Author Topic: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?  (Read 10678 times)

Bob W

  • Magnum Stache
  • ******
  • Posts: 2942
  • Age: 65
  • Location: Missouri
  • Live on minimum wage, earn on maximum
I searched but didn't find my answer.

I'm not one to agree that there is anything realistic about the SWR (it is all based on hope my friends) but let's assume I am and some others here are.

I see 3-4% as an oft stated rate.  Could someone detail the formula for me?

Is it 3% of the original number forever or is it 3% of the original number plus x% inflation per year. Or, and I'm assuming it is not, is it 3% of the trending value each year?

So I'm thinking I start with 1000.

The first year I draw down 30 (3% SWR) ---- so the next year I draw down 31, then 32,  then 33,  then 34.5 and so on indefinitely to take into account a 3% inflation rate?    Is that how you guys see it?




arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Yes, the classic 4% SWR as done in the Trinity study takes 4% of the initial portfolio amount and adjusts it upwards for inflation each year, regardless of market performance.

Most of us think that's a bit unrealistic.

www.cfiresim.com lets you play with withdrawal scenarios.

I searched but didn't find my answer.

That's weird.  I searched Google for "where does the 4% safe withdrawal rate come from" -- whaddya know, MMM is the first link.   There's a bunch of other ones though (many questioning if it's still "safe") that will help explain it.

https://www.google.com/search?q=where+does+the+4%25+safe+withdrawal+rate+come+from

The Wikipedia page on the Trinity study (4th link on that search) specifically mentions:
Quote
The 4% refers to the portion of the portfolio withdrawn during the first year; it is assumed that the portion withdrawn in subsequent years will increase with the consumer price index (CPI) to keep pace with the cost of living.

What did you search for that you couldn't find an explanation of it?
« Last Edit: January 27, 2015, 03:46:24 PM by arebelspy »
I am a former teacher who accumulated a bunch of real estate, retired at 29, spent some time traveling the world full time and am now settled with three kids.
If you want to know more about me, this Business Insider profile tells the story pretty well.
I (rarely) blog at AdventuringAlong.com. Check out the Now page to see what I'm up to currently.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
The math behind the SWR is solid. It takes into account the past performance of the market going back to the creation of the S&P500 Index: There have been very, very few periods of time over which an S&P500 Index position would exhaust it's self at a ~4% withdrawal rate. The math holds (althought it is a slightly higher risk) if you randomize time periods as FIRECalc does.

If you take the withdrawal rate down to 3%, exhausting said position is almost impossible. You need to string together nearly every financial disaster of the last 150 years back to back. There is little non-emotional reason to strive for a lower SWR than this.

A 2% withdrawal rate is impossible to exhaust. Period. Even if you string said disasters together. We're talking depression, followed by 2008, followed by 2000, followed by whatever that shit in the 80's was. That's why this SWR is the ultimate goal of the risk adverse amongst us. There is literally no mathmatical justification for going any more conservative.

Don't confuse the confidence of the informed with the hope of the ignorant.
« Last Edit: January 27, 2015, 04:05:38 PM by Chuck »

dandarc

  • Walrus Stache
  • *******
  • Posts: 5454
  • Age: 41
  • Pronouns: he/him/his
Don't confuse the confidence of the informed with the hope of the ignorant.
That's a great line.

WildJager

  • Bristles
  • ***
  • Posts: 440
  • Age: 37
    • Can't complain.
What those above me said checks with my research.  Even at higher withdrawal rates of up to 7%, there was still an 85% success rate according to the study.

With all this said though, we are looking in the past to guide our future strategy.  Like anything in life, the future could potentially not hold true to this model.  However, if you are investing at all, you are putting in some faith that the entire economy won't collapse.  Scarily enough, it almost did as a result of the subprime lending shenanigans of 2009.  Hence the government bailout of banks "too big to fail.". Everyone would have been fucked, not just the banks.

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2204
  • Age: 43
There is literally no mathmatical justification for going any more conservative.

This statement assumes that one's withdrawal rate (as adjusted for inflation) is set in stone.  The reason the more risk-averse among us shoot for lower withdrawal rates is not solely because they are afraid of investment underperformance but also because they are afraid of unexpectedly high expenses.  Although I personally favor the approach of using a more aggressive withdrawal rate with some built-in flexibility in my retirement plan (i.e., recognizing that if necessary I may have to cut expenses, go back to work, etc.), unless you have a functioning crystal ball that can predict the future in terms of both market returns and expense levels, you can't write off the more conservative strategies as having absolutely no mathematical justification.

