Are risk-based guardrails approach to retirement withdrawals even desirable?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
silent cal
Posts: 29
Joined: Wed Feb 24, 2021 8:40 am

Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by silent cal »

Hello everyone,

I have been doing a deep dive into withdrawal strategies while I'm still a ways from retirement. I came across this article on Kitces.com https://www.kitces.com/blog/risk-based- ... n-hatchet/ describing a risk-based guardrails approach to portfolio withdrawals. It explains the downsides of a simple withdrawal-rate guardrails approach and the advantage of a probability-of-success guardrail strategy which factors in longevity risk, cash flows, income, and Monte Carlo results. There is even a new financial planning software to implement this exact method https://incomelaboratory.com.

I then also came across this article from Early Retirement Now https://earlyretirementnow.com/2017/02/ ... n-klinger/ which I thought laid out a pretty solid issue with variable withdrawal strategies in general.


My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).

Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.

Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
Parkinglotracer
Posts: 4057
Joined: Fri Dec 20, 2019 2:49 am
Location: Upstate NY

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Parkinglotracer »

You sound like a very analytical person which is good because we need them. Here is a view from an mba educated not so analytical person. Don’t think I don’t appreciate your analysis.

To answer your question, I read the executive summary of kitces article and I don’t know how to explain it any better than he did. If I was dependent upon a regular withdrawal for my retirement survival I would have some sort of guard rail to cut my spending to prevent me from running out of money.

While an I am an engineering major with years experience in a strict defense contractor environment - I am more of a TLAR (that looks about right) decision maker from my AF flying experience. Coming up with an exact formula seems unnecessary - like trying to make jello stick to a wall as we used to say in the military.

I agree planners might use complicated strategies to make their services seem essential. I wouldn’t throw the strategy idea out with the bath water just because some complicate it. That is what they get paid to do - complicate it.

If I started withdrawing 4%; a year and my portfolio performance made that amount eventually 6% of my portfolio that would set off a grinding noise in my brain thinking my retirement bus was hitting a guard rail and get me to move to a lower spending level or return to work.

What practical guard rail do you think you would use if any? maybe a strategy that is simple enough to explain to a non technical spouse who wants to understand what the plan is?

Call it common sense or a guardrail - if your lump sum is disappearing one might want to not spend as much if they want to have any left in retirement.
Wanderingwheelz
Posts: 3219
Joined: Mon Mar 04, 2019 8:52 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Wanderingwheelz »

I agree with parkinglotracer in all of the points he made.

A guard rails approach is simply another way of making sure you don’t crash and burn which is the basic bottom like to any plan that is to unfold over three decades or longer.

It’s doesn’t need to be strictly rules based since you’re already wired as a human being to survive.
Being wrong compounds forever.
Topic Author
silent cal
Posts: 29
Joined: Wed Feb 24, 2021 8:40 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by silent cal »

That makes a lot of sense. Maybe using a simple guardrails approach (that a spouse could understand) has a benefit of not necessarily guaranteeing you will get to enjoy a higher SWR in retirement, but putting a system in place that has an emotional benefit to cut expenses when the market is taking a dive. I think I like the idea of simplicity which is why the constant-dollar withdrawal approach coupled with a lower SWR appeals to me, but it seems there are very simple guardrail methods I could put in place too. Definitely something I will have to explore further :happy
User avatar
Ben Mathew
Posts: 2743
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Ben Mathew »

silent cal wrote: Sat Aug 26, 2023 7:51 pm My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).
The optimal allocation and withdrawal strategy coming out of economics is the lifecycle investing model. There is a long line of literature around this starting with Merton (1969) and Samuelson (1969). The withdrawal strategy in this model is a variable amortization based withdrawal (ABW) strategy. This is a flexible withdrawal method that can accommodate a wide range of reasonable preferences. By contrast, fixed SWR withdrawals and guardrail based variable withdrawals are ad hoc methods that are hard to reconcile with reasonable preferences.

Amortization based withdrawals (ABW) are easy to understand and implement. ABW does not imply high starting withdrawals. That’s controlled by a spending growth parameter (g) that depends on preferences. Starting withdrawals can be made as low as you want by amortizing the portfolio with a high g,

See the wiki on ABW for more on the ABW strategy. TPAW has an online planner that implements the lifecycle model and ABW withdrawals.
silent cal wrote: Sat Aug 26, 2023 7:51 pm Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.
It’s easy to implement an ABW strategy with a calculator or spreadsheet (the ABW wiki has several spreadsheets you can use.) Even a printed withdrawal rate table would work if you don’t plan to updated expected returns.The RMD withdrawal rates for example comes from a specific amortization. You could create such tables for any ABW strategy.

