Kitces article on lower prob of retirement success ok

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Exchme
Posts: 134
Joined: Sun Sep 06, 2020 3:00 pm

Re: Kitces article on lower prob of retirement success ok

Post by Exchme »

willthrill81 wrote: Sun Jan 10, 2021 9:56 pm
WoodSpinner wrote: Sun Jan 10, 2021 7:29 pm
Dottie57 wrote: Sun Jan 10, 2021 7:04 pm
Marseille07 wrote: Sun Jan 10, 2021 7:00 pm
Normchad wrote: Sun Jan 10, 2021 6:56 pm His articles are great. I was just reading that one today. The main point is also echoed frequently in here. Failure doesn’t have to mean you die broke and starving in an alley. Be flexible and adjust as you go. Or more to the point, make sure you have a plan that allows for flexibility if things go sideways.

For me, I’m seriously looking at the VPW spreadsheet to guide my spending levels into retirement. It will adjust annually with market changes, and looks like something that will work well for us.
VPW is good if you don't care to leave anything on the table. My concern is, VPW will run into a death spiral where your capital depletes so much to the point that you have to withdraw a high percentage of your portfolio just to survive; and there's no turning back at that point as you can't rebuild the portfolio.
My concern too.
Why do you say that? VPW or ABW will not go into a Death Spiral under any circumstances if you use these tools as intended and update at least yearly. What may happen though is the suggested spending may be lower than what you hoped for — if you don’t adjust then you could be in trouble.

WoodSpinner
Correct. Also, there is no rule saying that you must withdraw what VPW or ABW says that you can. You can always withdraw less.
I thought the two points mentioned above undermined the usefulness of the study. Here, the modelers must have put their heads down and blinders on and continued to plan for 30 more years throughout life or they could not have ended up with an average of 60% of starting assets still in place at the end for the 95% confidence case. At least after the first 10-15 years SORR risk window, real world advisors for the 95% confidence case would not have asked for the severe cut backs that this method required.

The other point mentioned above, anyone capable of the discipline to make the cutbacks in bad times would not spend all the excess in the good times. Humans don't work that way.

So they asked an interesting question that I will rephrase as - "would continuous updating of the plan and spending correct for an initial high withdrawal rate?" The answer is yes, any plan that is updated regularly for new facts will probably work out. But to avoid complexities and confounding factors, they had to do the analysis in a very artificial way that makes it not very useful in the real world.
jj45
Posts: 93
Joined: Tue Jan 02, 2018 4:02 pm

Re: Kitces article on lower prob of retirement success ok

Post by jj45 »

This is a terrible article. The correct conclusion of the Monte Carlo experiment is that you can follow this strategy if you are willing to deplete your portfolio and live on your pension and social security alone. But if you don't need your portfolio why bother with all the calculations? Just do whatever you want with it, you don't need it.

They hide this conclusion by how they set up and talk about the experiment. They use 95% safe withdrawal rate and their "new" method is to recalculate the 95% rate every month. Bogleheads who have been reading this board know what will happen. We know sequence of returns risk is the cause of failure and we know the risk is in the early years of retirement. 95% safe means you have a 5% chance of retiring at a time when you will be hit by a bad sequence of returns. Recalculating the withdrawal rate resets the clock on sequence of return risk. Every year is now an early year and you ratchet yourself into a bad sequence of returns.

Then they hide the results. The original 95% safe spending is $6769/mo but most of that, 52%, is from $3500 in social security. They graph the minimum spending but don't show the distribution. Looking at the graph, the minimum spending is mostly under $6k, and, by my eye, under $5k more than half the time. They do state the minimum is $3926, which they call a 42% reduction. This is only such a small reduction because of the assumed social security. They never mention what actually happens to portfolio withdrawals. The initial portfolio withdrawal is $3296 and the minimum is mostly then under $2500, a cut of 24%, more than half the time under $1500, a cut of 54%, and the minimum minimum portfolio withdrawal is $426, a cut of 87%. No thanks, not interested in an 87% cut in my portfolio retirement income.

This is using a 95% withdrawal rate. They then discuss reducing the safety and seem surprised that the minimum doesn't change much. How can it change much when it is already close to only taking social security.