Also, don't forget that failure rates tend to be somewhat higher for extremely long retirements (the Trinity study and its progeny are based on 30 year retirements).

Capsu78

  • Pencil Stache
  • ****
  • Posts: 765
  • Location: Chicagoland
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?    Or does that mean with market gains, there willstill be $1000ish Wahooties left for pass along to his cat?

Timmmy

  • Bristles
  • ***
  • Posts: 439
  • Age: 40
  • Location: Madison Heights, Michigan
I'm interested in following this discussion. 

I think I understand the SWR but hashing it out a few times in a few different ways will help me to understand it better. 

Psychstache

  • Handlebar Stache
  • *****
  • Posts: 1594
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?    Or does that mean with market gains, there willstill be $1000ish Wahooties left for pass along to his cat?

It could be. Or it culd be less. Or it could be more. Year to year, the fluctuations are unpredictable, that's why you have to consider the larger picture.

In a recent test of cfiresim, my scenario indicated final balances when I die at anywhere from 102M to -800K.

Capsu78

  • Pencil Stache
  • ****
  • Posts: 765
  • Location: Chicagoland
... And the SWR is a good guideline, until applied to many real world scenarios.  My mothers end of life days included 27 days of hospital and 4 days of hospice- all mostly covered under my parents HC provisions.  My father, OTOH, spent his last 3.5 years in assisted living.  Fortunately his biggest insurance win in his life was buying long term care 3 years prior to being diagnosed with Parkinsons.  While that "full coverage at $140/day" was a God send, the cost of care was closer to $200/day.   Certainly a different burn rate than 3%.

annod

  • 5 O'Clock Shadow
  • *
  • Posts: 50
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #10 on: January 27, 2015, 05:05:04 PM »
I am wondering about SWR withdrawal rate too. Stock market fluctuates, so what would people do in a bear stock market scenario?
 For example, if you have a million dollars in your portfolio, and you are withdrawing 4%...so, $40K. So if there is a huge stock market downturn and your portfolio is down to $700k this year. Then do you withdraw $28K this year?  For example, you normally need 40K to cover expenses, in a bear market, you tighten your belt (let's say, no more traveling or selling a car) and you still find yourself needing 32K for expenses. Do you cheat yourself and withdraw extra 4K? (so you are withdrawing 4.57% that year), or should you have a separate emergency/bear Market funds account (for example $10K in a savings account)?


MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #11 on: January 27, 2015, 05:07:13 PM »
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?
Yes, this is the Trinity Study "definition of success."  Replacing "statistically" with "historically, based on ~70 years of data" is even more correct, as statistics doesn't often return "never" as a result.

Quote
Or does that mean with market gains, there will still be $1000ish Wahooties left for pass along to his cat?
Because the answer above is "yes," the answer here is "no."

Check2400

  • Stubble
  • **
  • Posts: 175
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #12 on: January 27, 2015, 05:10:11 PM »
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?    Or does that mean with market gains, there willstill be $1000ish Wahooties left for pass along to his cat?

My basic understanding is this
1000 Wahooties will earn on average a 7% Return, annually.  This is not the year you are FI on SWR.  After one year, though, there will be 1070 Wahooties.
If you withdraw 4% (SWR), you will withdraw 43.8 wahooties, leaving you 1027.2 Wahooties.  Assuming 43.80 Wahooties covers expenses, congrats, you're FI.   
The next year, those Wahooties grow at 7% again, but with the extra 27 Wahooties.  This means you end the year with 1099.10 Wahooties.
From there, 4% gets you 43.96 Wahooties (Wahoo! A raise!).

In the most terrible of economies, not the historical 7% average growth, 4% is the known worst case scenario.  Meaning
1000 Wahooties in a bad year leaves you at 1040 Wahooties.
4% withdrawal gets you 41.60 Wahooties, leaving you 998.4.  You're using principal now, albeit slowly.  This is why the SWD rate is lower than the historical average. 

In other words, the SWR is less than the increases in your assets historically.  So you essentially live off the interest, and never truly have to touch the corpus of the money, just the growth. 

This could be all wrong, but this is what I've picked up one month into MMM.  In truth, I just wanted to say Wahooties as often as possible.  It will henceforth be my new catchall term for investments or money.  Saying Widgets is just no fun anymore...


matchewed

  • Magnum Stache
  • ******
  • Posts: 4422
  • Location: CT
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #14 on: January 27, 2015, 05:19:47 PM »
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?    Or does that mean with market gains, there willstill be $1000ish Wahooties left for pass along to his cat?