Software is useful for running Monte Carlo or historical simulations that can help establish what parameters match your preferences—i.e. what asset allocation and g to use in the strategy. But it’s not needed for simply calculating or looking up withdrawal rates.
silent cal wrote: Sat Aug 26, 2023 7:51 pm Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
The benefit of variable withdrawals benefit is robustness to planning assumptions. You keep adjusting your spending as the future reveals itself, so bad assumptions at the start of retirement do less damage. If the markets do worse than expected, your spending will decline but you won’t run out of money. That’s much better than relying on what you thought the market was likely to do at the start of retirement and worrying about whether you were right as the market drops.

And as described earlier, you can always reduce the risk of spending declines in ABW by choosing a high g for the amortization. That will start you off with lower spending at the start of retirement. Spending would then be more likely to go up than down over the course of retirement.
Total Portfolio Allocation and Withdrawal (TPAW)
ralph124cf
Posts: 2993
Joined: Tue Apr 01, 2014 11:41 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by ralph124cf »

Hi Parking.

While I fully understand the TLAR bombing strategy, I disagree with your example because it does not consider time.

The 4% "rule" is for a 30 year retirement, and therefore allows for a 100% depletion of all remaining funds during year 30. If I was spending 6% of funds in year 2, I might be concerned, but if it was in year 15, this might be right on schedule.
Florida Orange
Posts: 1200
Joined: Thu Jun 16, 2022 2:22 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Florida Orange »

The problem with the fixed percentage strategy is that if the percentage is low enough to be truly safe, it's very likely that you will die with a lot more money than you had when you retired. The guardrails strategy is an attempt to be able to spend more on discretionary items when times are good while still being sure that you have enough to cover the necessities when times are bad.

If dying with too much money is a concern, I think the guardrails strategy is a reasonable approach.
User avatar
dogagility
Posts: 3273
Joined: Fri Feb 24, 2017 5:41 am
Location: Del Boca Vista - Phase 3

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by dogagility »

Lots of discussion on Bogleheads about withdrawal strategies.

My conclusions from these threads are that few people actually use the 4% rule as a strategy. It's a potentially useful planning tool to determine readiness to retire, but I think planning/modeling retirement using software tools like TPAWPlanner and VPW retirement worksheet are much better.

I agree with Ben in his post above that variable withdrawal strategies (ABW and VPW are two) are better withdrawal strategy methods than the fixed percentage withdrawal method.
Make sure you check out my list of certifications. The list is short, and there aren't any. - Eric 0. from SMA
User avatar
nisiprius
Advisory Board
Posts: 52439
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by nisiprius »

What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
exodusing
Posts: 2287
Joined: Thu Oct 13, 2022 7:32 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by exodusing »

nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
How does one test "adjusted by seat-of-the-pants"? It lacks a certain degree of precision that seems a requirement of testing.
Broken Man 1999
Posts: 8645
Joined: Wed Apr 08, 2015 11:31 am
Location: West coast of Florida, near Champa Bay !

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Broken Man 1999 »

silent cal wrote: Sat Aug 26, 2023 7:51 pm Hello everyone,

I have been doing a deep dive into withdrawal strategies while I'm still a ways from retirement. I came across this article on Kitces.com https://www.kitces.com/blog/risk-based- ... n-hatchet/ describing a risk-based guardrails approach to portfolio withdrawals. It explains the downsides of a simple withdrawal-rate guardrails approach and the advantage of a probability-of-success guardrail strategy which factors in longevity risk, cash flows, income, and Monte Carlo results. There is even a new financial planning software to implement this exact method https://incomelaboratory.com.

I then also came across this article from Early Retirement Now https://earlyretirementnow.com/2017/02/ ... n-klinger/ which I thought laid out a pretty solid issue with variable withdrawal strategies in general.


My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).

Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.

Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
OP, if you are taking a deep dive on withdrawal methods, you might order a copy of Living Off Your Money: The Modern Mechanics of Investing During Retirement with Stock and Bonds by Michael H McClung.

The author does a deep dive on withdrawal strategies. That is probably an understatement.

Broken Man 1999
“If I cannot drink Bourbon and smoke cigars in Heaven then I shall not go." - Mark Twain
User avatar
nisiprius
Advisory Board
Posts: 52439
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by nisiprius »

exodusing wrote: Mon Aug 28, 2023 7:28 am
nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
How does one test "adjusted by seat-of-the-pants"? It lacks a certain degree of precision that seems a requirement of testing.
It might be more important to try to research how people behave in real life than to investigate whether how a guardrail approach would have worked on paper. In financial economics there seems to be a lot of losing-a-car-key-over-there-but-looking-for-it-over-here-where-the-light's-better.

How do you test the variable withdrawal system used by Vanguard in their Managed Payout fund--when they changed the calculation five years in, and then canned the product entirely eleven years after inception? We don't even have close to a full-retirements'-worth of data from that experiment.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

silent cal wrote: Sat Aug 26, 2023 7:51 pm Hello everyone,

I have been doing a deep dive into withdrawal strategies while I'm still a ways from retirement. I came across this article on Kitces.com https://www.kitces.com/blog/risk-based- ... n-hatchet/ describing a risk-based guardrails approach to portfolio withdrawals. It explains the downsides of a simple withdrawal-rate guardrails approach and the advantage of a probability-of-success guardrail strategy which factors in longevity risk, cash flows, income, and Monte Carlo results. There is even a new financial planning software to implement this exact method https://incomelaboratory.com.

