in fixed is the more dependable method for handling sequence of return risk.

Catastrophic events in the market are an actuarial certainty. Problem is you won't know what they are or when they come. All the more reason to think in terms of years in living expenses as your fail safe. You'll have to pick the number of years with which you are comfortable.

All the best
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
theorist
Posts: 1329
Joined: Sat Sep 28, 2019 11:39 am

Re: A question about glide paths.

Post by theorist »

vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
I like the look this method. In my investing journey I have heard repeatedly that it is better to set and work toward financial goals rather than "wealth maximisation"

Doing some back of the envelope calcs, if I set my retirement goals to be $1.5 million (PV) and 60/40 AA on retirement, my current total wealth suggests a 92 percent stock allocation for now (However I am currently only about 76 percent stocks). That also suggests that if there is a big crash in the next year then I might have to have more than 100 percent in stocks. This additionally suggests that my retirement goal is unrealistic unless I save more, work longer or get lucky with returns. The estimated returns I use in my model are about 3 percent real. Hard truths but really good to think about.
User avatar
Sandtrap
Posts: 19591
Joined: Sat Nov 26, 2016 5:32 pm
Location: Hawaii No Ka Oi - white sandy beaches, N. Arizona 1 mile high.

Re: A question about glide paths.

Post by Sandtrap »

Jaymover wrote: Wed Jun 09, 2021 4:49 am Hi

I am 52 year old DIY investor who takes an interest, and has chosen a moderately aggressive glide path to follow with my investments as I get older and move toward retirement (maybe 2035).

The question I have is; after a major correction and a subsequent gradual recovery, say a 50 percent fall and a 7 year recovery period, should I keep following the glide path during each rebalance or should I somehow "suspend" the glide path until the market fully recovers and only then glide more aggressively toward my target retirement allocation?

For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
This book might be helpful to you.
The Ages of the Investor: A Critical Look at Life-cycle Investing (Investing for Adults; [Book 1]) Paperback – August 28, 2012
by W. Bernstein
The print edition is better than the Kindle because of the charts.
Here it is on Amazon.com
https://smile.amazon.com/Ages-Investor- ... 218&sr=8-1
It covers all that you mention including "Sequence of Returns Risk" (also many excellent threads on this so search the archives).

j :D
Wiki Bogleheads Wiki: Everything You Need to Know
Elysium
Posts: 4120
Joined: Mon Apr 02, 2007 6:22 pm

Re: A question about glide paths.

Post by Elysium »

Jaymover wrote: Wed Jun 09, 2021 7:08 am
I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
I like the look this method. In my investing journey I have heard repeatedly that it is better to set and work toward financial goals rather than "wealth maximisation"

Doing some back of the envelope calcs, if I set my retirement goals to be $1.5 million (PV) and 60/40 AA on retirement, my current total wealth suggests a 92 percent stock allocation for now (However I am currently only about 76 percent stocks). That also suggests that if there is a big crash in the next year then I might have to have more than 100 percent in stocks. This additionally suggests that my retirement goal is unrealistic unless I save more, work longer or get lucky with returns. The estimated returns I use in my model are about 3 percent real. Hard truths but really good to think about.
I wrote about this method here. The important thing is to set realistic goals, as you said if it says your stock allocation should be higher then it means you have under-accumulated relative to your goals, and you need to save more, reduce your goals, or get lucky with returns. I would look at first two as the practical options, and the last one is built-in to the calculation, because if you get lucky with returns then it will tell you that you can reduce risk.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

theorist wrote: Wed Jun 09, 2021 6:22 am If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Yes, that's the basic idea. There are some refinements someone could add to the approach, of course, but that's the core premise.

For instance, an investor might change the 100% stocks to something lower if they were unusually risk-averse (e.g. using an 80/20 fund like Vanguard LifeStrategy Growth [VASGX] instead of a stock fund).

Or the investor might keep the equity target defined above as their goal but use savings bonds or a ladder of individual TIPS for the majority of their bond allocation. In that case, they might prefer to use what S&P calls a "lockbox" approach in which the savings bonds or TIPS are not used to rebalance once purchased.

Both of those refinements add a more conservative bent to the approach and therefore probably dictate increasing the savings rate to compensate.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Morse Code
Posts: 374
Joined: Wed Mar 14, 2007 12:02 pm
Location: Northern Michigan

Re: A question about glide paths.

Post by Morse Code »

theorist wrote: Wed Jun 09, 2021 6:22 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Here's a similar method that I am using which may be helpful to consider. If your target balance is 1.5M and your target retirement allocation is 60/40, you intend to have $900,000 in equity at retirement (1.5M * 0.6) and conversely, $600,000 in fixed income. So buy your stocks first. Maintain 100% equity until you reach $900,000, then begin to acquire fixed income with new contributions and rebalancing until you arrive at retirement with $1.5M in your nest egg and a 60/40 allocation.
Livin' the dream
absolute zero
Posts: 1244
Joined: Thu Dec 29, 2016 3:59 pm

Re: A question about glide paths.

Post by absolute zero »

Morse Code wrote: Wed Jun 09, 2021 7:55 am
theorist wrote: Wed Jun 09, 2021 6:22 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Here's a similar method that I am using which may be helpful to consider. If your target balance is 1.5M and your target retirement allocation is 60/40, you intend to have $900,000 in equity at retirement (1.5M * 0.6) and conversely, $600,000 in fixed income. So buy your stocks first. Maintain 100% equity until you reach $900,000, then begin to acquire fixed income with new contributions and rebalancing until you arrive at retirement with $1.5M in your nest egg and a 60/40 allocation.
I don’t think that’s just a similar method - it sounds like what you are using is precisely what vineviz suggested.

Personally I find it appealing since the concept is very intuitive to me.
absolute zero
Posts: 1244
Joined: Thu Dec 29, 2016 3:59 pm

Re: A question about glide paths.