The issue is not the safety percentage but the recalculation. There are umpteen Boglehead threads on the impact of recalculation. Recalculating increases risk.
User avatar
corn18
Posts: 1859
Joined: Fri May 22, 2015 6:24 am

Re: Kitces article on lower prob of retirement success ok

Post by corn18 »

Yeah me! We are planning to retire @ 96% Ps.
Don't do something, just stand there!
esteen
Posts: 123
Joined: Thu May 23, 2019 12:31 am

Re: Kitces article on lower prob of retirement success ok

Post by esteen »

Regardless of the rest of the article, from a communication perspective I appreciate how he stressed that Monte Carlo simulations' probability of "success" doesn't mean the other side of the coin is abject failure/living in squalor. More often, it means adjustments. This can be relatively minor adjustments if you are tracking your spending and depleting nest egg regularly. Since this blog is geared toward investment advisors, I think that is an important concept for them to convey to clients (without going so rosy as to endorse a plan with 50% success rate).
Escapevelocity
Posts: 318
Joined: Mon Feb 18, 2019 8:32 am

Re: Kitces article on lower prob of retirement success ok

Post by Escapevelocity »

esteen wrote: Mon Jan 11, 2021 4:06 pm Regardless of the rest of the article, from a communication perspective I appreciate how he stressed that Monte Carlo simulations' probability of "success" doesn't mean the other side of the coin is abject failure/living in squalor. More often, it means adjustments. This can be relatively minor adjustments if you are tracking your spending and depleting nest egg regularly. Since this blog is geared toward investment advisors, I think that is an important concept for them to convey to clients (without going so rosy as to endorse a plan with 50% success rate).
You're right. The key for retirees is to understand how much of their expenses are purely discretionary (vacations, gifts, toys, etc.) and then bake that into their modeling.
Reamus294
Posts: 108
Joined: Wed Jul 24, 2019 8:54 am

Re: Kitces article on lower prob of retirement success ok

Post by Reamus294 »

Knowing I am not the target audience for that blog, it made me think more about how important variability of withdraws can be and you don't always need a 95% success rate. I have heard the information here before but an in-depth analysis with examples that are well explained, as in the article, helps me understand the concept more.
marcopolo
Posts: 3854
Joined: Sat Dec 03, 2016 10:22 am

Re: Kitces article on lower prob of retirement success ok

Post by marcopolo »

Escapevelocity wrote: Mon Jan 11, 2021 5:01 pm
esteen wrote: Mon Jan 11, 2021 4:06 pm Regardless of the rest of the article, from a communication perspective I appreciate how he stressed that Monte Carlo simulations' probability of "success" doesn't mean the other side of the coin is abject failure/living in squalor. More often, it means adjustments. This can be relatively minor adjustments if you are tracking your spending and depleting nest egg regularly. Since this blog is geared toward investment advisors, I think that is an important concept for them to convey to clients (without going so rosy as to endorse a plan with 50% success rate).
You're right. The key for retirees is to understand how much of their expenses are purely discretionary (vacations, gifts, toys, etc.) and then bake that into their modeling.
Seems the biggest discretionary expense in the article was the 1.2% AUM fee. Get rid of that and the retiree could likely keep the vacations, gifts, toys, etc.
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
Horton
Posts: 860
Joined: Mon Jan 21, 2008 3:53 pm

Re: Kitces article on lower prob of retirement success ok

Post by Horton »

Ken Steiner recently wrote an article comparing and contrasting the article from the OP with the “Actuarial Approach” (essentially amortization based withdrawals).

http://howmuchcaniaffordtospendinretire ... l.html?m=1

Some take aways:
In summary, their model is less conservative (more aggressive) than the ABC with default assumptions, in that it produces higher initial total spending budgets.
Spending budgets determined under either the Kitces’ model or the Actuarial Approach are self-correcting as actual future experience emerges. Therefore, as noted in the Background section above, if spending under the Actuarial Approach is lower initially than spending under the Kitces’ model, it will catch up in time and eventually exceed the Kitces’ spending budget if assets are invested similarly, (which we have assumed).
Therefore, you have to decide how much risk you are willing to take to fund your future expenses. And the more risk you are willing to take, the more you need to be willing to reduce future expenses when necessary.
Whether you use a budget prepared by your financial advisor, you use one of our workbooks or you use some other approach, we suggest that you compare your final spending budget with the Actuarial Budget Benchmark determined with default assumptions so that you can gauge how conservative or how aggressive your budget may be and how much risk you (and not your financial advisor) are assuming with your spending plan.
User avatar
willthrill81
Posts: 22562
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Kitces article on lower prob of retirement success ok