I then also came across this article from Early Retirement Now https://earlyretirementnow.com/2017/02/ ... n-klinger/ which I thought laid out a pretty solid issue with variable withdrawal strategies in general.


My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).

Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.

Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
It’s a thorny problem and one of the approaches I really like to bolt-on is using a Funded Ratio to determine if I need to cut back or not.
viewtopic.php?t=392120

Essentially The Funded Ratio is a financial calculation that helps you understand the costs of your Retirement/Estate Plans in todays dollars and compares that to your current portfolio. I find this insight extremely helpful in making the decision to decrease or increase my spending and prefer it to a probability based guardrails approach.

Essentially, I won’t force a spending cut unless my Funded Ratio drops below 110%. Even then, I might shift my Estate Plans (we prioritize our Retirement Plans over the Estate Plans) to keep our Retirement well funded. There are only so many years I have left to pursue some of my passions in Retirement and I want to make the most of them.

Lastly, there are a number of tools to help build your own Monte Carlo plan, I use Flexible Retirement Planner.
https://www.flexibleretirementplanner.com/wp/

Personally, I find that using multiple approaches to help build and refine a strategy leads to better decision making and stronger signposts to monitor.

Thanks for the thread — it’s a fascinating topic.

WoodSpinner
WoodSpinner
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
Interesting question ….

Here are a couple of my thoughts:

1. VPW/ABW withdrawal strategies match my real world spending desires in Retirement and Estate Planning much better than 3-4% rule. I prioritize being able to spend and enjoy my passions while my body and mind still allows it,

2. VPW/ABW are financial, mathematical models and aren’t dependent on data mining from past performance. I am always a bit suspect of data mining the past since today (and the future) is such a different world.

3. I would not have the intestinal fortitude to continue to spend based on the 3-4% guidelines while my portfolio balances approach $0. It would freak me out….

Not sure this is evidence from your perspective and am looking forward to your perspective.

Lastly, I find that bolting-on a Funded Ratio Signpost to VPW/ABW (see previous post) gives me more confidence to ride out some of turbulence.

WoodSpinner
WoodSpinner
exodusing
Posts: 2287
Joined: Thu Oct 13, 2022 7:32 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by exodusing »

nisiprius wrote: Mon Aug 28, 2023 7:44 am
exodusing wrote: Mon Aug 28, 2023 7:28 am
nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
How does one test "adjusted by seat-of-the-pants"? It lacks a certain degree of precision that seems a requirement of testing.
It might be more important to try to research how people behave in real life than to investigate whether how a guardrail approach would have worked on paper. In financial economics there seems to be a lot of losing-a-car-key-over-there-but-looking-for-it-over-here-where-the-light's-better.

How do you test the variable withdrawal system used by Vanguard in their Managed Payout fund--when they changed the calculation five years in, and then canned the product entirely eleven years after inception? We don't even have close to a full-retirements'-worth of data from that experiment.
Seeing what people do in real life would be interesting. An issue is that backtesting would be hard to the extent it relies on recollection, which is notoriously faulty, and going forward would take a long time. I suspect people tend to adjust down in bad times and up in good times and that there aren't too many doctrinaire 4% SWR (or other formula) types, but that's just a guess.

Yes, the Managed Payout fund was a massive failure.

These are among the issues that lead me to recommend a TIPS ladder out to a good age to cover necessary expenses and investing the balance in a risk portfolio, perhaps rather equity heavy, now that a 30 year ladder will support a 4% government guaranteed, inflation (CPI) adjusted withdrawal rate, with a fair amount left over. A safe floor seems, well, safer than a probabilistic portfolio. https://www.tipsladder.com/build?income ... =BestYield
Parkinglotracer
Posts: 4057
Joined: Fri Dec 20, 2019 2:49 am
Location: Upstate NY

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Parkinglotracer »

ralph124cf wrote: Sun Aug 27, 2023 2:58 pm Hi Parking.

While I fully understand the TLAR bombing strategy, I disagree with your example because it does not consider time.

The 4% "rule" is for a 30 year retirement, and therefore allows for a 100% depletion of all remaining funds during year 30. If I was spending 6% of funds in year 2, I might be concerned, but if it was in year 15, this might be right on schedule.
Good point

yes indeed what looks about right (TLAR) at age 62 like I am now will and should look a lot different for TLAR for me at age 77; 15 years from now. One could reference a conservative IRS RMD schedule over time and keep an eye ball on their own health / life expectancy too as time goes. Easier calls to be made in a bull market than in a long bear market.
randomguy
Posts: 11313
Joined: Wed Sep 17, 2014 9:00 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by randomguy »

WoodSpinner wrote: Mon Aug 28, 2023 8:33 am
nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
Interesting question ….