Post by absolute zero »

vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I recall, you (and others including Ben Mathew and Steve Reading) have talked about this approach, but have typically incorporated a process whereby the target stock allocation is recalculated each year using a discount rate and a couple other variables like future savings.

I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

2) Your retirement asset allocation.
Another important decision is how do you choose the target allocation on retirement? Me choosing 60/40 was arbitrary. However when I work it out, 40 percent of my retirement target is about 10 years of expenses in bonds which seems about right. If I found out during accumulation years that my expenses are more than I predicted (eg due to higher than expected forecast medical costs)then I might change my retirement allocation to 50/50 which would result in me tapering my AA more aggressively. Lower returns but reduces the sequencing risk of possibly running out of money way to early.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: A question about glide paths.

Post by Random Walker »

Sandtrap wrote: Wed Jun 09, 2021 7:14 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am Hi

I am 52 year old DIY investor who takes an interest, and has chosen a moderately aggressive glide path to follow with my investments as I get older and move toward retirement (maybe 2035).

The question I have is; after a major correction and a subsequent gradual recovery, say a 50 percent fall and a 7 year recovery period, should I keep following the glide path during each rebalance or should I somehow "suspend" the glide path until the market fully recovers and only then glide more aggressively toward my target retirement allocation?

For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
This book might be helpful to you.
The Ages of the Investor: A Critical Look at Life-cycle Investing (Investing for Adults; [Book 1]) Paperback – August 28, 2012
by W. Bernstein
The print edition is better than the Kindle because of the charts.
Here it is on Amazon.com
https://smile.amazon.com/Ages-Investor- ... 218&sr=8-1
It covers all that you mention including "Sequence of Returns Risk" (also many excellent threads on this so search the archives).

j :D
“This book might be helpful to you” is a major league big time understatement! It’s mandatory reading to address the exact question you are asking! The part you will be most interested in is near the end of the book.

Dave
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

“This book might be helpful to you” is a major league big time understatement! It’s mandatory reading to address the exact question you are asking! The part you will be most interested in is near the end of the book.
WIll definitely get a copy!
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: A question about glide paths.

Post by Marseille07 »

Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.
Imo, if you do a glidepath then you should stick with it. Obviously in this case the glidepath looks bad, but what if there was another crash instead of a 7 years of recovery? That's why you were gliding to begin with.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

absolute zero wrote: Wed Jun 09, 2021 8:13 am I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
This simpler method results in a glide path that is more sensitive to your assumption about future market returns (since that goes directly into the estimate for "target retirement savings"). If you use an inaccurate estimate for expected returns (either much higher or lower than what you actually experience) you'll get bigger swings in the asset allocation. I also seem to recall that the simpler method typically generates a slightly steeper glide path, but overall the outcomes will be very similar.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Morse Code
Posts: 374
Joined: Wed Mar 14, 2007 12:02 pm
Location: Northern Michigan

Re: A question about glide paths.

Post by Morse Code »

absolute zero wrote: Wed Jun 09, 2021 8:03 am
Morse Code wrote: Wed Jun 09, 2021 7:55 am
theorist wrote: Wed Jun 09, 2021 6:22 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Here's a similar method that I am using which may be helpful to consider. If your target balance is 1.5M and your target retirement allocation is 60/40, you intend to have $900,000 in equity at retirement (1.5M * 0.6) and conversely, $600,000 in fixed income. So buy your stocks first. Maintain 100% equity until you reach $900,000, then begin to acquire fixed income with new contributions and rebalancing until you arrive at retirement with $1.5M in your nest egg and a 60/40 allocation.
I don’t think that’s just a similar method - it sounds like what you are using is precisely what vineviz suggested.

Personally I find it appealing since the concept is very intuitive to me.
Okay, maybe I didn't read vineviz' careful enough. At any rate, I rephrased it for clarity. I agree. I struggled with a glide path that made sense to me until I stumbled upon this method and I knew it was for me. The idea of targeting a balance rather than an age is key for me.
Livin' the dream
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

Morse Code wrote: Wed Jun 09, 2021 10:28 am
absolute zero wrote: Wed Jun 09, 2021 8:03 am
Morse Code wrote: Wed Jun 09, 2021 7:55 am
theorist wrote: Wed Jun 09, 2021 6:22 am
vineviz wrote: Wed Jun 09, 2021 6:05 am

The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I understand the basic idea here, you’d suggest 100% stocks up until the time you reach a portfolio value exceeding the estimated stock portion of your target amount. Is this right? (Not disputing, just trying to understand.)
Here's a similar method that I am using which may be helpful to consider. If your target balance is 1.5M and your target retirement allocation is 60/40, you intend to have $900,000 in equity at retirement (1.5M * 0.6) and conversely, $600,000 in fixed income. So buy your stocks first. Maintain 100% equity until you reach $900,000, then begin to acquire fixed income with new contributions and rebalancing until you arrive at retirement with $1.5M in your nest egg and a 60/40 allocation.
I don’t think that’s just a similar method - it sounds like what you are using is precisely what vineviz suggested.

Personally I find it appealing since the concept is very intuitive to me.
Okay, maybe I didn't read vineviz' careful enough. At any rate, I rephrased it for clarity. I agree. I struggled with a glide path that made sense to me until I stumbled upon this method and I knew it was for me. The idea of targeting a balance rather than an age is key for me.
This method makes sense. But figuring out the balance at the start of retirement can get complicated if the investor does not intend to keep working till the balance is reached. So for example if they plan to retire at age 65 with whatever they have accumulated, then figuring out how much they will have accumulated by the start of retirement may require some modeling.

Also, you would need to discount the $900,000 (present value, not inflation adjust) to calculate current year's stock target.
Last edited by Ben Mathew on Wed Jun 09, 2021 12:47 pm, edited 1 time in total.
Total Portfolio Allocation and Withdrawal (TPAW)
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

absolute zero wrote: Wed Jun 09, 2021 8:13 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I recall, you (and others including Ben Mathew and Steve Reading) have talked about this approach, but have typically incorporated a process whereby the target stock allocation is recalculated each year using a discount rate and a couple other variables like future savings.