Post by willthrill81 »

Escapevelocity wrote: Mon Jan 11, 2021 5:01 pm
esteen wrote: Mon Jan 11, 2021 4:06 pm Regardless of the rest of the article, from a communication perspective I appreciate how he stressed that Monte Carlo simulations' probability of "success" doesn't mean the other side of the coin is abject failure/living in squalor. More often, it means adjustments. This can be relatively minor adjustments if you are tracking your spending and depleting nest egg regularly. Since this blog is geared toward investment advisors, I think that is an important concept for them to convey to clients (without going so rosy as to endorse a plan with 50% success rate).
You're right. The key for retirees is to understand how much of their expenses are purely discretionary (vacations, gifts, toys, etc.) and then bake that into their modeling.
I too think that the essential/discretionary split is an important one that doesn't get enough attention among retirees. IMHO, the relative proportion of one's expenses that are essential may have a substantial impact on both a retiree's investment strategy and withdrawal strategy. In general, the smaller that proportion, the more flexible the withdrawals become, the less impact that sequence of returns risk has, the more volatility that can be safely taken on, and the more income that the retiree can likely take over the long-term from the portfolio.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
AlohaJoe
Posts: 5655
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Kitces article on lower prob of retirement success ok

Post by AlohaJoe »

willthrill81 wrote: Mon Jan 11, 2021 9:50 pm I too think that the essential/discretionary split is an important one that doesn't get enough attention among retirees.
I would rephrase this. (I'm retired.) The split is well understood among retirees. They live with it every day. Especially after events like this year.

But it isn't worth wasting breath trying to debate/discuss/inform pre-retirees who are 110% convinced that they have perfectly planned out the next 35 years of their spending and the slightest deviation from that spending plan is, literally, a fate worse than death. "You mean there's a 1% chance I might have take up a free hobby and cancel my Netflix, HBO Max, Hulu, Spotify, and cable subscriptions? I'd rather work another 2 years just in case."
User avatar
corn18
Posts: 1859
Joined: Fri May 22, 2015 6:24 am

Re: Kitces article on lower prob of retirement success ok

Post by corn18 »

AlohaJoe wrote: Mon Jan 11, 2021 10:05 pm
willthrill81 wrote: Mon Jan 11, 2021 9:50 pm I too think that the essential/discretionary split is an important one that doesn't get enough attention among retirees.
I would rephrase this. (I'm retired.) The split is well understood among retirees. They live with it every day. Especially after events like this year.

But it isn't worth wasting breath trying to debate/discuss/inform pre-retirees who are 110% convinced that they have perfectly planned out the next 35 years of their spending and the slightest deviation from that spending plan is, literally, a fate worse than death. "You mean there's a 1% chance I might have take up a free hobby and cancel my Netflix, HBO Max, Hulu, Spotify, and cable subscriptions? I'd rather work another 2 years just in case."
You nailed me to a tee. Except I am moderately terrified of the whole prospect of no income. Planning to retire in Mar and I just went into my spreadsheet and zeroed out all my income and savings. A little poo came out just then. Lots of red numbers in parenthesis.
Don't do something, just stand there!
User avatar
willthrill81
Posts: 22562
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Kitces article on lower prob of retirement success ok

Post by willthrill81 »

AlohaJoe wrote: Mon Jan 11, 2021 10:05 pm
willthrill81 wrote: Mon Jan 11, 2021 9:50 pm I too think that the essential/discretionary split is an important one that doesn't get enough attention among retirees.
I would rephrase this. (I'm retired.) The split is well understood among retirees. They live with it every day. Especially after events like this year.

But it isn't worth wasting breath trying to debate/discuss/inform pre-retirees who are 110% convinced that they have perfectly planned out the next 35 years of their spending and the slightest deviation from that spending plan is, literally, a fate worse than death. "You mean there's a 1% chance I might have take up a free hobby and cancel my Netflix, HBO Max, Hulu, Spotify, and cable subscriptions? I'd rather work another 2 years just in case."
I blame that sort of behavior on the idea that one cannot be happy and content without spending all of the money that one currently does and preferably more.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
DaufuskieNate
Posts: 424
Joined: Wed May 28, 2014 11:53 am

Re: Kitces article on lower prob of retirement success ok

Post by DaufuskieNate »

I view discretionary expenses in two different categories. First are the expenses that I can cut out tomorrow. This is the first line of defense and frankly it's something I've exercised off and on my entire life. Spend a little more in good times, a little less when money isn't as plentiful.

