What Success % do you use?

[youbet]

It's amazing to me that such a smart guy didn't conclude that his clients had a behavorial problem that needed fixing but rather, jumped to a conclusion that the principles that he previously espoused were broken and needed fixing.

While I tend to agree, micro economics is a behavioral science where human behavior and tendencies are key. Think of the shape of demand curves, etc. As you change the price of widgets and note the variation in quantity demanded you also note that you must change the widgets (or peoples' impression of them) in order to shift the demand curve (as opposed to moving along the demand curve).

audreyh1 hit it on the nose:

I think he adjusted his recommendations, because he did not believe he could adjust his clients' behavior.

 

[pb4uski]

I concede that he couldn't have anticipated that his clients would bail, but a wholesale change in approach is an overreaction. I'm guessing that not all of his clients bailed. There are hundreds or thousands of us here who stayed the course.... perhaps we didn't have the courage to sell bonds and buy stocks as our AA was screaming at us to do but we did have the courage to stand pat and divert withdrawals to bonds.

If he had decided to retain his old approach but added his new approach for "conservative investors" (aka those who bailed) then that would have been a more credible adjusment.

That said, I suspect that his clients that bailed tended to be more affluent. Let's say that they had a 3% WR... 20x would be 60% bonds and 40% stocks so while that isn't for me, it is within the realm of reasonableness I guess.

 

[Dtail]

Yes, he did and that's brought up just about every time he posts it. Regardless, there is a point to be made that there were people retiring in the past without a pension and for which backtesting of any kind didn't exist and they survived. And there were talking heads back then claiming that one could withdraw 8% and still be just fine. Regardless, now that we have some data available, and the tools to use them (and sometimes abuse them, unfortunately), I'd say we've made progress.

Agree with the above, but obviously in anyone's retirement survival, luck plays some role.
I think Peter Lynch was the SWR of 7% reference until Bill Bengen came along.

 

[Dtail]

Someone else asked something like this recently (or stated that they would just adjust spending). I asked - and when and by how much and how long would you need to cut spending. They did not report back.

A lot of people just assume some adjustments will do it. It's not so easy. So here's some actual data. Hold onto your hats (you may need to sell those hats)...

I ran a 40 year profile in FIRECalc, using a $1,000,000 starting portfolio for easy math. "INVESTIGATE" tells me that an initial $46,182 (4.62%) spend will result in an ~ 70% success rate.

Then I cut the the years to 5, 4 then 3 to try to find where it dropped by 50%. I assumed that would be the 'alarm' to cut spending - but use whatever point you want. At year 4, one line dipped to $470K, close enough.

So then I added $23,000 of 'off-chart' inflation adjusted income in year 2022 (that might be one year off?). That would be the same as cutting spending in half. How long did I need to do that to reach 100% for the 40 year period?

Any guesses? 5 years? No! 7, 10? No! 14 years!

So if you retired at 55 with a plan to cover the possibility of reaching age 95 (not uncommon for one of a married couple), that means cutting spending in half from age 59 to age 73. And you can't get those years back.

Here's a link to play with: https://goo.gl/NQVNpB

For reference, a 100% successful WR for those 40 years was $33,413 (3.34%). Up to each person, but I'd prefer to start with $33,413 with a high confidence that I will not need to cut spending, versus $46,182, and possibly need to cut spending in half for a decade and a half.

edit/add: Another interesting point - while the 70% path starts out spending 38% more than the 100% path, but over the full 40 year period (after cuts to reach 100%), it amounts to an average of only 14.7% more. And the 70% path doesn't catch up in total amount spent until age 80.

-ERD50

Didn't Big ERN run some numbers which are in conclusion with your findings in a general sense?

 

[ERD50]

It's amazing to me that such a smart guy didn't conclude that his clients had a behavorial problem that needed fixing but rather, jumped to a conclusion that the principles that he previously espoused were broken and needed fixing.

I guess it doesn't surprise me. How would he know his clients would bail? Until faced with something that appeared to be armageddon, few clients perhaps had reacted such a way.

How would you fix such a behavioral problem? I don't think you can. I think he adjusted his recommendations, because he did not believe he could adjust his clients' behavior. People coming to him for help were already perhaps more vulnerable and they didn't have confidence to invest independently?

But I still feel that the principles and the behaviors should be kept separate. A principle is a principle. If behaviors mean you have trouble following that principle, the principle isn't wrong, but you may need to make adjustment to the implementation to account for behaviors.

