Trinity Study update

[Running_Man]

On the blog a reoccurring view of Bernstein's quote "A wildly optimistic historian might give us another few centuries of economic, political, and military continuity. Back-of-the-envelope, that’s about an 80% survival rate over the next 40 years. Thus, any estimate of long-term financial success greater than about 80% is meaningless." is presented.

This thought has been quoted to imply that 80 percent success rates are equal in survival probability to 100 percent success rates which they surely are not.
This mathematical error has been repeated enough to become integrated into financial planning by apparently the Trinity study. Unless the financial failures calculated were due to economic, political or military continuity which is certainly not the case used in the TRINITY study then that is an additional chance of failure is not a subset of the Bernstein's withdrawl from his mule calculation but a unique seperate calculation.

If you accept the calculation from Bernstein on that part -- the calculation by Trinity is the remainder of the other 80% of which the 75% success yield's a total expected success rate =
( .2 X 0% <Bernstein's rate of non continuity of financial society>) + (.8 X .75% <financial continuity expected success rate>) = 60% success rate for 7% non inflation indexed withdrawals. Therefore to utilize a withdrawal calculation near 100% would actually improve your chances of not going bust in your retirement by 33.33% which is very statistically significant.

 

[Moemg]

Why doesn't some group do a study on retirees who regularly take 3 or 4 % but one year and only one year they go crazy and take 7% for the trip of a life time ? That would be interesting !

 

[MasterBlaster]

On the blog a reoccurring view of Bernstein's quote "A wildly optimistic historian might give us another few centuries of economic, political, and military continuity. Back-of-the-envelope, that’s about an 80% survival rate over the next 40 years. Thus, any estimate of long-term financial success greater than about 80% is meaningless." is presented.

This thought has been quoted to imply that 80 percent success rates are equal in survival probability to 100 percent success rates which they surely are not.
This mathematical error has been repeated enough to become integrated into financial planning by apparently the Trinity study. Unless the financial failures calculated were due to economic, political or military continuity which is certainly not the case used in the TRINITY study then that is an additional chance of failure is not a subset of the Bernstein's withdrawal from his mule calculation but a unique seperate calculation.

If you accept the calculation from Bernstein on that part -- the calculation by Trinity is the remainder of the other 80% of which the 75% success yield's a total expected success rate =
( .2 X 0% <Bernstein's rate of non continuity of financial society>) + (.8 X .75% <financial continuity expected success rate>) = 60% success rate for 7% non inflation indexed withdrawals. Therefore to utilize a withdrawal calculation near 100% would actually improve your chances of not going bust in your retirement by 33.33% which is very statistically significant.

I don't get this analysis.

If there is a 20% chance of societal calamity, then in that case there is no safe withdrawal rate that works.

What is the point of calculating a SWR if the world changes dramatically for the worse? Would any resonable amount of money always help in that case ?

 

[Gearhead Jim]

I don't get this analysis.

If there is a 20% chance of societal calamity, then in that case there is no safe withdrawal rate that works.

What is the point of calculating a SWR if the world changes dramatically for the worse? Would any resonable amount of money always help in that case ?

My casual study of history, especially in the last hundred years, suggests that entering a period of calamity with a lot of money works out much better for most than entering the period with little.

You may need to do investment and financial maneuvers that would be considered crazy in normal times, but if times are not normal...

 

[DblDoc]

Dr Bernstein's point is quoted here at the end of Part III:

So live a little, and enjoy your money, for tomorrow we may be consumed by the ghosts of Hitler, Lenin, and Attila the Hun.

Your planned 40 yr retirement can be ruined by:

1) You die
2) Your spouse dies
3) You both die
4) The world as we know it ends
5) Sh*t happens

Do you want to spend an extra 10 years working and saving to increase the probability of your portfolio surviving from 80% to 95% if there is a > 20% chance it won't matter? Or would you rather take your chances knowing you may have to reduce your spending or go back to w&*k if you have an unfortunate sequence of returns.

DD

 

[Ed_The_Gypsy]

Your planned 40 yr retirement can be ruined by:

1) You die
2) Your spouse dies
3) You both die
4) The world as we know it ends
5) Sh*t happens

Well worth keeping in mind.

 

[haha]

My thought is that these things are pseudo issues. In reality, there is no way to rationally think through it, so most people will knowingly or unknowingly let their impusles and prejudices and habits of mind decide for them. Just like we decide most other things.