The second category involves more significant lifestyle changes that take some time to implement. This is the second line of defense that I think about as Plan B. For me, this would entail tapping into the substantial equity in my house, downsizing and moving to a LCOL area. I have no current plans to pull the trigger on Plan B, but I recognize that life throws curve balls and we need to be flexible and adjust.

Having seen others wait too long to implement Plan B, I am most interested in understanding how to see the signs early enough to soften the blow.
MikeG62
Posts: 3424
Joined: Tue Nov 15, 2016 3:20 pm
Location: New Jersey

Re: Kitces article on lower prob of retirement success ok

Post by MikeG62 »

Random Walker wrote: Sun Jan 10, 2021 10:09 pm
Dottie57 wrote: Sun Jan 10, 2021 9:19 pm I think I will probably need MORE money when older - For helpers, medicine, other healthcare.
I tend to strongly agree with you. My advisor consistently tells me that it generally doesn’t work out that way. The costs of healthcare and helpers is offset by lack of other expenses. Perplexes me a bit.

Dave
I suspect this is true for those whose retirement expenses includes a large amount of discretionary spending. Add in as well that if health care spending is significant (such as for someone in an assisted living facility), there may be a meaningful tax benefit from the medical expense Federal income tax deduction (current rules).
Real Knowledge Comes Only From Experience
MikeG62
Posts: 3424
Joined: Tue Nov 15, 2016 3:20 pm
Location: New Jersey

Re: Kitces article on lower prob of retirement success ok

Post by MikeG62 »

59Gibson wrote: Mon Jan 11, 2021 12:53 pm I'm just curious how an article like this can be reconciled with what is frequently seen on this board of 1.5-2.5%SWR for sometimes less than 30 years. Seems like a huge disconnect.
People on this board (myself included) are ridiculously conservative. This is very likely to result in the passing on of a large % of their current assets to their heirs to protect against the very small risk of running out of money. What Tharp is suggesting is an option for folks to start out with a higher initial withdrawal rate (for discretionary spend) and enjoy more of their money, especially in the go-go early years of retirement.

I think it was an excellent article which opens up some possibilities that people may not have thought possible or reasonable.
Real Knowledge Comes Only From Experience
esteen
Posts: 123
Joined: Thu May 23, 2019 12:31 am

Re: Kitces article on lower prob of retirement success ok

Post by esteen »

Random Walker wrote: Sun Jan 10, 2021 10:09 pm
Dottie57 wrote: Sun Jan 10, 2021 9:19 pm I think I will probably need MORE money when older - For helpers, medicine, other healthcare.
I tend to strongly agree with you. My advisor consistently tells me that it generally doesn’t work out that way. The costs of healthcare and helpers is offset by lack of other expenses. Perplexes me a bit.

Dave
I think key word is "generally". It generally does, but hopefully your advisor is not advising you based on generalities but based on your specific situation: your current spending habits, goals in retirement, life/family obligations, etc.
esteen
Posts: 123
Joined: Thu May 23, 2019 12:31 am

Re: Kitces article on lower prob of retirement success ok

Post by esteen »

willthrill81 wrote: Mon Jan 11, 2021 11:07 pm
AlohaJoe wrote: Mon Jan 11, 2021 10:05 pm
willthrill81 wrote: Mon Jan 11, 2021 9:50 pm I too think that the essential/discretionary split is an important one that doesn't get enough attention among retirees.
I would rephrase this. (I'm retired.) The split is well understood among retirees. They live with it every day. Especially after events like this year.