I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
The starting retirement balance that is an input into the simple formula would be a function of future savings and expected return. Trying to estimate that carefully would get you back to the more complicated approach.

Also, the simple formula above is not discounting the dollar allocation to stocks--i.e $60 in stocks at retirement is equivalent to less than $60 in stocks today. So the correct allocation to stocks would be less than $60 today--a lot less if there are many years left to retirement.
Total Portfolio Allocation and Withdrawal (TPAW)
absolute zero
Posts: 1244
Joined: Thu Dec 29, 2016 3:59 pm

Re: A question about glide paths.

Post by absolute zero »

Ben Mathew wrote: Wed Jun 09, 2021 12:39 pm
absolute zero wrote: Wed Jun 09, 2021 8:13 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I recall, you (and others including Ben Mathew and Steve Reading) have talked about this approach, but have typically incorporated a process whereby the target stock allocation is recalculated each year using a discount rate and a couple other variables like future savings.

I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
The starting retirement balance that is an input into the simple formula would be a function of future savings and expected return. Trying to estimate that carefully would get you back to the more complicated approach.

Also, the simple formula above is not discounting the dollar allocation to stocks--i.e $60 in stocks at retirement is equivalent to less than $60 in stocks today. So the correct allocation to stocks would be less than $60 today--a lot less if there are many years left to retirement.
This makes sense to me now. If targeting a number (and not an age) for retirement then can just use the number itself in the calculation and keep it easy. Thanks.
esteen
Posts: 439
Joined: Thu May 23, 2019 12:31 am

Re: A question about glide paths.

Post by esteen »

vineviz wrote: Wed Jun 09, 2021 10:13 am
absolute zero wrote: Wed Jun 09, 2021 8:13 am I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
This simpler method results in a glide path that is more sensitive to your assumption about future market returns (since that goes directly into the estimate for "target retirement savings"). If you use an inaccurate estimate for expected returns (either much higher or lower than what you actually experience) you'll get bigger swings in the asset allocation. I also seem to recall that the simpler method typically generates a slightly steeper glide path, but overall the outcomes will be very similar.
Do I understand the bolded section above to mean that in your suggested method, the "target retirement savings" number is a future value? I typically think of my retirement "target" in today's dollars. How would this change your method?

Thanks!
This post is for entertainment or information only, and should not be construed as professional financial advice. | | "Invest your money passively and your time actively" -Michael LeBoeuf
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

Ben Mathew wrote: Wed Jun 09, 2021 12:39 pm Also, the simple formula above is not discounting the dollar allocation to stocks--i.e $60 in stocks at retirement is equivalent to less than $60 in stocks today. So the correct allocation to stocks would be less than $60 today--a lot less if there are many years left to retirement.
It's true that the more years until retirement the more differences you might see between the formal approach and the simpler approach I outlined above.

It's also true that with many years until retirement there's so much uncertainty about all the parameters that it's going to be hard to ascertain which approach is more "correct". And with many years to retirement, the impact of being a little more conservative or a little more aggressive isn't going to matter much since there's typically enough human capital left to smooth over the errors as retirement gets closer.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

esteen wrote: Wed Jun 09, 2021 1:33 pm
vineviz wrote: Wed Jun 09, 2021 10:13 am
absolute zero wrote: Wed Jun 09, 2021 8:13 am I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
This simpler method results in a glide path that is more sensitive to your assumption about future market returns (since that goes directly into the estimate for "target retirement savings"). If you use an inaccurate estimate for expected returns (either much higher or lower than what you actually experience) you'll get bigger swings in the asset allocation. I also seem to recall that the simpler method typically generates a slightly steeper glide path, but overall the outcomes will be very similar.
Do I understand the bolded section above to mean that in your suggested method, the "target retirement savings" number is a future value? I typically think of my retirement "target" in today's dollars. How would this change your method?
There are ways to estimate the glide path that use a bit more care with dealing with proper discount rates and growth rates. The simple method I outlined above is effectively expressing the future value in current dollars which, as Ben alluded to, generates a slightly higher equity allocation for longer .
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Horton
Posts: 1480
Joined: Mon Jan 21, 2008 2:53 pm
Contact:

Re: A question about glide paths.

Post by Horton »

vineviz wrote: Wed Jun 09, 2021 6:05 am
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.
I haven’t crunched any numbers, but intuition tells me this might be a challenging (or inefficient) approach if you are risk averse and want a low equity allocation in retirement (eg, 30%). You could fill up the equity allocation early if you are a saver and then spend a long time (decades?) just buying bonds. What do you think?
80% global equities (faith-based tilt) + 20% TIPS (LDI)
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

Horton wrote: Wed Jun 09, 2021 5:02 pm
vineviz wrote: Wed Jun 09, 2021 6:05 am
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.
I haven’t crunched any numbers, but intuition tells me this might be a challenging (or inefficient) approach if you are risk averse and want a low equity allocation in retirement (eg, 30%). You could fill up the equity allocation early if you are a saver and then spend a long time (decades?) just buying bonds. What do you think?
A highly risk averse investor will definitely pivot to bonds earlier than an investor who is less risk averse, but that’s exactly what a risk-averse investor SHOULD do.