The former. That is statistically if the future is like the past. If you want to make it more successful then you allow yourself wiggle room around that %. The latter is never a guarantee.

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #15 on: January 27, 2015, 05:21:17 PM »
...
In the most terrible of economies, not the historical 7% average growth, 4% is the known worst case scenario.  Meaning
1000 Wahooties in a bad year leaves you at 1040 Wahooties.
...
This could be all wrong, but this is what I've picked up one month into MMM.
It's not all wrong, but it is not correct.  Kudos for having the self-awareness to acknowledge the possibility.  Stick with it and your knowledge will accumulate.

Back to the point: the definition of "known worst case scenario" depends very much on the time duration assumed.  Don't need to go back very far to see a negative ~40% for a year, in which case 1000 drops to 600.

The Trinity (and related) studies found, however, that if one doesn't panic but merely remains invested, things will turn around fast enough that you don't run out of money within 30 years if your withdrawals have been equal to or less than the Safe Withdrawal Rate.  Reading the original articles, as linked above, is worthwhile.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #16 on: January 27, 2015, 05:23:46 PM »
OK, since we have dumbed this concept down enough that even the mathematically challenged can participate, lets go back to Bob's scenario.  He starts with $1000 Wahooties, takes out $30 Wahooties plus a few more Wahooties every year for inflation and he statistically will never run out of Wahooties for 30 years, correct?    Or does that mean with market gains, there willstill be $1000ish Wahooties left for pass along to his cat?

My basic understanding is this
You're close.

The "worst case scenario" is not a 4% gain. It's a 45+% loss.

The gains (many of which are 10, 15 or even 30%) simply outweigh these occasional huge decreases, and you can safely withdraw 4% from your balance each year and will almost never run out of money.

Capsu78

  • Pencil Stache
  • ****
  • Posts: 765
  • Location: Chicagoland
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #17 on: January 27, 2015, 05:24:11 PM »
Actually Check, the term Wahooties can be applied to any foreign currency, and it was our term of endearment when we travelled on business to Canada... until those polite Canadians flip the equation on us and we turned Dollars into Wahooties.

seattlecyclone

  • Walrus Stache
  • *******
  • Posts: 7254
  • Age: 39
  • Location: Seattle, WA
    • My blog
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #18 on: January 27, 2015, 05:28:48 PM »
My father, OTOH, spent his last 3.5 years in assisted living.  Fortunately his biggest insurance win in his life was buying long term care 3 years prior to being diagnosed with Parkinsons.  While that "full coverage at $140/day" was a God send, the cost of care was closer to $200/day. Certainly a different burn rate than 3%.

...depends on how big your stash is. $60/day is $20.9k/year, which is 3% of $730k. I imagine many of us are planning to support expenses at least that high during retirement.

It's important to remember that there's a lot of variation in the back-tested scenarios. When you pick a withdrawal rate low enough to make it almost impossible to run out of money if you keep withdrawing at that rate, you're also more likely than not to die a multi-millionaire. The multi-year assisted living expense only becomes a problem if you end up in one of the few retirement scenarios where your chosen withdrawal rate would leave you almost but not quite broke at the end of life without the assisted living expense. You're compounding an unlikely event on top of another unlikely event. At the end of the day you'll have to decide how much longer you're willing to work to prepare for this scenario. Remember that it's impossible to protect against every possible risk.

Chuck

  • Bristles
  • ***
  • Posts: 407
  • Age: 35
  • Location: Northern VA
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #19 on: January 27, 2015, 05:38:06 PM »
My father, OTOH, spent his last 3.5 years in assisted living.  Fortunately his biggest insurance win in his life was buying long term care 3 years prior to being diagnosed with Parkinsons.  While that "full coverage at $140/day" was a God send, the cost of care was closer to $200/day. Certainly a different burn rate than 3%.

...depends on how big your stash is. $60/day is $20.9k/year, which is 3% of $730k. I imagine many of us are planning to support expenses at least that high during retirement.

It's important to remember that there's a lot of variation in the back-tested scenarios. When you pick a withdrawal rate low enough to make it almost impossible to run out of money if you keep withdrawing at that rate, you're also more likely than not to die a multi-millionaire. The multi-year assisted living expense only becomes a problem if you end up in one of the few retirement scenarios where your chosen withdrawal rate would leave you almost but not quite broke at the end of life without the assisted living expense. You're compounding an unlikely event on top of another unlikely event. At the end of the day you'll have to decide how much longer you're willing to work to prepare for this scenario. Remember that it's impossible to protect against every possible risk.
+1

deborah

  • Senior Mustachian
  • ********
  • Posts: 15958
  • Age: 14
  • Location: Australia or another awesome area
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #20 on: January 27, 2015, 05:47:44 PM »
Another point is that 4% works for the US past performance. It doesn't work for most other countries past performance - for Australia it is 3.6%. Given that 4% is possibly a maximum, that may not hold for the US or any other country in the future, it is reasonable to have a lower SWR.