But it isn't worth wasting breath trying to debate/discuss/inform pre-retirees who are 110% convinced that they have perfectly planned out the next 35 years of their spending and the slightest deviation from that spending plan is, literally, a fate worse than death. "You mean there's a 1% chance I might have take up a free hobby and cancel my Netflix, HBO Max, Hulu, Spotify, and cable subscriptions? I'd rather work another 2 years just in case."
I blame that sort of behavior on the idea that one cannot be happy and content without spending all of the money that one currently does and preferably more.
Agree. I think for a lot of us humans it's against our nature to be comfortable with the unknown. That's why marketing like "what's your number?" works so well. It's an easier pill to swallow, even if it's not very accurate. But once we mix our basic planning with a level of acceptance for the unknown we get comfortable with flexibility. Those who remain inflexible usually make themselves work for far longer than they need to.

I'm by nature a numbers person who likes to solve equations. So although I espouse "nobody knows nothing", I am still working on getting fully comfortable with the unknowable future.

-es
User avatar
willthrill81
Posts: 22562
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Kitces article on lower prob of retirement success ok

Post by willthrill81 »

MikeG62 wrote: Tue Jan 12, 2021 10:00 am
59Gibson wrote: Mon Jan 11, 2021 12:53 pm I'm just curious how an article like this can be reconciled with what is frequently seen on this board of 1.5-2.5%SWR for sometimes less than 30 years. Seems like a huge disconnect.
People on this board (myself included) are ridiculously conservative. This is very likely to result in the passing on of a large % of their current assets to their heirs to protect against the very small risk of running out of money. What Tharp is suggesting is an option for folks to start out with a higher initial withdrawal rate (for discretionary spend) and enjoy more of their money, especially in the go-go early years of retirement.
I agree that those with comparatively substantial portfolios, which is certainly true for most BHs, can very likely plan on withdrawing more when they are younger and less when they are older, which coincides with retirees' actual spending behavior. The overwhelming majority of people are more energetic and interested in buying experiences at age 60 than they are at age 80, for instance. Relatively few 80 year olds are still doing much globetrotting, spending much on vehicles, buying high-end clothing, etc. The data we have are clear that total spending tends to declines with age, even though healthcare spending tends to increases with age. Some counter that with saying that a general tendency does not mean that it will hold true for any single person, which is true, but those who started with substantial portfolios are still very likely to have enough resources to cover their essential spending later in life, even if they have spent down a significant chunk of their portfolio along the way.

By comparison, those who retire with barely adequate retirement funding don't have as much flexibility and should possibly plan on level spending in retirement in order to have enough margin of safety.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
Rob1
Posts: 233
Joined: Thu May 04, 2017 2:57 pm

Re: Kitces article on lower prob of retirement success ok

Post by Rob1 »

Vanguard CEO Tim Buckley:

"...most Vanguard shareholders in retirement, they live below their means. You both have seen that time and time again. And when times get tough, they cut down...which they should do—but they do that too much. And they worry in the good times and are not spending. And you know what, maybe they could be living a little bit better."

Edit to include source:
Vanguard’s podcast ‘The Planner and the Geek’. Archived transcript here:
https://web.archive.org/web/20200421164 ... m-buckley/
Last edited by Rob1 on Tue Jan 12, 2021 1:21 pm, edited 1 time in total.
Exchme
Posts: 134
Joined: Sun Sep 06, 2020 3:00 pm

Re: Kitces article on lower prob of retirement success ok

Post by Exchme »

Horton wrote: Mon Jan 11, 2021 5:57 pm Ken Steiner recently wrote an article comparing and contrasting the article from the OP with the “Actuarial Approach” (essentially amortization based withdrawals).

http://howmuchcaniaffordtospendinretire ... l.html?m=1

Some take aways:
In summary, their model is less conservative (more aggressive) than the ABC with default assumptions, in that it produces higher initial total spending budgets.
Spending budgets determined under either the Kitces’ model or the Actuarial Approach are self-correcting as actual future experience emerges. Therefore, as noted in the Background section above, if spending under the Actuarial Approach is lower initially than spending under the Kitces’ model, it will catch up in time and eventually exceed the Kitces’ spending budget if assets are invested similarly, (which we have assumed).
Therefore, you have to decide how much risk you are willing to take to fund your future expenses. And the more risk you are willing to take, the more you need to be willing to reduce future expenses when necessary.
Whether you use a budget prepared by your financial advisor, you use one of our workbooks or you use some other approach, we suggest that you compare your final spending budget with the Actuarial Budget Benchmark determined with default assumptions so that you can gauge how conservative or how aggressive your budget may be and how much risk you (and not your financial advisor) are assuming with your spending plan.
Thank you for the pointer, the spreadsheets there are interesting and the actuarial method with updates would certainly be easy enough for DIY'ers to follow without the expensive specialists doing Monte Carlo that are implied in Kitces method. I do quibble with Steiner's default return assumption of 3% on assets, that seems awfully conservative, but it's selectable in his sheet.
User avatar
willthrill81
Posts: 22562
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Kitces article on lower prob of retirement success ok