So I don’t think it’s inefficient so much as it is appropriate for their risk profile.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

Horton wrote: Wed Jun 09, 2021 5:02 pm
vineviz wrote: Wed Jun 09, 2021 6:05 am
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.
I haven’t crunched any numbers, but intuition tells me this might be a challenging (or inefficient) approach if you are risk averse and want a low equity allocation in retirement (eg, 30%). You could fill up the equity allocation early if you are a saver and then spend a long time (decades?) just buying bonds. What do you think?
The model on which that vineviz is basing this rule-of-thumb is a lifecycle model where risk is being evenly spread across time. From that perspective, it's appropriate that the risk averse person you describe who wants a low 30% equity allocation in retirement should stick to a correspondingly low equity during working years. One way of looking at it is, do you care whether you lose 30% of your retirement funds to a stock market crash at age 45 or at age 65? If you don't care, then take equal risk at age 45 and at age 65. You should be as likely to lose 30% of your retirement income at age 45 as you are at age 65. Equalizing stock exposure at different ages does just that.

If on the other hand you feel that you would rather lose 30% of your retirement income at age 45 than at age 65--maybe you feel you can adjust better by working longer or saving more--that's an argument for maintaining a higher stock allocation at younger ages than at older ages. Vanguard's glidepath for target date funds broadly follows the shape suggested by the lifecycle model--flat at 100% in early career, sharp slope down in late career, turning flat at target AA at retirement. But they modify it in a couple different ways. One is that they cap stock allocation at 90% in a nod to human psychology. But the more interesting modification in the context of this discussion is that they maintain a higher stock allocation in early retirement than the eventual target in mid and late retirement. They justify this by saying that retirees may still have the option to go back to work in early retirement, so are able to take more risk than they can in mid to late retirement. That would be an example of taking on more stock risk earlier because people have more options to adjust in case of poor market performance.
Total Portfolio Allocation and Withdrawal (TPAW)
Horton
Posts: 1480
Joined: Mon Jan 21, 2008 2:53 pm
Contact:

Re: A question about glide paths.

Post by Horton »

Ben Mathew wrote: Wed Jun 09, 2021 5:46 pm
If on the other hand you feel that you would rather lose 30% of your retirement income at age 45 than at age 65--maybe you feel you can adjust better by working longer or saving more--that's an argument for maintaining a higher stock allocation at younger ages than at older ages. Vanguard's glidepath for target date funds broadly follows the shape suggested by the lifecycle model--flat at 100% in early career, sharp slope down in late career, turning flat at target AA at retirement. But they modify it in a couple different ways. One is that they cap stock allocation at 90% in a nod to human psychology. But the more interesting modification in the context of this discussion is that they maintain a higher stock allocation in early retirement than the eventual target in mid and late retirement. They justify this by saying that retirees may still have the option to go back to work in early retirement, so are able to take more risk than they can in mid to late retirement. That would be an example of taking on more stock risk earlier because people have more options to adjust in case of poor market performance.
Thanks Ben. Makes me wonder how Fidelity arrived at their glidepath.
80% global equities (faith-based tilt) + 20% TIPS (LDI)
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

vineviz wrote: Wed Jun 09, 2021 4:57 pm
esteen wrote: Wed Jun 09, 2021 1:33 pm
vineviz wrote: Wed Jun 09, 2021 10:13 am
absolute zero wrote: Wed Jun 09, 2021 8:13 am I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
This simpler method results in a glide path that is more sensitive to your assumption about future market returns (since that goes directly into the estimate for "target retirement savings"). If you use an inaccurate estimate for expected returns (either much higher or lower than what you actually experience) you'll get bigger swings in the asset allocation. I also seem to recall that the simpler method typically generates a slightly steeper glide path, but overall the outcomes will be very similar.
Do I understand the bolded section above to mean that in your suggested method, the "target retirement savings" number is a future value? I typically think of my retirement "target" in today's dollars. How would this change your method?
There are ways to estimate the glide path that use a bit more care with dealing with proper discount rates and growth rates. The simple method I outlined above is effectively expressing the future value in current dollars which, as Ben alluded to, generates a slightly higher equity allocation for longer .
Thanks for the insight here. I have modelled my accumulation over the next 13 years and have also calculated present Value of my total wealth to the point where it is (hopefully) enough. If I were to use future values to calculate my stock allocation then I dont think the model works.

Eg
Current value of holdings at 52 years = $1,000,000
Value of holdings in 13 years (with saving) = $2,000,000
PV of holdings in 13 years = $1,500,000
Target allocation on retirement = 60/40
Amount in stocks required now using VINEVIZs method = $1.200,000 (120 percent stock allocation needed today - no way)
Amount in stocks required now using VINEVIZs method based on future PV = $900,000 (90 percent stocks needed today - doable but pretty aggressive)

So it seems as if you need to calculate the PV of retirement amount for the VINEVIZ system to work (unless you are really well off)

I think what I like about the Vineviz system is that it encourages investors to go all in stocks after major corrections. Even in the Japan scenario this will work as if capital values remain terminally low then at least all in stocks will be do slightly better than the alternatives (eg FTSE index has been low for years but yields are currently around 5%). It would be hard to ignore the disappointment however.

It also gives me a goal eg. "By the time I drop out of the workforce I want $900,000 in growth assets (in PV) and 10 years of living expenses in bonds"
klaus14
Posts: 953
Joined: Sun Nov 25, 2018 6:43 pm

Re: A question about glide paths.

Post by klaus14 »

I am using* below method that vineviz mentioned in another thread.
It doesn't require estimating retirement savings.

1- Estimate annual retirement expenses: e.g.: $50k
2- Your current savings: e.g.: $1M
3- Then your bond allocation is: $1M/$50k = 20 -> 20%

Then you can retire when you have $1.5M @ 70/30 or $2M @ 60/40 or somewhere in between depending on your risk tolerance.

I think this is a good method today due to very low bond yields. That is: In order to "afford" a large bond allocation, you need to save more. Or you need to take more risk with stocks and/or have flexibility.
And i think this is a good method for folks aiming early retirement since it is independent of age.

*I also modify this based on current ERP. Current forward ERP estimation is 4.24%. Fair ERP is 5%. Difference is 0.76%. I multiply this by 10 and add it to bond alloc above. So: 20% + 7.6% = 27.6% is my current bond allocation target. I update this every month as Prof. Damodaran updates and direct new savings accordingly.
My investment algorithm: https://www.bogleheads.org/forum/viewtopic.php?f=10&t=351899&p=6112869#p6112869
aristotelian
Posts: 12277
Joined: Wed Jan 11, 2017 7:05 pm

Re: A question about glide paths.