The last 8 years of life are the most expensive according to multiple things I have read, so you may wish to think about the funds for them as a separate option.

Capsu78

  • Pencil Stache
  • ****
  • Posts: 765
  • Location: Chicagoland
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #21 on: January 27, 2015, 05:54:30 PM »
My father, OTOH, spent his last 3.5 years in assisted living.  Fortunately his biggest insurance win in his life was buying long term care 3 years prior to being diagnosed with Parkinsons.  While that "full coverage at $140/day" was a God send, the cost of care was closer to $200/day. Certainly a different burn rate than 3%.

...depends on how big your stash is. $60/day is $20.9k/year, which is 3% of $730k. I imagine many of us are planning to support expenses at least that high during retirement.

It's important to remember that there's a lot of variation in the back-tested scenarios. When you pick a withdrawal rate low enough to make it almost impossible to run out of money if you keep withdrawing at that rate, you're also more likely than not to die a multi-millionaire. The multi-year assisted living expense only becomes a problem if you end up in one of the few retirement scenarios where your chosen withdrawal rate would leave you almost but not quite broke at the end of life without the assisted living expense. You're compounding an unlikely event on top of another unlikely event. At the end of the day you'll have to decide how much longer you're willing to work to prepare for this scenario. Remember that it's impossible to protect against every possible risk.
+1
Dad retired in 1988, with a paid off home valued at maybe $140,000.  His stache wasn't anything like what many here have or most need.  Yes Seattle, the $60 per day gap was manageable... the problem is the "expensive decisions" had to be made 20 years into his retirement and 7 years after his wife passed.  He also bought the policies on him and Mom in 1986 and I am guessing they would be considered Cadillac by todays standards.   

Bob W

  • Magnum Stache
  • ******
  • Posts: 2942
  • Age: 65
  • Location: Missouri
  • Live on minimum wage, earn on maximum
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #22 on: January 27, 2015, 08:29:32 PM »
Yes, the classic 4% SWR as done in the Trinity study takes 4% of the initial portfolio amount and adjusts it upwards for inflation each year, regardless of market performance.

Most of us think that's a bit unrealistic.

www.cfiresim.com lets you play with withdrawal scenarios.

I searched but didn't find my answer.

That's weird.  I searched Google for "where does the 4% safe withdrawal rate come from" -- whaddya know, MMM is the first link.   There's a bunch of other ones though (many questioning if it's still "safe") that will help explain it.

https://www.google.com/search?q=where+does+the+4%25+safe+withdrawal+rate+come+from

The Wikipedia page on the Trinity study (4th link on that search) specifically mentions:
Quote
The 4% refers to the portion of the portfolio withdrawn during the first year; it is assumed that the portion withdrawn in subsequent years will increase with the consumer price index (CPI) to keep pace with the cost of living.

What did you search for that you couldn't find an explanation of it?
the. MMM search bar

Bob W

  • Magnum Stache
  • ******
  • Posts: 2942
  • Age: 65
  • Location: Missouri
  • Live on minimum wage, earn on maximum
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #23 on: January 27, 2015, 08:49:39 PM »
Don't confuse the confidence of the informed with the hope of the ignorant.
That's a great line.
It is a great quote.  Unfortunately it doesn't change the fact that the future does not predict the past 50 years ahead.   Thanks everyone for weighing in.   

dividendman

  • Handlebar Stache
  • *****
  • Posts: 1899
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #24 on: January 27, 2015, 11:00:57 PM »
Don't confuse the confidence of the informed with the hope of the ignorant.
That's a great line.
It is a great quote.  Unfortunately it doesn't change the fact that the future does not predict the past 50 years ahead.   Thanks everyone for weighing in.
Yup. So just plan for the most likely scenarios and a lot of the less likely ones so you're planning for like 99% of em and fuck the rest. If nuclear war comes or aliens enslave us or the economy implodes or the cure for cancer actually turns us into zombies, the fact that we didn't have big enough retirement accounts probably won't mean shit anyway.