Here are a couple of my thoughts:

1. VPW/ABW withdrawal strategies match my real world spending desires in Retirement and Estate Planning much better than 3-4% rule. I prioritize being able to spend and enjoy my passions while my body and mind still allows it,

2. VPW/ABW are financial, mathematical models and aren’t dependent on data mining from past performance. I am always a bit suspect of data mining the past since today (and the future) is such a different world.

3. I would not have the intestinal fortitude to continue to spend based on the 3-4% guidelines while my portfolio balances approach $0. It would freak me out….

Not sure this is evidence from your perspective and am looking forward to your perspective.

Lastly, I find that bolting-on a Funded Ratio Signpost to VPW/ABW (see previous post) gives me more confidence to ride out some of turbulence.

WoodSpinner
What are you spending desires? ABW is both more aggressive (in good times) and conservative ( in bad times) than something like the 4% rule. When I retire I don't want to live on 30k for 15 years and then live on 50k/year for 15 years. That is the exact opposite of what I want. I would much rather do 40k/year for 15 years and 30k for 15 years. It less money overall but the value of money when you are young is a lot higher than when you are old. Now in good times ABW gives you more money. That is good. But 10k more of income brings a lot less joy than the pain of 10k of less income. Making the worst case better is more valuable than making the best case better.

Obviously you can feel differently. You need to define what better means to you and pick a scheme that meets your needs. There are no free lunches. You have to pick off what tradeoffs you are happy with. The ABW/VPW crowd don't worry about underspending when young. They focus on the fear of being broke when old. The fixed spending goes the other way. Again it isn't better or worse. It is different.

And in the end you really hope you never hit a time when this matters. I don't care what scheme you are using but when 1966-1981 hits (or even 2000-9), it basically doesn't matter what scheme you use until you get to absurdly low SWR. Watching your portfolio drop 50% is going to be gut wretching. The difference between that and say 40% isn't going to change how well I sleep at night.
Grt2bOutdoors
Posts: 25646
Joined: Thu Apr 05, 2007 8:20 pm
Location: New York

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Grt2bOutdoors »

Parkinglotracer wrote: Sun Aug 27, 2023 7:19 am You sound like a very analytical person which is good because we need them. Here is a view from an mba educated not so analytical person. Don’t think I don’t appreciate your analysis.

To answer your question, I read the executive summary of kitces article and I don’t know how to explain it any better than he did. If I was dependent upon a regular withdrawal for my retirement survival I would have some sort of guard rail to cut my spending to prevent me from running out of money.

While an I am an engineering major with years experience in a strict defense contractor environment - I am more of a TLAR (that looks about right) decision maker from my AF flying experience. Coming up with an exact formula seems unnecessary - like trying to make jello stick to a wall as we used to say in the military.

I agree planners might use complicated strategies to make their services seem essential. I wouldn’t throw the strategy idea out with the bath water just because some complicate it. That is what they get paid to do - complicate it.

If I started withdrawing 4%; a year and my portfolio performance made that amount eventually 6% of my portfolio that would set off a grinding noise in my brain thinking my retirement bus was hitting a guard rail and get me to move to a lower spending level or return to work.

What practical guard rail do you think you would use if any? maybe a strategy that is simple enough to explain to a non technical spouse who wants to understand what the plan is?

Call it common sense or a guardrail - if your lump sum is disappearing one might want to not spend as much if they want to have any left in retirement.
Totally agree. However, common sense is not as common as we'd like to believe.
"One should invest based on their need, ability and willingness to take risk - Larry Swedroe" Asking Portfolio Questions
furnace
Posts: 357
Joined: Tue Oct 20, 2015 3:38 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by furnace »

Keeping your cash outlays as low as possible is probably the safest retirement strategy.
unbiased
Posts: 195
Joined: Sat Apr 25, 2020 5:46 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by unbiased »

To answer the OP's question, I think these guardrails are useful. I plan on using Guyton-Klinger as a way to "check-in" yearly on my portfolio's health. If I can spend more, I'd want to. If I need to spend less because the G-K rules are indicating it's time to conserve, I'd find that useful also.

I doubt any retiree would just blindly withdrawal 4%+inflation for decades while watching their portolio deplete if there is an extended market decline. However, I think planning for this possibility, whether by using a guardrail approach or an amortization schedule or some other means, is really necessary for one of the most consequential financial decisions we'll ever make.
User avatar
Ben Mathew
Posts: 2743
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Ben Mathew »

randomguy wrote: Fri Sep 01, 2023 8:01 pm ABW is both more aggressive (in good times) and conservative ( in bad times) than something like the 4% rule.
A more accurate statement is that ABW (with the right amortization parameters that fit your preferences) does not underspend during good times or overspend during bad times. During good times and bad times, ABW is neither aggressive nor conservative, but withdraws from the portfolio in keeping with what's in it and how you expect it to grow. Yes, this happens to be more if the market did well and less if the market did badly, but to call this conservative vs aggressive is misleading. The same risk preferences would lead to withdrawing more if the portfolio did well or expected returns went up, and less if the portfolio did badly or expected returns went down.