Post by aristotelian »

I see glide path and rebalancing policy as separate issues. Glide path determines your target allocation at any given time. Rebalancing policy determined how closely you stick to it. I don't see any one size fits all for all investors. If you wait until the market recovers you will miss buying opportunities at the bottom. If you rebalance on the way down you are taking risk the market goes down further.

You could split the difference and rebalance during the crash up to a certain amount, putting an absolute limit on the amount you will rebalance. E.g. if you are starting with 12 years expenses in bonds you might say that you will rebalance during a crash as long as bond allocation is > 10 years expenses.
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: A question about glide paths.

Post by Random Walker »

I started this thread a while back, based on William Bernstein’s book. Customizing the Glidepath. Curious what people think.

viewtopic.php?t=236967

Dave
Elysium
Posts: 4120
Joined: Mon Apr 02, 2007 6:22 pm

Re: A question about glide paths.

Post by Elysium »

Jaymover wrote: Wed Jun 09, 2021 7:55 pm
Thanks for the insight here. I have modelled my accumulation over the next 13 years and have also calculated present Value of my total wealth to the point where it is (hopefully) enough. If I were to use future values to calculate my stock allocation then I dont think the model works.

Eg
Current value of holdings at 52 years = $1,000,000
Value of holdings in 13 years (with saving) = $2,000,000
PV of holdings in 13 years = $1,500,000
Target allocation on retirement = 60/40
Amount in stocks required now using VINEVIZs method = $1.200,000 (120 percent stock allocation needed today - no way)
Amount in stocks required now using VINEVIZs method based on future PV = $900,000 (90 percent stocks needed today - doable but pretty aggressive)

So it seems as if you need to calculate the PV of retirement amount for the VINEVIZ system to work (unless you are really well off)

I think what I like about the Vineviz system is that it encourages investors to go all in stocks after major corrections. Even in the Japan scenario this will work as if capital values remain terminally low then at least all in stocks will be do slightly better than the alternatives (eg FTSE index has been low for years but yields are currently around 5%). It would be hard to ignore the disappointment however.

It also gives me a goal eg. "By the time I drop out of the workforce I want $900,000 in growth assets (in PV) and 10 years of living expenses in bonds"
** Corrected **

Or, you can use my system linked upthread.

Target AA = (Current Balance / Target Balance) * Target Fixed Income in retirement.
using your numbers:
Target AA = (1,000,000 / 2,000,000) * 40%

Results in 20% in fixed income, 80% in equities.

For me, this works out to about 70% equities and 30% fixed income at the moment. As you noted, it will dynamically shift whenever there is a major correction, as it did for me in March 2020, and then again other way at present.
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

Elysium wrote: Thu Jun 10, 2021 8:59 am
Jaymover wrote: Wed Jun 09, 2021 7:55 pm
Thanks for the insight here. I have modelled my accumulation over the next 13 years and have also calculated present Value of my total wealth to the point where it is (hopefully) enough. If I were to use future values to calculate my stock allocation then I dont think the model works.

Eg
Current value of holdings at 52 years = $1,000,000
Value of holdings in 13 years (with saving) = $2,000,000
PV of holdings in 13 years = $1,500,000
Target allocation on retirement = 60/40
Amount in stocks required now using VINEVIZs method = $1.200,000 (120 percent stock allocation needed today - no way)
Amount in stocks required now using VINEVIZs method based on future PV = $900,000 (90 percent stocks needed today - doable but pretty aggressive)

So it seems as if you need to calculate the PV of retirement amount for the VINEVIZ system to work (unless you are really well off)

I think what I like about the Vineviz system is that it encourages investors to go all in stocks after major corrections. Even in the Japan scenario this will work as if capital values remain terminally low then at least all in stocks will be do slightly better than the alternatives (eg FTSE index has been low for years but yields are currently around 5%). It would be hard to ignore the disappointment however.

It also gives me a goal eg. "By the time I drop out of the workforce I want $900,000 in growth assets (in PV) and 10 years of living expenses in bonds"
** Corrected **

Or, you can use my system linked upthread.

Target AA = (Current Balance / Target Balance) * Target Fixed Income in retirement.
using your numbers:
Target AA = (1,000,000 / 2,000,000) * 40%

Results in 20% in fixed income, 80% in equities.

For me, this works out to about 70% equities and 30% fixed income at the moment. As you noted, it will dynamically shift whenever there is a major correction, as it did for me in March 2020, and then again other way at present.

I like this approach actually. If there is a major correction then it tilts toward a higher allocation but not as aggressively as the VINEVIZ method. Kind of a compromise.

It also encourages contrarianism, ie if you have to rebalance during a bear market then you will be encouraged to do it a bit more aggressively. The next question is when should one rebalance? Yearly seems to be the suggestion, or when your AA varies by 5 percent or more (eg during or after a selloff or boom. Of course the above method means that we will be needing to rebalance more often after a bull or bear market as the target will be moving in the opposite direction to the market so maybe the variance needs to be 10 percent.

This has been a very helpful post. Thanks!
Elysium
Posts: 4120
Joined: Mon Apr 02, 2007 6:22 pm

Re: A question about glide paths.

Post by Elysium »

Jaymover wrote: Fri Jun 11, 2021 2:06 am
Elysium wrote: Thu Jun 10, 2021 8:59 am
Jaymover wrote: Wed Jun 09, 2021 7:55 pm
Thanks for the insight here. I have modelled my accumulation over the next 13 years and have also calculated present Value of my total wealth to the point where it is (hopefully) enough. If I were to use future values to calculate my stock allocation then I dont think the model works.