Mustache Fatty

  • 5 O'Clock Shadow
  • *
  • Posts: 56
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #25 on: January 27, 2015, 11:08:05 PM »
I know the article is linked above and I could find out for myself if I took the time to read it.  However, it is late and I just thought I would throw the question out in case someone knows the answer because they know the study inside and out.  I read in another forum that the Trinity study found the 4% withdrawal rate to be safe while simultaneously assuming a 1% expense ratio.  Is this true? 

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #26 on: January 27, 2015, 11:39:28 PM »
I read in another forum that the Trinity study found the 4% withdrawal rate to be safe while simultaneously assuming a 1% expense ratio.  Is this true?

Not true.  At least, one finds two quotes:
Quote
The study did not adjust for taxes or transaction costs.  An investor’s own experience would differ depending on how much of his assets were in tax-deferred accounts, and the extent to which transaction costs could be held to a minimum using low-cost index funds.
Quote
The stock returns in the analysis are total monthly returns to the Standard & Poor's 500 Index.

Not 100% sure of the implications, but one could reasonably infer:
1.  They ignored taxes completely, and simply assumed withdrawals were used to pay all annual expenses.  Thus one needs to include annual tax expense when applying these principles to one's personal situation.
2.  They assumed a 0% expense ratio and assumed stock returns equal to the S&P 500 with dividends reinvested.  Thus a 1% expense ratio would lead to a lower success rate (equivalently, a higher SWR).

AJDZee

  • Bristles
  • ***
  • Posts: 310
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #27 on: January 28, 2015, 07:21:02 AM »
If you want a year-by-year look at the theoretical numbers...
4% SWR
6% Return on Investment
2% inflation
Starting with $1M portfolio this year

Year   Portfolio Start   4% Withdrawl   Market Return (6%)   Portfolio End   Living Expenses
2015   $1,000,000               $40,000   $60,000                  $1,020,000   $40,000
2016   $1,020,000               $40,800   $61,200                  $1,040,400   $40,800
2017   $1,040,400               $41,616   $62,424                  $1,061,208   $41,616
2018   $1,061,208               $42,448   $63,672                  $1,082,432   $42,448
2019   $1,082,432               $43,297   $64,946                  $1,104,081   $43,297
2020   $1,104,081               $44,163   $66,245                  $1,126,162   $44,163
2021   $1,126,162               $45,046   $67,570                  $1,148,686   $45,046
2022   $1,148,686               $45,947   $68,921                  $1,171,659   $45,947
2023   $1,171,659               $46,866   $70,300                  $1,195,093   $46,866
2024   $1,195,093               $47,804   $71,706                  $1,218,994   $47,804
2025   $1,218,994               $48,760   $73,140                  $1,243,374   $48,760
2026   $1,243,374               $49,735   $74,602                  $1,268,242   $49,735
2027   $1,268,242               $50,730   $76,095                  $1,293,607   $50,730
2028   $1,293,607               $51,744   $77,616                  $1,319,479   $51,744
2029   $1,319,479               $52,779   $79,169                  $1,345,868   $52,779
2030   $1,345,868               $53,835   $80,752                  $1,372,786   $53,835


Your expenses go up by 2% compounded year after year, but your 4% annual withdraw matches it exactly every year.
Obviously you won't get a smooth 6% return every year, this is simplified, but that's why the study is significant, it showed with real market fluctuations, 4% each and every year is sustainable, majority of the time.
« Last Edit: January 28, 2015, 07:29:17 AM by AJDZee »

nereo

  • Senior Mustachian
  • ********
  • Posts: 17496
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #28 on: January 28, 2015, 07:52:30 AM »

That's weird.  I searched Google for "where does the 4% safe withdrawal rate come from" -- whaddya know, MMM is the first link.   There's a bunch of other ones though (many questioning if it's still "safe") that will help explain it.
arebelspy - common man... of COURSE you have MMM's explanation is listed 1st on your google search.  You're a moderator on the MMM forum and google tailors their results to the user.  Tried the same search on a work computer and MMM's post came up third.  Wikipedia was 1st, forbes article was 2nd. 

To address the Bob W and the thread at hand (I just can't resist)
The trinity study specifically looked at how various portfolios would have survived at different withdraw rates using historical data.  Note the past tense.  The withdraw rate was real-adjusted dollars based on the annual inflation rate (every year it was adjusted upwards by the % inflation from the previous year).  It assumed that a portfolio "succeeded" if it never dropped below $0 during a continuous 30 year period.   The study makes the distinct point that if the future market conditions are similar to historical conditions, the results should be equatable. 