It's like the analogy I've used with fear of flying. If flying really disturbs you, avoid it. But don't try to tell me that flying is more dangerous than driving. They are two separate things, and it only confuses things by blending them.

-ERD50

 

[GravitySucks]

So I'm trying to redefine Success (to me) in this area.

If my health remains good, I have a paid off house, and make it to 70 before starting SS, I think just having $300K of today's dollars at age 70 would still put me in a better situation than 80% of my fellow Americans. This could really increase what I could spend from 60 to 70.
Honestly how many cars will i buy after 70? 90% of the people I know over 80 don t want to travel much. Even just SS at that level covers all expenses and one of two cruises a year.
This success has a known end date as opposed to trying to look for my expiration tag.

If course that puts long term care pretty much into the hand of Medicaid. And even more a problem is what would I spend the extra on? Four months of travel and gifting some to DM and DDs this year and I'm still under last year's restricted spending.

 

[HI Bill]

...having $300K of today's dollars at age 70 would still put me in a better situation than 80% of my fellow Americans. This could really increase what I could spend from 60 to 70.
Honestly how many cars will i buy after 70? 90% of the people I know over 80 don t want to travel much. Even just SS at that level covers all expenses and one of two cruises a year.
This success has a known end date as opposed to trying to look for my expiration tag.

Being better off than 80% of fellow Americans is irrelevant, as my wife always tells me, even when I'm talking about the 95 percentile. (The lesson is that comparing yourself to others is not what's important).

The question is, are you comfortable with the standard of living the asset level you're talking about will provide? Does it afford you the ability to do the things you want to do? $300K at 4% is $12K annual spend. There's not much cusion for emergencies, and even a couple of dental crowns could wipe out some of that, or an expensive hospital stay, or a new house roof. If that's all you need on top of SS, then you're good, except the whole Medicare/LTC thing, and the lack of budget for contingencies. Maybe a reverse mortgage would be in order.

BTW, my 81 year-old dad's girlfriend's father is 94 and is still driving. He just bought a new car last year! Of course, a $30K car in relation to his $3M+ in assets is pretty insignificant!

 

[Retch The Grate]

Anyone know how much cutting spending during down markets increases your % of success? Could a 10% decrease in spending during a bear bump you from 70% to 100%?

The three posts starting here: https://earlyretirementnow.com/2017...-safe-withdrawal-rates-part-9-guyton-klinger/ cover it (post 9-11 of the series). Basically, no. If you are going to rely on cutting withdrawals in down markets to make the plan work you run the risk of extended periods of significantly reduced withdrawals.

 

[flintnational]

100%, but my preferred mode is using FIRECalc to find the spending level at which I'll never (historically) run out of money. When I see that that "safe" number is 1½ - 2X what my annual spending projections are, then I feel pretty confident. So I'd guess that translates roughly into 150 - 200% success?

This is also how I look at SWR. And, our success percentages are similar.

 

[audreyh1]

But I still feel that the principles and the behaviors should be kept separate. A principle is a principle. If behaviors mean you have trouble following that principle, the principle isn't wrong, but you may need to make adjustment to the implementation to account for behaviors.

It's like the analogy I've used with fear of flying. If flying really disturbs you, avoid it. But don't try to tell me that flying is more dangerous than driving. They are two separate things, and it only confuses things by blending them.

-ERD50

Staying the course with investments is ultimately all about personal psychology.

An investor must know themselves well enough to choose an investment strategy that will keep them invested during very stressful times. I suppose for many, they can’t determine that ahead of time.

 

[GravitySucks]

Being better off than 80% of fellow Americans is irrelevant, as my wife always tells me, even when I'm talking about the 95 percentile. (The lesson is that comparing yourself to others is not what's important).

The question is, are you comfortable with the standard of living the asset level you're talking about will provide? Does it afford you the ability to do the things you want to do? $300K at 4% is $12K annual spend. There's not much cusion for emergencies, and even a couple of dental crowns could wipe out some of that, or an expensive hospital stay, or a new house roof. If that's all you need on top of SS, then you're good, except the whole Medicare/LTC thing, and the lack of budget for contingencies. Maybe a reverse mortgage would be in order.

BTW, my 81 year-old dad's girlfriend's father is 94 and is still driving. He just bought a new car last year! Of course, a $30K car in relation to his $3M+ in assets is pretty insignificant!

Better off than 80% has relevance. I do not see 80% of seniors doing poorly. The upper half seems to be doing pretty good.