Post by willthrill81 »

Horton wrote: Mon Jan 11, 2021 5:57 pm Ken Steiner recently wrote an article comparing and contrasting the article from the OP with the “Actuarial Approach” (essentially amortization based withdrawals).

http://howmuchcaniaffordtospendinretire ... l.html?m=1

Some take aways:
In summary, their model is less conservative (more aggressive) than the ABC with default assumptions, in that it produces higher initial total spending budgets.
Spending budgets determined under either the Kitces’ model or the Actuarial Approach are self-correcting as actual future experience emerges. Therefore, as noted in the Background section above, if spending under the Actuarial Approach is lower initially than spending under the Kitces’ model, it will catch up in time and eventually exceed the Kitces’ spending budget if assets are invested similarly, (which we have assumed).
Therefore, you have to decide how much risk you are willing to take to fund your future expenses. And the more risk you are willing to take, the more you need to be willing to reduce future expenses when necessary.
Whether you use a budget prepared by your financial advisor, you use one of our workbooks or you use some other approach, we suggest that you compare your final spending budget with the Actuarial Budget Benchmark determined with default assumptions so that you can gauge how conservative or how aggressive your budget may be and how much risk you (and not your financial advisor) are assuming with your spending plan.
So another situation where ABW is at least no worse and potentially significantly better than an alternative? :wink:
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings
User avatar
vineviz
Posts: 8500
Joined: Tue May 15, 2018 1:55 pm

Re: Kitces article on lower prob of retirement success ok

Post by vineviz »

Ben Mathew wrote: Mon Jan 11, 2021 1:20 pm Failure rate calculations have never made much sense because it assumes a withdrawal strategy that does not make sense--withdraw a fixed dollar amount without adjusting for portfolio performance.
Such calculations make sense specifically BECAUSE they are they assume a fixed withdrawal rate. What the "failure rate" tells you is essentially the probability that you will NEED to adjust withdrawals for portfolio performance.

In other words, a failure rate of 10% is basically telling you that there is a 10% chance that your withdrawals will need to be adjusted at some point. It's agnostic about WHEN and HOW DRAMATICALLY the adjustment should be: that's what your financial plan and/or investment policy statement is for. Some people might prefer small adjustments made earlier and/or with greater frequency than other people.

Moshe Milevsky's “Retirement Sustainability Quotient” or RSQ is more or less a scaled transformation of the failure rate (.e.g 1 minus failure rate) except it includes income from annuities and Social Security (so the lower bound is not zero). He discusses this in several of his books, but this white paper includes an overview.
"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: 1401
Joined: Tue Mar 13, 2018 11:41 am
Location: Seattle
Contact:

Re: Kitces article on lower prob of retirement success ok

Post by Ben Mathew »

vineviz wrote: Tue Jan 12, 2021 1:34 pm
Ben Mathew wrote: Mon Jan 11, 2021 1:20 pm Failure rate calculations have never made much sense because it assumes a withdrawal strategy that does not make sense--withdraw a fixed dollar amount without adjusting for portfolio performance.
Such calculations make sense specifically BECAUSE they are they assume a fixed withdrawal rate. What the "failure rate" tells you is essentially the probability that you will NEED to adjust withdrawals for portfolio performance.