Many smart people question whether the next 50+ years will see gains as large as the previous 50 years.  Fair enough - maybe they re right, maybe not.  I'd just add that during every decade a group of smart, intelligent, economically-minded individuals have argued that the end of prosperity is coming.    Prudence dictates that we consider slower earnings over the next 30-50 years, but there's a fine line between being cautious and irrational fear.  FWIW, the last 5, 10, and 20 year periods showed annual, real-growth of 12.8%, 5.5% and 7.2%.  The latter two include the 'great-recession'.  Pretty good returns, historically speaking. 

Also, when you start looking into the historical simulations an interesting pattern emerges (fireCalc is great ofr this). Of the portfolios that fail, all drop below their initial value during the first 10 years (most within the first 5).  Certainly not all simulations that drop below the initial value fail; some recover and go on an absolute tear.  But it's relatively easy to label a portfolio as 'at risk' very early on in retirement, which means small course corrections can alleviate problems.  Also, while some historical portfolios failed at 4%, the overwhelming majority earned money over a 30 year period, several increasing over 4x.

Finally, Bob W you brought up whether the 4% was fixed or variable.  For the trinity study it was fixed to the original portfolio and annually adjusted to inflation, but that brings up a good point - by adjusting withdraw rates to the current portfolio value it is possible to ensure that it can never be exhausted. For example, "4% of the original portfolio or 4% of the current portfolio, whichever is less."  That is one strategy.  Others prefer to find some 'bulletproof' SWR of 3.5%, 3.0%, 2.8% - but one can never be certain where that 'bulletproof' SWR will actually be.  In the end we are just left asking "how similar to the past will the future be?"

dude

  • Handlebar Stache
  • *****
  • Posts: 2369
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #29 on: January 28, 2015, 09:18:08 AM »
... And the SWR is a good guideline, until applied to many real world scenarios.  My mothers end of life days included 27 days of hospital and 4 days of hospice- all mostly covered under my parents HC provisions.  My father, OTOH, spent his last 3.5 years in assisted living.  Fortunately his biggest insurance win in his life was buying long term care 3 years prior to being diagnosed with Parkinsons.  While that "full coverage at $140/day" was a God send, the cost of care was closer to $200/day.   Certainly a different burn rate than 3%.

Scott Burns at AssetBuilder just did a good column on this:

http://assetbuilder.com/scott_burns/will_nursing_care_take_all_your_money_maybe_not


dude

  • Handlebar Stache
  • *****
  • Posts: 2369
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #30 on: January 28, 2015, 09:24:57 AM »
I know the article is linked above and I could find out for myself if I took the time to read it.  However, it is late and I just thought I would throw the question out in case someone knows the answer because they know the study inside and out.  I read in another forum that the Trinity study found the 4% withdrawal rate to be safe while simultaneously assuming a 1% expense ratio.  Is this true?

You are probably referencing Wade Pfau's more recent stuff on the Trinity Study, claiming that 4% might be unsustainable, and suggesting a 3% burn instead.  Over at the Early Retirement Homepage, it was pointed out that, buried in Pfau's research, was an assumed 1% expense ratio.  it was accordingly mocked.

To my knowledge, by the way, it was William Bengen who started all this, and the Trinity Study confirmed his findings.

dude

  • Handlebar Stache
  • *****
  • Posts: 2369
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #31 on: January 28, 2015, 09:29:21 AM »
I am wondering about SWR withdrawal rate too. Stock market fluctuates, so what would people do in a bear stock market scenario?
 For example, if you have a million dollars in your portfolio, and you are withdrawing 4%...so, $40K. So if there is a huge stock market downturn and your portfolio is down to $700k this year. Then do you withdraw $28K this year?  For example, you normally need 40K to cover expenses, in a bear market, you tighten your belt (let's say, no more traveling or selling a car) and you still find yourself needing 32K for expenses. Do you cheat yourself and withdraw extra 4K? (so you are withdrawing 4.57% that year), or should you have a separate emergency/bear Market funds account (for example $10K in a savings account)?

The 4% literature accounts for this generally -- however, there is a higher failure rate if that bear market occurs in the early part of your retirement (see "sequence of returns risk"), which is why many in this forum intend to either keep cash on the side to cover a 1-3 year downturn, or work part-time, or cut expenses to the bone; in each case not selling any equities during the downturn.