Why would you be looking at 4% withdraw? I doubt I'll live last 85 but planning to 92. That leaves 22 years. That would be a 5% + SWR. $15 k on top of SS would be fun money.

I befriended a 90 year old gent last winter. He still went to lunch out once a week. Dinner out once a week. Enjoyed minigolf. Drives his 15 year old Lexus. He said he pretty much spends nothing anymore.
Reality check shows my driving after 85 will not be a real good idea.

Not saying $300k is ideal, but if I get to 70 and I'm there I'll consider it a win. SUCCESS!
(I also have a 65 year old success level. Medicare and local tax relief are big deals for me. Also had a 59.5 year old success level.)

 

[HI Bill]

Not saying $300k is ideal, but if I get to 70 and I'm there I'll consider it a win. SUCCESS!
(I also have a 65 year old success level. Medicare and local tax relief are big deals for me. Also had a 59.5 year old success level.)

With your $15K spend, SOR risks are real, if you were to live past 90 (~20 years out), depending on your AA. But you're right, most do live with less than your assets at 70; it's about budget and cash flow, so it looks like you'll probably be successful!

 

[GravitySucks]

With your $15K spend, SOR risks are real, if you were to live past 90 (~20 years out), depending on your AA. But you're right, most do live with less than your assets at 70; it's about budget and cash flow, so it looks like you'll probably be successful!

I think I got my point lost in minutiae.

My point is there can be other measures of success.

And in reality I'm struggling to spend my super safe budget of the first 5 years.

 

[big-papa]

The three posts starting here: https://earlyretirementnow.com/2017...-safe-withdrawal-rates-part-9-guyton-klinger/ cover it (post 9-11 of the series). Basically, no. If you are going to rely on cutting withdrawals in down markets to make the plan work you run the risk of extended periods of significantly reduced withdrawals.

For those of us who use or plan on using a variable withdrawal method, that's actually a "yes". If the definition of success is that the portfolio survives, then variable withdrawal methods can meet that definition. That's the tradeoff to be made with a variable withdrawal method - one must be able to survive a possible extended period of reduced withdrawals for a higher likelihood of portfolio survivability. Not everybody can, but many can, especially if they have a steady floor of income (SS, Pension, Rental income, etc.). And depending on the size of the income floor, one may choose to go more conservative in the AA which will most likely reduce the depth of the reduced withdrawals.

 

[kcowan]

That is not what FIRECalc does! It reports every 30 year period in its history (age has nothing to do with it). That's way, way different, and far more useful. (ooops, cross posted with MB)

I agree that I stated it badly.

Not sure what you mean by "the current generation". But many retirees and some considering retiring lived through the inflation of the 80's (I had a 17% mortgage, adjustable, so it ended well).

At any rate, anyone running FIRECalc will be tested against that 80's inflation. 1966 is one of the 'killer' starting years, largely due to the 80's inflation....-ERD50

I was referring to my kids who never appreciated 20+% interest rates. And yes the starting year makes a huge difference. My 30 year look-back started in 1972. I am ok but mostly because of higher equity holdings and significant LBYM due to spending 6 months in Mexico. Neither of these could have been forecast in 2002.

 

[HaveEnough]

I look at how much I can spend and still hit 95% until 95 years old. The number is over 5 times what we currently spend. I'm not concerned unless I pick up some expensive arm candy in the future.

 

[ychuck46]

Always conservative

100%. And I like Monte Carlo

 

[W2R]

Nine pages and I haven't weighed in yet? OOPS Well, here goes:

I generally use 95% - 100%. Right now, I have saved the results of 5 runs done over the past year. Three of these were done using 95%, and two were done using 100%.

Of these runs, two were done to age 95, one was done to age 100, and two were done to age 117, which I mistakenly thought was the age of the oldest living human. But I was wrong. Wikipedia says

Since the death of 117-year-old Chiyo Miyako of Japan on 22 July 2018, 115-year-old Kane Tanaka, also of Japan, born 2 January 1903, is the oldest living person in the world whose age has been documented.

So, 115 is the age of the oldest living human now that Chiyo Miyako has passed away.

Right now my purpose in running FIRECalc is to persuade myself to loosen the purse strings and spend a bit more. Even with these inputs, FIRECalc says I can spend more and that is what I wanted to know. My guess is that I have the chance of a snowball in h*ll of living to 115, but the point is, even if I did, I'd still have plenty of money at that age, even if I started spending more than I am presently spending.