SWR, on the other hand, withdraws the same amount regardless of how the portfolio did. This is conservative if the portfolio did well and aggressive if the portfolio did badly.
randomguy wrote: Fri Sep 01, 2023 8:01 pm When I retire I don't want to live on 30k for 15 years and then live on 50k/year for 15 years. That is the exact opposite of what I want. I would much rather do 40k/year for 15 years and 30k for 15 years. It less money overall but the value of money when you are young is a lot higher than when you are old.
What you are describing here is a high subjective discount rate for consumption. This is incorporated in the basic lifecycle model as a parameter. You would simply amortize the portfolio with a high rate (low g in ABW). This gives you more in early retirement and less in late retirement.
randomguy wrote: Fri Sep 01, 2023 8:01 pm But 10k more of income brings a lot less joy than the pain of 10k of less income. Making the worst case better is more valuable than making the best case better.
You are describing decreasing marginal utility from consumption. This again is a standard assumption in economics and is fully incorporated in the lifecycle model. In fact, risk aversion comes from this assumption. If there were no diminishing marginal utility, we would be risk neutral.

Diminishing marginal utility is precisely why the lifecycle model leads to ABW and not SWR. SWR eschews small cuts during bad times (a small reduction in consumption) and instead takes on the risk of running out of money (a large reduction in consumption). This is not consistent with diminishing marginal utility. In contrast, ABW accepts small cuts when the market does badly, so it does not have to cut as severely as SWR if the market continues to do badly. This is consistent with diminishing marginal utility.
Last edited by Ben Mathew on Sat Sep 02, 2023 11:41 am, edited 1 time in total.
Total Portfolio Allocation and Withdrawal (TPAW)
randomguy
Posts: 11313
Joined: Wed Sep 17, 2014 9:00 am

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by randomguy »

Ben Mathew wrote: Fri Sep 01, 2023 9:36 pm
randomguy wrote: Fri Sep 01, 2023 8:01 pm ABW is both more aggressive (in good times) and conservative ( in bad times) than something like the 4% rule.
A more accurate statement is that ABW (with the right amortization parameters that fit your preferences) does not underspend during good times or overspend during bad times. During good times and bad times, SWR is neither aggressive nor conservative, but withdraws from the portfolio in keeping with what's in it and how you expect it to grow. Yes, this happens to be more if the market did well and less if the market did badly, but to call this conservative vs aggressive is misleading. The same risk preferences would lead to withdrawing more if the portfolio did well or expected returns went up, and less if the portfolio did badly or expected returns went down.

SWR, on the other hand, withdraws the same amount regardless of how the portfolio did. This is conservative if the portfolio did well and aggressive if the portfolio did badly.
We are using different reference points for the correct amount of spending.
- You are going the formula says I can spend 30k, so that is the right number
- I am going you could have spend 38k and not gone broke, that is the right number

Feel free to correct me if I am wrong about that. Show some ABW scheme for 1966-1981 where it doesn't go too conservative and spends less than 3.8% of the starting value.
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

randomguy wrote: Fri Sep 01, 2023 8:01 pm
WoodSpinner wrote: Mon Aug 28, 2023 8:33 am
nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
Interesting question ….

Here are a couple of my thoughts:

1. VPW/ABW withdrawal strategies match my real world spending desires in Retirement and Estate Planning much better than 3-4% rule. I prioritize being able to spend and enjoy my passions while my body and mind still allows it,

2. VPW/ABW are financial, mathematical models and aren’t dependent on data mining from past performance. I am always a bit suspect of data mining the past since today (and the future) is such a different world.

3. I would not have the intestinal fortitude to continue to spend based on the 3-4% guidelines while my portfolio balances approach $0. It would freak me out….

Not sure this is evidence from your perspective and am looking forward to your perspective.

Lastly, I find that bolting-on a Funded Ratio Signpost to VPW/ABW (see previous post) gives me more confidence to ride out some of turbulence.

WoodSpinner
What are you spending desires? ABW is both more aggressive (in good times) and conservative ( in bad times) than something like the 4% rule. When I retire I don't want to live on 30k for 15 years and then live on 50k/year for 15 years. That is the exact opposite of what I want. I would much rather do 40k/year for 15 years and 30k for 15 years. It less money overall but the value of money when you are young is a lot higher than when you are old. Now in good times ABW gives you more money. That is good. But 10k more of income brings a lot less joy than the pain of 10k of less income. Making the worst case better is more valuable than making the best case better.

Obviously you can feel differently. You need to define what better means to you and pick a scheme that meets your needs. There are no free lunches. You have to pick off what tradeoffs you are happy with. The ABW/VPW crowd don't worry about underspending when young. They focus on the fear of being broke when old. The fixed spending goes the other way. Again it isn't better or worse. It is different.