DoubleDown

  • Handlebar Stache
  • *****
  • Posts: 2075
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #32 on: January 28, 2015, 11:18:52 AM »
I am wondering about SWR withdrawal rate too. Stock market fluctuates, so what would people do in a bear stock market scenario?
 For example, if you have a million dollars in your portfolio, and you are withdrawing 4%...so, $40K. So if there is a huge stock market downturn and your portfolio is down to $700k this year. Then do you withdraw $28K this year?  For example, you normally need 40K to cover expenses, in a bear market, you tighten your belt (let's say, no more traveling or selling a car) and you still find yourself needing 32K for expenses. Do you cheat yourself and withdraw extra 4K? (so you are withdrawing 4.57% that year), or should you have a separate emergency/bear Market funds account (for example $10K in a savings account)?

The 4% literature accounts for this generally -- however, there is a higher failure rate if that bear market occurs in the early part of your retirement (see "sequence of returns risk"), which is why many in this forum intend to either keep cash on the side to cover a 1-3 year downturn, or work part-time, or cut expenses to the bone; in each case not selling any equities during the downturn.

Great summary, dude, on ways to mitigate sequence of returns risk.

I'll add that one doesn't even need to completely avoid selling equities -- doing any of the things you mentioned just to reduce the amount withdrawn can completely overcome the risk. For example, going to a 2% withdrawal in a down year (and making up the other 2% with part time work, cutting expenses, or saved cash) would be a pretty much bulletproof approach in a downturn.

Another mitigation is to get a pension, social security, or purchase an annuity to cover basic expenses. Then the checks keep coming no matter what the markets are doing. This is likely less optimal from strictly a rate-of-return view, but gives you all kinds of security. I'll take security over having my portfolio invested 100% in stocks.

So far I've been fortunate -- first year of ER had net worth going up quite a bit (an extra 2.5 years of living expenses), giving us a larger buffer to handle down years in the future. I don't ever want to have to go back to work (even part time) or cut expenses to make it.

nereo

  • Senior Mustachian
  • ********
  • Posts: 17496
  • Location: Just south of Canada
    • Here's how you can support science today:
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #33 on: January 28, 2015, 11:55:44 AM »
I'll take security over having my portfolio invested 100% in stocks.

So far I've been fortunate -- first year of ER had net worth going up quite a bit (an extra 2.5 years of living expenses), giving us a larger buffer to handle down years in the future. I don't ever want to have to go back to work (even part time) or cut expenses to make it.
Congrats on your first year of ER.  If you don't like the volatility of 100% stocks and your portfolio has increased by an extra 2.5 years of living expenses, why not take those gains and shuttle them aside in a 'safe' account and earmark them for the next recession? Odds are it'll happen sometime in the next 0-4 years.  Just a thought....

Capsu78

  • Pencil Stache
  • ****
  • Posts: 765
  • Location: Chicagoland
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #34 on: January 28, 2015, 12:57:33 PM »
Dude,
Good article on LTC perspectives and reality.  Thanks for sharing.

DoubleDown

  • Handlebar Stache
  • *****
  • Posts: 2075
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #35 on: January 28, 2015, 01:42:17 PM »
I'll take security over having my portfolio invested 100% in stocks.

So far I've been fortunate -- first year of ER had net worth going up quite a bit (an extra 2.5 years of living expenses), giving us a larger buffer to handle down years in the future. I don't ever want to have to go back to work (even part time) or cut expenses to make it.
Congrats on your first year of ER.  If you don't like the volatility of 100% stocks and your portfolio has increased by an extra 2.5 years of living expenses, why not take those gains and shuttle them aside in a 'safe' account and earmark them for the next recession? Odds are it'll happen sometime in the next 0-4 years.  Just a thought....

Yes, a very good thought and thanks for that! I'm actually already in a balanced AA, I didn't mean to imply I'm invested in 100% stocks or that was where the gains came from (it came from overall gains in stocks/bonds, real estate, etc.). You are also very right that a downturn is bound to happen, and for that reason I have not increased my spending $.01 but instead plan to do exactly as you've said -- keep the funds as a buffer for the inevitable down years.

I feel blessed that I didn't have to contend in my first year of ER with a 2008/09 situation, and contemplating riding it out according to Trinity Study predictions that everything will be all right, in the long run (intellectually I know this, but glad I didn't have to put it to the test the first year).

RetiredAt63

  • CMTO 2023 Attendees
  • Senior Mustachian
  • *
  • Posts: 20742
  • Location: Eastern Ontario, Canada
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #36 on: January 28, 2015, 04:33:44 PM »
Semi-theoretical question.  Background - I haven't worried too much about the SWR because most of my retirement income is pension, and a chunk of the rest is a RRIF, which has a mandated minimum based on the value on January 1 of that year.  The mandated minimum goes up every year so at 94 I have to take out 20% - it is set up that way.  4% is the minimum at age 65, by the way.