Inputs to models are best chosen by thinking about your intended usage for the model.

 

[TwoByFour]

Anyone know how much cutting spending during down markets increases your % of success? Could a 10% decrease in spending during a bear bump you from 70% to 100%?

I wrote my own RIP tool so it is easy for me to program into it questions like yours. I did that and here is what I found:

With a starting asset level of $1M and a withdrawal rate of 5.4% the success rate is close to 70% for a 30 year life expectancy. That is a SWR type of withdrawal - inflation adjust a fixed dollar amount of $54k every year.

If I program in what you said, that you lower the withdrawal rate by 10% if the market is bear (i.e., current S&P 500 is at least 20% off of the high) then the success rate only goes up to 74%.

If I program in that you lower the withdrawal amount to $44k if the market is corrected (i.e., current S&P 500 is at least 10% off the high) then the success rate goes up to 95%.

I didn't experiment any further than that. But you can see the trend - you can indeed improve your success rate but you will have to be a lot more aggressive in how much you cut your WR and the extent of the "down" market that triggers your cut.

 

[RobLJ]

Thanks to the geeks on the blog, who reminded me again that you could use FireCalc to "solve" for spending level, based on the historic data, so I finally did it.

At a 99% success level (with 1/4 of the portfolio always available), FireCalc indicates I could spend approximately 40% more than what I planned. (And we're not spending that amount this year after DW's retirement).

It's 55% more if I'm willing to spend down the portfolio completely.

 

[pb4uski]

If you have used retirement calculators, what success percentage are you comfortable using / targeting?

Do you prefer historical (firecalc, ********) or Monte Carlo?

(Being ridiculously conservative, I always go for 100% success.)

95% or better is as good as 100%. Used a number of them.

 

[ERD50]

95% or better is as good as 100%. Used a number of them.

I don't agree with this at all. Per FIRECalc:

A 40 year profile and a 1M portfolio, a 100% historical safe WR starts at $33,413.
A 40 year profile and a 1M portfolio, a x95% historical safe WR starts at $36,582.

If being able to spend 10% more was just as good, why wouldn't we all do it?

Of course spending less will help any portfolio. How can that be denied?

Sure, we can't know the variability of the future, A 95% success rate may succeed the future. Heck, a 50% success rate succeeds half the time! ( ). But we can know that no matter what, the person spending less will be able to hold on longer if things are worse than the worst of the past.

-ERD50

 

[pb4uski]

My point was that there is enough wiggle room in the data and process that a 95% success rate is for all intents not very different from a 100% success rate.... so if you are at 95% there is no real need to continue to work until you get to 100%.... or put another way, 100% is belts and suspenders compared to 95%.

Those with a mathematical bent can find endless fascination in these effects, but unless you just enjoy the math, keep in mind that when the issue is how close to the edge one can get over the 30-50 years of a retirement, the mathematician might measure with a micrometer, but the retiree will be cutting with an axe.

 

[ERD50]

My point was that there is enough wiggle room in the data and process that a 95% success rate is for all intents not very different from a 100% success rate.... so if you are at 95% there is no real need to continue to work until you get to 100%.... or put another way, 100% is belts and suspenders compared to 95%.

Those with a mathematical bent can find endless fascination in these effects, but unless you just enjoy the math, keep in mind that when the issue is how close to the edge one can get over the 30-50 years of a retirement, the mathematician might measure with a micrometer, but the retiree will be cutting with an axe.

I still don't agree that any of that is useful.

The 'argument' is one of absolute accuracy, versus relative accuracy. And I agree that things like FIRECalc don't give us much in the way of absolute accuracy, but that is no reason to dismiss the relative accuracy that it provides.

An analogy to illustrate:

I have two ropes, one 95' long, one 100' long. Without something to measure them with, you'd be hard pressed to tell me how long each is with any accuracy. But if you lay them side by side, you could easily tell me which one is longer, and by approximately how much ( 5' is easier to estimate than 100').

Now say we are on a boat, and you fall overboard, and drift ~ 100' away before I can grab one of those ropes and tie it to a life preserver. Now are you going to tell me it makes no difference if I chose the 95' rope or the 100' rope? It may not make a difference, depending, but clearly, the 100' rope is the safest choice, and it could make the difference between life and death. For safety, you choose the 100' long rope, every time.

-ERD50

 

[pb4uski]

I guess we'll have to agree to disagree. By the way, the rope analogy was really silly.