Eg
Current value of holdings at 52 years = $1,000,000
Value of holdings in 13 years (with saving) = $2,000,000
PV of holdings in 13 years = $1,500,000
Target allocation on retirement = 60/40
Amount in stocks required now using VINEVIZs method = $1.200,000 (120 percent stock allocation needed today - no way)
Amount in stocks required now using VINEVIZs method based on future PV = $900,000 (90 percent stocks needed today - doable but pretty aggressive)

So it seems as if you need to calculate the PV of retirement amount for the VINEVIZ system to work (unless you are really well off)

I think what I like about the Vineviz system is that it encourages investors to go all in stocks after major corrections. Even in the Japan scenario this will work as if capital values remain terminally low then at least all in stocks will be do slightly better than the alternatives (eg FTSE index has been low for years but yields are currently around 5%). It would be hard to ignore the disappointment however.

It also gives me a goal eg. "By the time I drop out of the workforce I want $900,000 in growth assets (in PV) and 10 years of living expenses in bonds"
** Corrected **

Or, you can use my system linked upthread.

Target AA = (Current Balance / Target Balance) * Target Fixed Income in retirement.
using your numbers:
Target AA = (1,000,000 / 2,000,000) * 40%

Results in 20% in fixed income, 80% in equities.

For me, this works out to about 70% equities and 30% fixed income at the moment. As you noted, it will dynamically shift whenever there is a major correction, as it did for me in March 2020, and then again other way at present.

I like this approach actually. If there is a major correction then it tilts toward a higher allocation but not as aggressively as the VINEVIZ method. Kind of a compromise.

It also encourages contrarianism, ie if you have to rebalance during a bear market then you will be encouraged to do it a bit more aggressively. The next question is when should one rebalance? Yearly seems to be the suggestion, or when your AA varies by 5 percent or more (eg during or after a selloff or boom. Of course the above method means that we will be needing to rebalance more often after a bull or bear market as the target will be moving in the opposite direction to the market so maybe the variance needs to be 10 percent.

This has been a very helpful post. Thanks!
Rebalance based on percentage variation from target, minimum 5% and no more than 10%, giving a band like that allows for flexibility because sometimes you do not want to trigger a rebalance at 5% and may allow it to go to 7% instead, but keeping an upper band will allow keeping risk in check. That's what I do, last March I let it go 7% off target before buying equities and that helped. As of now I am perhaps again 7% off target, but haven't sold equities yet but it gets closer to that 10% band then I will.
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: A question about glide paths.

Post by HootingSloth »

Ben Mathew wrote: Wed Jun 09, 2021 12:39 pm
absolute zero wrote: Wed Jun 09, 2021 8:13 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I recall, you (and others including Ben Mathew and Steve Reading) have talked about this approach, but have typically incorporated a process whereby the target stock allocation is recalculated each year using a discount rate and a couple other variables like future savings.

I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
The starting retirement balance that is an input into the simple formula would be a function of future savings and expected return. Trying to estimate that carefully would get you back to the more complicated approach.

Also, the simple formula above is not discounting the dollar allocation to stocks--i.e $60 in stocks at retirement is equivalent to less than $60 in stocks today. So the correct allocation to stocks would be less than $60 today--a lot less if there are many years left to retirement.
Sorry if I missed this, but what discount rate is appropriate for the slightly-more-complicated-than-simple approach? I have effectively done something with results quite close to having a slowly-growing emergency fund plus this method using a 3% real discount rate, but never dove much into the lifecycle approach to understand it well.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

HootingSloth wrote: Fri Jun 11, 2021 6:45 pm
Ben Mathew wrote: Wed Jun 09, 2021 12:39 pm
absolute zero wrote: Wed Jun 09, 2021 8:13 am
vineviz wrote: Wed Jun 09, 2021 6:05 am
Jaymover wrote: Wed Jun 09, 2021 4:49 am For instance, if over the 7 years of recovery the unlucky investor glides from a 70/30 allocation to a 60/40 allocation then it looks as if they miss out on some of the opportunity through the 7 year period by tilting more and more each year toward bonds.

I see that some Bogleheads model various scenarios and there may be some evidence in doing things differently to the typical set and forget target date fund. Or perhaps there is a different principle out there that one could use to manage sequencing risk.
The glide path used by target date funds is a generalized approximation: it works well on average, but you can manage your own glide path simply (IMHO) and easily with two numbers:

1) Your target retirement savings;
2) Your retirement asset allocation.


Let's say you expect to retire in 10 years with $100 and that you've determined that 60/40 is the asset allocation you want for the duration of your retirement.

Take your retirement equity allocation (60%) and multiply it by your target retirement savings ($100) . This tells you how much money to have invested in equities at any given point in time.

In this case it is $60.

Let's say you're 10 years from retirement and your portfolio has only grown (so far to $90). Allocate $60 to stocks and $30 to bonds, for an AA of 66/33.

Let's say the portfolio drops to $80 next week. Rebalance so you have $60 in stocks and $20 in bonds, for an AA of 75/25.

Let's say that two years later the portfolio has rebounded to $105 due to contributions and market growth. Allocate $60 to stocks and $45 to bonds, for an AA of 57/43.

Etc.

I'm glossing over a few nuances, but honestly this simple method works quite well as long as you are at least moderately confident about your estimate for the two numbers.
If I recall, you (and others including Ben Mathew and Steve Reading) have talked about this approach, but have typically incorporated a process whereby the target stock allocation is recalculated each year using a discount rate and a couple other variables like future savings.

I find the simpler method that you’ve described in this current thread to be way more appealing and easy to understand. Any thoughts on the potential downfalls of this method vs the more “sophisticated” approach? The simple method just produces a slightly more aggressive allocation during the glide years, correct?
The starting retirement balance that is an input into the simple formula would be a function of future savings and expected return. Trying to estimate that carefully would get you back to the more complicated approach.