Ha

 

[Nords]

Why doesn't some group do a study on retirees who regularly take 3 or 4 % but one year and only one year they go crazy and take 7% for the trip of a life time ? That would be interesting !

You should write up your experiences and send them to Trinity for a starting point of their next research paper...

 

[Moemg]

Dr Bernstein's point is quoted here at the end of Part III: Your planned 40 yr retirement can be ruined by:

2) Your spouse dies

5) Sh*t happens

Been there done that !

 

[clifp]

Reading the Boglehead thread and this one, one thing struck on discussing SWR that we often forget the very real possibility that you will die before the age of 90 or 100.

Intercast did a post on this years ago but I haven't seen much else on the subject. I think it would be very nice addition to FIRECalc or another calculator.

I think it is safe to say that the SWR forum members here and also at Bogleheads are using have become more conservative in the last few years, which is entirely understandable. So I think it is valuable that Trinity authors have updated their work and are showing that 4 and even 5% withdrawal rates are reasonably safe. For those of us lucky/obsessed/disciplined enough to save more than we need it doesn't matter much. But for the zillions of people who's portfolio haven't recovered, seen their home values drop, and often had job loss, pay cut etc it is helpful to know that even with this crisis. A million dollars plus SS will still provide a comfortable retirement for most places in the US.

 

[Running_Man]

Dr Bernstein's point is quoted here at the end of Part III: Your planned 40 yr retirement can be ruined by:

1) You die
2) Your spouse dies
3) You both die
4) The world as we know it ends
5) Sh*t happens

Do you want to spend an extra 10 years working and saving to increase the probability of your portfolio surviving from 80% to 95% if there is a > 20% chance it won't matter? Or would you rather take your chances knowing you may have to reduce your spending or go back to w&*k if you have an unfortunate sequence of returns.

DD

All of the above could happen and 1, 3 &4 are beyond my control. #2 should be planned for and #5 is the purposes of developing a plan for retirement. However to go from an 80% to 95% chance using the trinity study at a 75/25 stock bond ratio means going from a 5% inflation adjusted withdrawl to 4 percent withdrawl rate.


If I assume I want 100K per year then a 4% withdrawl I would require a 2.5 million portfolio instead of 2 million. This would probably require a 3 year additional work time not 10 during my peak years and rather than reducing my spending to 50K per year at a relatively small downdraft (dividend and bond income on my portfolio) as the Trinity study would suggest I do, for myself I would find the safety margin prudent. This is ironic right now as I think about this because when I was younger I planned on retiring when I was 55 but after lurking here awhile 6 years ago I learned here that a 4 percent withdrawl is about the maximum I should plan on and I developed a plan to retire at 58 (3 years from now) I am still right on that plan.

I find it odd that numbers and calculations are the basis for academic excellence of a thesis but a obvious mathmatical equation is ignored in using that thesis in reviewing the expected terminal values for a retirement portfolio in favor of an emotional arguement in making a final decision.

If one wants to retire on a 60% expected chance of success after consulting with the most brilliant minds in publishing I have no issue with that, but the inference that an 80% firecalc expected chance is the same as 100% chance is misleading and to me seems wrong.

 

[chinaco]

On the blog a reoccurring view of Bernstein's quote "A wildly optimistic historian might give us another few centuries of economic, political, and military continuity. Back-of-the-envelope, that’s about an 80% survival rate over the next 40 years. Thus, any estimate of long-term financial success greater than about 80% is meaningless." is presented.

This thought has been quoted to imply that 80 percent success rates are equal in survival probability to 100 percent success rates which they surely are not.
This mathematical error has been repeated enough to become integrated into financial planning by apparently the Trinity study. Unless the financial failures calculated were due to economic, political or military continuity which is certainly not the case used in the TRINITY study then that is an additional chance of failure is not a subset of the Bernstein's withdrawl from his mule calculation but a unique seperate calculation.

If you accept the calculation from Bernstein on that part -- the calculation by Trinity is the remainder of the other 80% of which the 75% success yield's a total expected success rate =
( .2 X 0% <Bernstein's rate of non continuity of financial society>) + (.8 X .75% <financial continuity expected success rate>) = 60% success rate for 7% non inflation indexed withdrawals. Therefore to utilize a withdrawal calculation near 100% would actually improve your chances of not going bust in your retirement by 33.33% which is very statistically significant.

But his rhetorical quote was Not a suggestion to included it in a calculation for probability. Besides.... weaving the calamity into it could be higher at a personal level (which is much more likely)... your personal probability of calamity could be 90% because of making foolish investment move (to say the least of it).