So - the rest of my income is conservative income-oriented mutual funds and a REIT - and my cash-in is zero.  I am living on dividends.  Is the 4% (I know, 7% yield less 3% inflation) based on selling things or taking out dividends?  Barring extra expenses I could die (30 years down the road) and still leave all my stocks for my estate.  And as long as a fund has productive companies, I don't really care what the stock value is, I am not selling it anyway.

seattlecyclone

  • Walrus Stache
  • *******
  • Posts: 7254
  • Age: 39
  • Location: Seattle, WA
    • My blog
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #37 on: January 28, 2015, 04:41:16 PM »
Is the 4% (I know, 7% yield less 3% inflation) based on selling things or taking out dividends?

Either. Both. It doesn't matter. If your 4% number is $30k and you have $25k of dividends, you can spend those dividends and sell $5k of shares to make up the difference. Alternatively you could reinvest the $25k dividends in an asset class that you don't have enough of (according to your desired asset allocation) and sell $30k of some fund you have too much of. It's six of one, half a dozen of the other.

Minion

  • Stubble
  • **
  • Posts: 210
  • Location: Brisbane -> London -> Berlin
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #38 on: January 28, 2015, 04:48:33 PM »
Another point is that 4% works for the US past performance. It doesn't work for most other countries past performance - for Australia it is 3.6%. Given that 4% is possibly a maximum, that may not hold for the US or any other country in the future, it is reasonable to have a lower SWR.

The last 8 years of life are the most expensive according to multiple things I have read, so you may wish to think about the funds for them as a separate option.

Good point on Australian past perfomance :)

MDM

  • Senior Mustachian
  • ********
  • Posts: 11477
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #39 on: January 28, 2015, 04:57:12 PM »
Semi-theoretical question.  Background - I haven't worried too much about the SWR because most of my retirement income is pension, and a chunk of the rest is a RRIF, which has a mandated minimum based on the value on January 1 of that year.  The mandated minimum goes up every year so at 94 I have to take out 20% - it is set up that way.  4% is the minimum at age 65, by the way.
So - the rest of my income is conservative income-oriented mutual funds and a REIT - and my cash-in is zero.  I am living on dividends.
You should subtract any guaranteed payments from your annual expenses and use the difference in any SWR-related calculations.

Quote
Is the 4% (I know, 7% yield less 3% inflation)...
Some times we "know" things that just aren't so - this is one example.  Yes, 7 - 3 = 4, just as 1 + 1 = 2, but that is not how the Trinity Study authors (and related researchers) determined that 4% was a "Safe" Withdrawal Rate.

Quote
...based on selling things or taking out dividends?
Yes.  Either.  Both.*  Whatever it takes to meet your annual expense needs.

Quote
Barring extra expenses I could die (30 years down the road) and still leave all my stocks for my estate.  And as long as a fund has productive companies, I don't really care what the stock value is, I am not selling it anyway.
Makes sense.  Enjoy!

*And I wrote this before seeing seattlecyclone's post.

RetiredAt63

  • CMTO 2023 Attendees
  • Senior Mustachian
  • *
  • Posts: 20742
  • Location: Eastern Ontario, Canada
Re: Could someone explain the SWR (safe withdrawal rate) in a bit more detail?
« Reply #40 on: January 28, 2015, 05:19:25 PM »
Thanks for the replies.  I think taxes play a part here (Canada) - I pay less income tax on a dividend than the same amount of capital gain.  So net a dividend is better - and the mutual fund will make me more money next year.  Of course if I wanted to re-balance I could start reinvesting dividends from one source and not from another.  However, pension and related (i.e. RRIF, CPP) are about 85% of income, investments are only about 15%, so not a lot of adjustment needed.

I guess I am in the fortunate position that my pensions and RIFF and dividends cover my expenses, which is what we are all aiming for.  I have gone the other MMM route, instead of increasing income I have cut expenses.  The divorce is done, no more lawyers' fees.   Can you see my huge grin on that one?  Don't ask how much it cost, I really do not want to do the math there. Car loan (0% interest) has two more months to go, on a car that I expect to keep another 5 years.  I have money tucked away for the roof re-shingling I need to do this summer.  The garden gets bigger and more productive every year. My $50 library membership more than covers the cost of books I would otherwise buy, even if they were second hand.  And so on.