Also, the simple formula above is not discounting the dollar allocation to stocks--i.e $60 in stocks at retirement is equivalent to less than $60 in stocks today. So the correct allocation to stocks would be less than $60 today--a lot less if there are many years left to retirement.
Sorry if I missed this, but what discount rate is appropriate for the slightly-more-complicated-than-simple approach? I have effectively done something with results quite close to having a slowly-growing emergency fund plus this method using a 3% real discount rate, but never dove much into the lifecycle approach to understand it well.
The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Total Portfolio Allocation and Withdrawal (TPAW)
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: A question about glide paths.

Post by HootingSloth »

Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: A question about glide paths.

Post by vineviz »

HootingSloth wrote: Fri Jun 11, 2021 8:01 pm
Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.

And the rate should reflect, to some degree at least, thenriskiness of your human capital. That said, for reasons, you should probably use a rate no higher than the yield on investment grade corporate bonds.

If you view your human capital as particularly risky I’d still not use a rate above 3%.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

vineviz wrote: Sat Jun 12, 2021 5:46 am
HootingSloth wrote: Fri Jun 11, 2021 8:01 pm
Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.
Nominal estimates of future savings and pensions should be discounted by a nominal rate. Real (inflation adjusted) estimates of future savings and pensions should be discounted by a real rate. Both methods will result in the same present value.
Total Portfolio Allocation and Withdrawal (TPAW)
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: A question about glide paths.

Post by HootingSloth »

vineviz wrote: Sat Jun 12, 2021 5:46 am
HootingSloth wrote: Fri Jun 11, 2021 8:01 pm
Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.

And the rate should reflect, to some degree at least, thenriskiness of your human capital. That said, for reasons, you should probably use a rate no higher than the yield on investment grade corporate bonds.

If you view your human capital as particularly risky I’d still not use a rate above 3%.
After looking at this again, I think what I am doing ends up being quite similar to this method with a discount rate around 0.8% real. (It seems much easier for me to think in real than nominal terms when talking about decades in the future, and pensions and SS will have COLA). The biggest difference is that in early stages I never went above 80% equities (I count my emergency fund as part of my portfolio for AA purposes, so would not have wanted to be at 100%). It may be just a coincidence or may be that my more qualitative thought process ended up getting me to a fairly similar place to this approach. Anyway, thanks to both of you for explaining more about the method. Always good to have different ways of looking at and evaluating my plan.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
Ben Mathew
Posts: 2720
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: A question about glide paths.

Post by Ben Mathew »

HootingSloth wrote: Sat Jun 12, 2021 3:11 pm
vineviz wrote: Sat Jun 12, 2021 5:46 am
HootingSloth wrote: Fri Jun 11, 2021 8:01 pm
Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.

And the rate should reflect, to some degree at least, thenriskiness of your human capital. That said, for reasons, you should probably use a rate no higher than the yield on investment grade corporate bonds.

If you view your human capital as particularly risky I’d still not use a rate above 3%.
After looking at this again, I think what I am doing ends up being quite similar to this method with a discount rate around 0.8% real. (It seems much easier for me to think in real than nominal terms when talking about decades in the future, and pensions and SS will have COLA). The biggest difference is that in early stages I never went above 80% equities (I count my emergency fund as part of my portfolio for AA purposes, so would not have wanted to be at 100%). It may be just a coincidence or may be that my more qualitative thought process ended up getting me to a fairly similar place to this approach. Anyway, thanks to both of you for explaining more about the method. Always good to have different ways of looking at and evaluating my plan.
There are two ways to be conservative about the future savings assumption so as not to go overboard on stocks today:

1. Assume normal future savings, but go conservative with a high discount rate to discount it with.
2. Assume conservatively low future savings, but use the TIPS yield to discount it with.

I find the latter to be easier to wrap my head around. e.g. If I expect savings to be about $30,000 per year, then maybe $20,000 feels quite safe. Or maybe $15,000 does. I can understand and evaluate this dollar figure. The discount rate on the other hand is harder to get a feel for. Is 0.5% above TIPS conservative enough? Does it need to be 1% above TIPS? Does it depend on how old I am and how many years of savings I have left? Etc.
Total Portfolio Allocation and Withdrawal (TPAW)
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: A question about glide paths.

Post by HootingSloth »

Ben Mathew wrote: Sat Jun 12, 2021 4:34 pm
HootingSloth wrote: Sat Jun 12, 2021 3:11 pm
vineviz wrote: Sat Jun 12, 2021 5:46 am
HootingSloth wrote: Fri Jun 11, 2021 8:01 pm
Ben Mathew wrote: Fri Jun 11, 2021 7:40 pm The goal is to maintain a fixed allocation on the total portfolio, which includes future savings and pensions counted as bonds. The discount rate needed to calculate the total portfolio is the appropriate bond rate to calculate the present value of future savings and pensions. If you use a conservative real estimate of future savings and pensions, you could use TIPS yields of matching duration as the discount rate (available here). Once you have this value, add it to current savings to get the value of the total portfolio. Apply the desired asset allocation to the total portfolio to get current stock allocation.

Example:

Current savings = $400,000
PV of future savings and pensions (discounted using bond rates) = $500,000
Total portfolio = $400,000 + $500,000 = $900,000
Asset allocation on total portfolio = 30/70 fixed
Stock exposure = 30% of $900,000 = $270,000

So asset allocation on current savings portfolio = $270,000/$130,000 which works out to 68/32. We are maintaining 30/70 on the total portfolio by holding 68/32 on the savings portfolio.

This approach is implemented in Total portfolio allocation and withdrawal (TPAW).
Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.

And the rate should reflect, to some degree at least, thenriskiness of your human capital. That said, for reasons, you should probably use a rate no higher than the yield on investment grade corporate bonds.