He is just making the observation that 40 years is a long time... anything could happen! There is no certainty... focusing on historical stock market fluctuations and a seemingly workable WR% does not contain the entire set of issues that might break the portfolio.

IMO - That general observation he made should not be used as a reason to kick up the WR% to stress the portfolio. IOW 80% means there is a 20% chance of needing to make adjustments.... and that may occur at a time that not convenient or tolerable.... unless of course... one has excess assets (rich) and can tolerate an unexpected reduction.

If one wants more certainty... they should look towards something like Treasuries or SPIAs from the highest rated insurance companies.... but even then... you might not get your money if the world comes to an end!

 

[obgyn65]

Well put DblDoc. That's the key question many of us on this website try to answer IMO. I know I do - everyday.

Do you want to spend an extra 10 years working and saving to increase the probability of your portfolio surviving from 80% to 95% if there is a > 20% chance it won't matter? Or would you rather take your chances knowing you may have to reduce your spending or go back to w&*k if you have an unfortunate sequence of returns.

DD

 

[rescueme]

1) You die
2) Your spouse dies
3) You both die
4) The world as we know it ends
5) Sh*t happens

#1 (For me) is not a problem. Money is for the living, not the dead. As long as I feel that I planned to provide for my family after I pass, I can't be concerned about it.

#2 Again, not a financial but an emotional problem. Unless you are counting the spouses income (j*b, SS, pension, etc.) as a necessity for you to continue to be able to "financially survive", then it may be. However, that's why they have life insurance to cover such situations.

#3 See #1. If you are both dead, there is no problem. If you have somebody you are caring for (financially) - such as young (e.g. non-adult) childern or disabled person you are supporting (as we are) than there may be an impact, if you/spouse have not planned for that possible scenario.

#4 Again, if we're all dead why would we need money? If there was a global war and some folks survive, I don't think that money will help your survival.

#5 I think that is the greatest possibility for us that remain living, and why we continue to discuss/ponder "financial if's" on this forum.

 

[rescueme]

Reading the Boglehead thread and this one, one thing struck on discussing SWR that we often forget the very real possibility that you will die before the age of 90 or 100.

How true.

However (at least for me), I would rather die with money than live without it. That's the main risk of not having a "robust" plan.

Sure, I would say that most of us will have income in our later years, but could we continue to live in the manner we wish on that possible SS or a small pension?

That's the curse of an "over planner" such as myself (ask my DW). I still carry the remembrance of the first half of my life when money was in short supply, even for the basic necessities in life. I don't wish to return to those days and remained "scarred" by the memory...

 

[ERD50]

my bold...

On the blog a reoccurring view of Bernstein's quote "A wildly optimistic historian might give us another few centuries of economic, political, and military continuity. Back-of-the-envelope, that’s about an 80% survival rate over the next 40 years. Thus, any estimate of long-term financial success greater than about 80% is meaningless." is presented.

This thought has been quoted to imply that 80 percent success rates are equal in survival probability to 100 percent success rates which they surely are not.

This mathematical error has been repeated enough to become integrated into financial planning by apparently the Trinity study. Unless the financial failures calculated were due to economic, political or military continuity which is certainly not the case used in the TRINITY study then that is an additional chance of failure is not a subset of the Bernstein's withdrawal from his mule calculation but a unique seperate calculation.

I didn't read this update, so I won't comment on the Trinity or the blog interpretation, but I agree with you that many people seem to accept the implication of the line I bolded above. I've read it many times on this forum.

I don't get this analysis.

If there is a 20% chance of societal calamity, then in that case there is no safe withdrawal rate that works.

What is the point of calculating a SWR if the world changes dramatically for the worse? Would any resonable amount of money always help in that case ?

I calculate it for the 80% chance that the calamity does not occur. Else, we are looking at a situation where maybe I run out of money, and then I'm saying "What happened to that 20% chance that we were all gong to be wiped out by some external calamity?".

Let's take another example - let's say that a Life Expectancy calculator says I only have a 20% chance of living past 80 YO. Does that mean I shouldn't plan my finances for the 20% chance that I live past 80? I don't think so. A conservative investor is going to plan for the sum of the longest life they think reasonably possible, and other reasonably good/bad scenarios. Then you punt, or keep working 'one more year'.

edit/add:

How true.

However (at least for me), I would rather die with money than live without it. That's the main risk of not having a "robust" plan.

+1


-ERD50