And in the end you really hope you never hit a time when this matters. I don't care what scheme you are using but when 1966-1981 hits (or even 2000-9), it basically doesn't matter what scheme you use until you get to absurdly low SWR. Watching your portfolio drop 50% is going to be gut wretching. The difference between that and say 40% isn't going to change how well I sleep at night.
Based on our modeling we are withdrawing significantly more from the portfolio before I hit 70 (5-6% withdraw rate) and much less after SS (2% Withdrawal Rate). Lastly we are spending more than the 4% Rule of Thumb but less than the ABW calculations. More importantly, building a Bottoms-up portfolio that includes all of our Retirement Plans, Estate Goals and Charitable desires shows a Funded Ratio of a 1.3.

For us having a plan and metrics to use for our Cashflow planning is key to navigating serious volatility in the market. You are right, it’s no-fun to go through!

I think we are on the right track and haven Had to adjust our spending down so far.

WoodSpinner
Last edited by WoodSpinner on Sat Sep 02, 2023 12:14 pm, edited 1 time in total.
WoodSpinner
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

furnace wrote: Fri Sep 01, 2023 9:10 pm Keeping your cash outlays as low as possible is probably the safest retirement strategy.
Are you concerned that this approach might leave a lot of your Hopes, Dreams and Ambitions in Retirement unfulfilled….

WoodSpinner
WoodSpinner
L84SUPR
Posts: 343
Joined: Sat Feb 01, 2014 12:28 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by L84SUPR »

Much of what Kitces and associates write appears to me to be leading edge beyond standard practice in their industry. I read an artle of theirs about rising equity glide paths and was surprised how little difference it made. They are self proclaimed geeks and passionate about their profession. I enjoy reading some of the articles but find I am drifting inextricably toward very simple approaches such as 1/N.
1/3rd VTWAX, 1/3rd Wellington, 1/3rd G Fund | All models are wrong. Some are useful.
furnace
Posts: 357
Joined: Tue Oct 20, 2015 3:38 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by furnace »

WoodSpinner wrote: Sat Sep 02, 2023 12:11 pm
furnace wrote: Fri Sep 01, 2023 9:10 pm Keeping your cash outlays as low as possible is probably the safest retirement strategy.
Are you concerned that this approach might leave a lot of your Hopes, Dreams and Ambitions in Retirement unfulfilled….

WoodSpinner
Not at all. I won't save all the good stuff for retirement; I will try to do them prior to retirement. Good health and old age are a flaky pairing.
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

silent cal wrote: Sat Aug 26, 2023 7:51 pm
Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
For us we plan to spend quite a bit more during the GoGo years of Retirement and definitely are not looking for consistency in spending.


That said, I think a true Variable Spending withdrawal plan using an Amortization based approach is a much better fit for our needs.


Have you looked at VPW, ABW, or TPAW in the WIKI?


WoodSpinner
WoodSpinner
User avatar
WoodSpinner
Posts: 3509
Joined: Mon Feb 27, 2017 12:15 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by WoodSpinner »

nisiprius wrote: Mon Aug 28, 2023 7:04 am What's the evidence that calculated variable withdrawal systems are better than 3-4% rule-of-thumb adjusted by seat-of-the-pants?
Assuming you are referring to ABW, VPW, TPAW approaches:

1. Doesn’t depend on data mining from historical returns
2. Mathematically sound
3. Automatically adjusts to actual market performance
4. Adjusts for smaller or larger withdrawals actual withdrawals
5. Accommodates longer Retirement timeframes
6. Much less risk of dying with too large a portfolio.

WoodSpinner
WoodSpinner
derekt
Posts: 1
Joined: Thu Jun 29, 2017 2:02 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by derekt »

I am one of the authors of the risk-based guardrails article you referenced, so I'm quite biased on this matter, but I just wanted to offer a few thoughts/clarifications based on your questions.
silent cal wrote: Sat Aug 26, 2023 7:51 pm I then also came across this article from Early Retirement Now https://earlyretirementnow.com/2017/02/ ... n-klinger/ which I thought laid out a pretty solid issue with variable withdrawal strategies in general.
I wouldn't equate risk-based guardrails with Guyton-Klinger's guardrails. GK is a distribution-rate-driven framework which has a number of issues such as not being able to account for the "retirement distribution hatchet" and not doing a good job of accounting for the ability to take larger percentage distributions from a portfolio over time as longevity decreases (in contrast to how a VPW-type approach will allow for increasing distribution rates over time). Risk-based guardrails do not face these same issues and, in my opinion, can account for total risk better than approaches such as ABW, VPW, etc. based on looking at a more complete profile of someone's financial resources.