If you view your human capital as particularly risky I’d still not use a rate above 3%.
After looking at this again, I think what I am doing ends up being quite similar to this method with a discount rate around 0.8% real. (It seems much easier for me to think in real than nominal terms when talking about decades in the future, and pensions and SS will have COLA). The biggest difference is that in early stages I never went above 80% equities (I count my emergency fund as part of my portfolio for AA purposes, so would not have wanted to be at 100%). It may be just a coincidence or may be that my more qualitative thought process ended up getting me to a fairly similar place to this approach. Anyway, thanks to both of you for explaining more about the method. Always good to have different ways of looking at and evaluating my plan.
There are two ways to be conservative about the future savings assumption so as not to go overboard on stocks today:

1. Assume normal future savings, but go conservative with a high discount rate to discount it with.
2. Assume conservatively low future savings, but use the TIPS yield to discount it with.

I find the latter to be easier to wrap my head around. e.g. If I expect savings to be about $30,000 per year, then maybe $20,000 feels quite safe. Or maybe $15,000 does. I can understand and evaluate this dollar figure. The discount rate on the other hand is harder to get a feel for. Is 0.5% above TIPS conservative enough? Does it need to be 1% above TIPS? Does it depend on how old I am and how many years of savings I have left? Etc.
That sounds totally fair. I did not get to this glidepath by thinking in terms of a discount rate.

I chose my glidepath by first thinking more generally about what types of problems were most important to solve, at the margin, with different dollars at different life stages, then thinking about what assets were best suited to solving those problems, and finally figuring out a way to smoothly and automatically steer around those goals at different points of time so that I had an objective formula for my spreadsheet and was not trying to make decisions on short timescales.

I thought it was interesting that this qualitative method produced an answer that seemed quite close to an entirely different and highly quantitative method, so was trying to see how close it came and with what kinds of parameters.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
Topic Author
Jaymover
Posts: 214
Joined: Wed May 12, 2021 8:19 pm

Re: A question about glide paths.

Post by Jaymover »

HootingSloth wrote: Sat Jun 12, 2021 4:46 pm
Ben Mathew wrote: Sat Jun 12, 2021 4:34 pm
HootingSloth wrote: Sat Jun 12, 2021 3:11 pm
vineviz wrote: Sat Jun 12, 2021 5:46 am
HootingSloth wrote: Fri Jun 11, 2021 8:01 pm

Thanks. It sounds like I am not too close to doing this after all, if the TIPS yield is the appropriate discount rate, as that would mean I should be overshooting my target to account for the negative real yields. From what you say, it seems to make sense that this might not be the right method for me, because modelling future savings as bond-like is not a great fit for my situation.
The discount rate used to compute the PV of future savings should be a nominal rate, not a real rate.

And the rate should reflect, to some degree at least, thenriskiness of your human capital. That said, for reasons, you should probably use a rate no higher than the yield on investment grade corporate bonds.

If you view your human capital as particularly risky I’d still not use a rate above 3%.
After looking at this again, I think what I am doing ends up being quite similar to this method with a discount rate around 0.8% real. (It seems much easier for me to think in real than nominal terms when talking about decades in the future, and pensions and SS will have COLA). The biggest difference is that in early stages I never went above 80% equities (I count my emergency fund as part of my portfolio for AA purposes, so would not have wanted to be at 100%). It may be just a coincidence or may be that my more qualitative thought process ended up getting me to a fairly similar place to this approach. Anyway, thanks to both of you for explaining more about the method. Always good to have different ways of looking at and evaluating my plan.
There are two ways to be conservative about the future savings assumption so as not to go overboard on stocks today:

1. Assume normal future savings, but go conservative with a high discount rate to discount it with.
2. Assume conservatively low future savings, but use the TIPS yield to discount it with.

I find the latter to be easier to wrap my head around. e.g. If I expect savings to be about $30,000 per year, then maybe $20,000 feels quite safe. Or maybe $15,000 does. I can understand and evaluate this dollar figure. The discount rate on the other hand is harder to get a feel for. Is 0.5% above TIPS conservative enough? Does it need to be 1% above TIPS? Does it depend on how old I am and how many years of savings I have left? Etc.
That sounds totally fair. I did not get to this glidepath by thinking in terms of a discount rate.

I chose my glidepath by first thinking more generally about what types of problems were most important to solve, at the margin, with different dollars at different life stages, then thinking about what assets were best suited to solving those problems, and finally figuring out a way to smoothly and automatically steer around those goals at different points of time so that I had an objective formula for my spreadsheet and was not trying to make decisions on short timescales.

I thought it was interesting that this qualitative method produced an answer that seemed quite close to an entirely different and highly quantitative method, so was trying to see how close it came and with what kinds of parameters.
Thanks everyone for your input as this has been a breakthrough post for me. This one post has helped me focus on my retirement financial goals (some 10-15 years off) rather than some notional idea of stocks versus bonds should be x at age y. I could only do this by modelling 15 years into the future using a simple Excel spreadsheet with conservative assumptions on returns, taxes, income, expenses, rent etc.

I might get there (25x future expenses in todays dollars). However I am going to work on getting just a little bit more income through a side hustle or doing OT etc to improve my chances. I realise that I also have to remain engaged with risk. However my assets doing better than expected will allow me to back off a little with the risk hopefully. I am aiming for a 60/40 balanced portfolio which I expect will be required by me and many others in 2035 and no less.

Some people are talking about discount rates, bonds = pension etc. However you are really better off working out what income you need over and above future estimated pensions to work out what you want to retire with. Then use that

So for me right now it is approx 1,000,000*0.4/2,000,000 = 20 percent bonds. This does not include my 1 year of expenses in a CD.

I think the important thing is checking in once a year or after a significant event (eg windfall, terminal diagnosis, retrenchment, huge promotion) and looking at not only my portfolio but my model and consider making some changes (probably with an hour or two with an independent financial advisor to assist before making the changes). The old model may need to be thrown out the window as all the assumptions that I based my future asset position on are probably toast.

Thanks once again and best of luck whatever financial situation you are in!
Post Reply