In other words, a failure rate of 10% is basically telling you that there is a 10% chance that your withdrawals will need to be adjusted at some point. It's agnostic about WHEN and HOW DRAMATICALLY the adjustment should be: that's what your financial plan and/or investment policy statement is for. Some people might prefer small adjustments made earlier and/or with greater frequency than other people.
For a withdrawal strategy that adjusts to portfolio performance (withdraw more when the portfolio does well, and less when it does poorly), the probability that you will need to adjust withdrawals is 100%. The only scenario where you don't have to adjust is if the realized returns equals expected returns every year, which is basically zero probability unless the portfolio is 100% duration matched real bonds/annuities. The "failure rate" of 10% does not correspond to anything meaningful for such withdrawal strategies. It's the probability distribution of withdrawals (e.g. 80% chance it's between $50,000 and $60,000) which conveys meaningful information.
User avatar
vineviz
Posts: 8500
Joined: Tue May 15, 2018 1:55 pm

Re: Kitces article on lower prob of retirement success ok

Post by vineviz »

Ben Mathew wrote: Tue Jan 12, 2021 2:08 pm
vineviz wrote: Tue Jan 12, 2021 1:34 pm
Ben Mathew wrote: Mon Jan 11, 2021 1:20 pm Failure rate calculations have never made much sense because it assumes a withdrawal strategy that does not make sense--withdraw a fixed dollar amount without adjusting for portfolio performance.
Such calculations make sense specifically BECAUSE they are they assume a fixed withdrawal rate. What the "failure rate" tells you is essentially the probability that you will NEED to adjust withdrawals for portfolio performance.

In other words, a failure rate of 10% is basically telling you that there is a 10% chance that your withdrawals will need to be adjusted at some point. It's agnostic about WHEN and HOW DRAMATICALLY the adjustment should be: that's what your financial plan and/or investment policy statement is for. Some people might prefer small adjustments made earlier and/or with greater frequency than other people.
For a withdrawal strategy that adjusts to portfolio performance (withdraw more when the portfolio does well, and less when it does poorly), the probability that you will need to adjust withdrawals is 100%. The only scenario where you don't have to adjust is if the realized returns equals expected returns every year, which is basically zero probability. The "failure rate" of 10% does not correspond to anything meaningful for such withdrawal strategies. It's the probability distribution of withdrawals (e.g. 80% chance it's between $50,000 and $60,000) which conveys meaningful information.
Choosing to adjust and needing to adjust are not the same thing.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
WoodSpinner
Posts: 1868
Joined: Mon Feb 27, 2017 1:15 pm

Re: Kitces article on lower prob of retirement success ok

Post by WoodSpinner »

vineviz wrote: Tue Jan 12, 2021 1:34 pm
Ben Mathew wrote: Mon Jan 11, 2021 1:20 pm Failure rate calculations have never made much sense because it assumes a withdrawal strategy that does not make sense--withdraw a fixed dollar amount without adjusting for portfolio performance.
Such calculations make sense specifically BECAUSE they are they assume a fixed withdrawal rate. What the "failure rate" tells you is essentially the probability that you will NEED to adjust withdrawals for portfolio performance.

In other words, a failure rate of 10% is basically telling you that there is a 10% chance that your withdrawals will need to be adjusted at some point. It's agnostic about WHEN and HOW DRAMATICALLY the adjustment should be: that's what your financial plan and/or investment policy statement is for. Some people might prefer small adjustments made earlier and/or with greater frequency than other people.

Moshe Milevsky's “Retirement Sustainability Quotient” or RSQ is more or less a scaled transformation of the failure rate (.e.g 1 minus failure rate) except it includes income from annuities and Social Security (so the lower bound is not zero). He discusses this in several of his books, but this white paper includes an overview.
My approach has been to look deeper into the results of the Monte Carlo simulation. I am very interested in the distribution of the magnitude of the adjustment and the timing.

For instance, I am much more comfortable with a 30% failure rate but a mean 20% adjustment that typically occurs in year 28 of my plan. That is a much lower risk (IMHO) than a 10% failure rate with a 30% adjustment 5 years into the plan. Especially since I want to spend more in the early part of retirement on experiences.

Finding a tool that exposes that detail can be very helpful.

WoodSpinner
AlwaysLearningMore
Posts: 78
Joined: Sun Jul 26, 2020 2:29 pm

Re: Kitces article on lower prob of retirement success ok

Post by AlwaysLearningMore »

In this video he talks about how he frames the discussion with his clients as 'probability of adjustment' to their withdrawal plan, rather than use the word "failure" with them. One technique he discusses (and calls 'guardrails') is if in the first 15 years of retirement the withdrawal rate rises 1% over the starting WR (e.g., goes from 4% of nest egg to 5%) then decrease spending by 10%. Conversely, if the WR drops by 1%, then the retiree may raise spending by 10%.

https://tinyurl.com/y3huczyj
Post Reply