In fact, we recently published a critique of Guyton-Klinger's guardrails with some historical simulation that demonstrates the extreme volatility of Guyton-Klinger guardrails highlighted by Early Retirement Now, while also demonstrating that risk-based guardrails provide much more stable income in retirement. What's driving this is largely an overreaction to market downturns by the GK approach.
silent cal wrote: Sat Aug 26, 2023 7:51 pm My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).
I would say that yes, risk-based guardrails (but not Guyton-Klinger guardrails) are desirable. They provide a mechanism for both increasing and decreasing spending to keep a plan on track, but the use of appropriate guardrails parameters will keep income far more stable. Also, I would disagree with the notion that RBGs would imply a high likelihood of needing to cut below 4% at some point in the future. This does depend heavily on the guardrails parameters you select, but distribution rates (e.g., 4%) are the wrong way to think about sustainable retirement spending to begin with. As the retirement distribution hatchet illustrates, it is not uncommon to see distribution rates of 8%+ for the first few years of a plan and then fall to 2-3% thereafter when you account for deferring Social Security, etc. With RBGs you are accounting for all of the unique factors of a given individual's situation and not forcing some potentially inaccurate assumptions on a plan as you often would with ABW, VPW, etc.
silent cal wrote: Sat Aug 26, 2023 7:51 pm Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.
I don't know that advocating a higher withdrawal rate really makes most advisory clients happier. In my experience, the types of people who have tendencies that lead them to be good savers in the first place, often need a lot of pushing to spend at rates they are capable of. I've seen people in their 70s with 7-figure portfolios that are living entirely off of Social Security and terrified to touch their portfolios. In my opinion, too many people who weren't explicitly aiming to leave a large portfolio end up doing so. If that is the goal, then great, but I would like to see more people enjoying their wealth earlier when their health allows them to do so.

I think it is true that there's no way to implement RBGs without financial planning software. However, there are free Monte Carlo tools available and anyone could implement a RBG framework without sophisticated planning tools. In my opinion, what the more sophisticated tools really add at this point is the ability to stress test a given set of guardrails plan parameters, etc. That is really hard to do that historical simulation (or other forms of stress testing) without tools built to do that, but someone could take parameters like we laid out in the GK critique article and implement those with any MC software. For instance, what we used in that article was:
  • Target initial spending at a PoS = 80%
  • Increase spending if PoS rises to 100%
  • Decrease spending if PoS falls to 25%
I don't think that is actually all that complicated for most people to follow, and it provides the benefits of RBGs and a better overall assessment of total risk than frameworks like ABW and VPW.
User avatar
Ben Mathew
Posts: 2743
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by Ben Mathew »

derekt wrote: Tue Apr 02, 2024 1:55 pm I am one of the authors of the risk-based guardrails article you referenced, so I'm quite biased on this matter, but I just wanted to offer a few thoughts/clarifications based on your questions.
I was going to say welcome to the forum, but looks like you've been here longer than me. :D

Anyways, thanks for sharing your thoughts on this topic.
derekt wrote: Tue Apr 02, 2024 1:55 pm I wouldn't equate risk-based guardrails with Guyton-Klinger's guardrails. GK is a distribution-rate-driven framework which has a number of issues such as not being able to account for the "retirement distribution hatchet" and not doing a good job of accounting for the ability to take larger percentage distributions from a portfolio over time as longevity decreases (in contrast to how a VPW-type approach will allow for increasing distribution rates over time). Risk-based guardrails do not face these same issues and, in my opinion, can account for total risk better than approaches such as ABW, VPW, etc. based on looking at a more complete profile of someone's financial resources.
[...]

With RBGs you are accounting for all of the unique factors of a given individual's situation and not forcing some potentially inaccurate assumptions on a plan as you often would with ABW, VPW, etc.

[...]
I don't think that is actually all that complicated for most people to follow, and it provides the benefits of RBGs and a better overall assessment of total risk than frameworks like ABW and VPW.
Can you expand on how the risk-based guardrails you favor account for total risk better than Amortization Based Withdrawals (ABW)?
Total Portfolio Allocation and Withdrawal (TPAW)
User avatar
nisiprius
Advisory Board
Posts: 52439
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by nisiprius »

Absent widely-available, cheap, automated ways for implementing these strategies, it all seems kind of sterile to me.

It's amazing and depressing how little is available.

The Vanguard Managed Payout Fund was tied to a specific portfolio and a specific withdrawal system--and is defunct.

I haven't looked carefully at Schwab Smart Income, but I believe it, too, is tied to a choice of Schwab-defined portfolios, and their own withdrawal system.

I don't believe any brokerage has the tools to automate even the 4% rule, as none of them have any provisions for tying automatic withdrawals to the CPI.

At the moment, the only practical automatic withdrawal systems seem to be:

1) Automated RMD services, and "spend the RMD."

2) Choose a portfolio of assets, including stock funds, bond funds, and individual stocks and bonds, and "spend the dividends."

Quite apart from theoretical objections, even "spend the dividends" is problematical--I've been poking into this and have been surprised this hasn't been commented on much--the quarterly distributions from mutual funds are seriously unequal and irregular, so much so as to be a budgeting problem if you aren't holding a couple of years' in cash as a buffer. This is true even if you spend the "dividends" and reinvest the "capital gains distribution."
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
dcabler
Posts: 4677
Joined: Wed Feb 19, 2014 10:30 am
Location: TX

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by dcabler »

silent cal wrote: Sat Aug 26, 2023 7:51 pm Hello everyone,

I have been doing a deep dive into withdrawal strategies while I'm still a ways from retirement. I came across this article on Kitces.com https://www.kitces.com/blog/risk-based- ... n-hatchet/ describing a risk-based guardrails approach to portfolio withdrawals. It explains the downsides of a simple withdrawal-rate guardrails approach and the advantage of a probability-of-success guardrail strategy which factors in longevity risk, cash flows, income, and Monte Carlo results. There is even a new financial planning software to implement this exact method https://incomelaboratory.com.

I then also came across this article from Early Retirement Now https://earlyretirementnow.com/2017/02/ ... n-klinger/ which I thought laid out a pretty solid issue with variable withdrawal strategies in general.


My question is basically: Are guardrail-based withdrawal strategies even desirable? It seems to me that the advantage is a potentially higher starting withdrawal rate, but as the Early Retirement Now article explains if 4% is the historic safe withdrawal rate, and you start at higher rate, you are likely going to have to cut spending to below 4% at some point in the future (obviously SWRs vary but his point stands).

Part of me wonders if the reason financial advisors seem to be adopting different guardrail strategies, especially probability of success guardrails like in the Kitces article above, is because a) it allows them to provide a data-driven argument for a higher withdrawal rate which makes clients happier (since they can spend more) and/or b) it potentially makes distribution planning seem too complicated for someone to implement on their own. I have no idea how you implement the risk-based guardrails approach without financial planing software for example.

Am I missing major benefits to variable withdrawal strategies? It seems a constant-withdrawal strategy with a SWR <4% might run the risk of you dying with a large portfolio, but reduces the risk you have to make dramatic cuts to spending throughout retirement if the market turns south? Isn't having the ability to maintain a consistent level of spending in retirement desirable?
I start from a position that anything that is based on SWR is probably not the best solution. Guardrails, Guyton-Klinger, Kitces Ratcheting methods, etc. are all hacks to SWR to try and improve it. And they all like will improve it. But because the future is unknown, starting with a method that considers only the past still leaves you with a possibility of prematurely running out of money.

I suggest that you look at several variable withdrawal methods on this forum which incorporate amortization in the withdrawal calculations: ABW, TPAW and VPW. With these methods, your money will last as long as you plan for it to last, but your withdrawals are going to vary from cycle to cycle. So instead of the risk of completely running out of money prematurely, you've now shifted the risk to any single withdrawal being too small for your expenses which is something that I know I'd rather try to manage vs. having nothing left at all. There are a number of ways to reduce that risk. Also, the math for this method (at least for ABW and TPAW) is very straightforward - It just uses the excel PMT equation to calculate the withdrawal. No if/then/else algorithmic stuff..

Cheers.
lazynovice
Posts: 3405
Joined: Mon Apr 16, 2012 10:48 pm
Location: Denver area. Former Texan.

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by lazynovice »

derekt wrote: Tue Apr 02, 2024 1:55 pm I am one of the authors of the risk-based guardrails article you referenced, so I'm quite biased on this matter, but I just wanted to offer a few thoughts/clarifications based on your questions….
Thanks for posting. I’ve watched several of your webinars on Risk Based Guardrails. I think the approach is very interesting. It’s interesting to hear the suggestions to advisors on adjusting the initial withdrawal rate if a client is like me- negative on dropping below a spending floor.
rustwood
Posts: 204
Joined: Mon Jan 17, 2022 2:05 pm

Re: Are risk-based guardrails approach to retirement withdrawals even desirable?

Post by rustwood »

derekt wrote: Tue Apr 02, 2024 1:55 pm In my experience, the types of people who have tendencies that lead them to be good savers in the first place, often need a lot of pushing to spend at rates they are capable of.
I believe this is 100% accurate for me, which is why I was very intrigued when I stumbled across your recent article comparing Guyton-Klinger guardrails to risk-based guardrails. I thought it was very thought provoking, but I didn't recall seeing anything about it here so I searched and found your comment. Thank you for taking the time to post it.

I am well aware that my planning is very conservative. I use very conservative assumptions in Maxifi Planner and our smooth consumption plan still shows that we could/should spend far more than we do. I've never been very interested in Monte Carlo retirement analyses before, but after reading your article I am curious to see how an initial withdrawal rate with an 80% chance of success compares to our conservative smooth consumption rate. Given the historical performance of risk-based guardrails shown in your article, seeing our 80% success withdrawal rate might help me get more comfortable with a higher spending rate.

FWIW, we are still in our 50's and started a phased retirement 5 years ago. I am now ~50% retired (largely for non-financial reasons) and my younger spouse fully retired 2 years ago. I hope we both still have long retirements ahead of us, so I am happy to err on the conservative side in these early years. With that said, I am trying to ramp up our spending as we approach starting Medicare and taking Social Security. I hope by the time we start taking SS we will have settled into a relatively stable and very comfortable spending rate.
Post Reply