Author Topic: Can someone explain SWR (safe withdrawal rate) to me in the context of this post  (Read 6903 times)


  • Bristles
  • ***
  • Posts: 291
I was reading this on early retirement extreme and wasn't sure how this rate was calculated.

Also, what does the acronym DW stand for?


  • 5 O'Clock Shadow
  • *
  • Posts: 13
"DW" typically means "Dear Wife."  Just a nice way to refer to your spouse.  You may also find "DH" which is "Dear Husband" as opposed to Designated Hitter.

It looked like the rates are simply what those people are (or would be) withdrawing from their portfolios each year.  For example, if I say I had a 4% SWR this year it means I needed to take out 4% of my portfolio to meet my expenses.


  • Bristles
  • ***
  • Posts: 291
ah so maybe those were people who already retired early and were talking about their withdrawal rates.

Or they could have been acting as if they were retired, based on net assets, and money spent that year, what their SWR would have been had they been retired.  Probably the latter I'm thinking.


  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28442
  • Age: -999
  • Location: Seattle, WA
That would be calculated as "your spending at that time divided by your portfolio value at that time" - as spending decreases and portfolio increases, SWR drops.  Most would argue you're FI at around 4% and can FIRE whenever you're at a SWR you feel comfortable with (4% for some, 3% for many with a sufficiently long ER time period, etc.. See many discussions about stability of SWRs).
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 Check out the Now page to see what I'm up to currently.


  • Magnum Stache
  • ******
  • Posts: 3371
  • Age: 62
  • Location: Oahu
    • Military Retirement & Financial Independence blog
ah so maybe those were people who already retired early and were talking about their withdrawal rates.
Or they could have been acting as if they were retired, based on net assets, and money spent that year, what their SWR would have been had they been retired.  Probably the latter I'm thinking.
To be excruciatingly technically correct, they're talking about what their annual withdrawal rates would have been had they retired that year.

The concept of the "safe withdrawal rate" comes from Bengen's 1994 "Safemax" study and the 1998 Trinity study-- that's the SWR.  Retirement spending starts at the SWR (4%) and raises that spending each year with inflation.

The withdrawal rates in that ERE thread are "just" an annual withdrawal rate based on that year's portfolio value, and it's not necessarily adjusted for inflation.  If you use the 4% SWR then you'll probably have enough money, but you could run out of principal.  However if you use a plain ol' 4% annual withdrawal rate then you'll never run out of principal.  You might not have enough money to live on, but you'll still have 96% of the portfolio left.

You'll run across this accidental terminology confusion in Bob Clyatt's "Work Less, Live More".  He's one of the first to model a variable retirement spending model.  His portfolio survivability study assumed that you could withdraw 4% of that year's portfolio value for each year that the market was up.  Ideally your portfolio would rise enough each year to keep up with inflation, so a constant 4% annual withdrawal initially worked out to about the same number of dollars as a 4% SWR. 

However Bob's system worked even better in down markets.  If the market was down, then your next 4% withdrawal would drop just as much as the market did.  This could be quite unpleasant alongside 2008-09, when a straight 4% withdrawal meant that you would have had to reduce your annual spending by ~37%.  Bob's variable spending algorithm said that if the market was down then you could spend the greater of <that year's 4% withdrawal> or <95% of last year's withdrawal amount>.  If the market dropped more than 5% then you went with 95% of last year's withdrawal amount, which meant that this year's annual withdrawal percentage would be higher than 4% of your (smaller) portfolio.  In the worst cases of his computer analysis (same FIRECalc data) it could get as high as 5-6% for a year or two.

The computer models showed that when retirees used this 4%/95% variable withdrawal system, the overall 30-year Bengen/Trinity SWR of their portfolio was about 4.5%. 

So despite all the doom & gloom about the 4% SWR being dead because future returns will be lower, the reality is that the SWR models don't account for variable withdrawals or annuity income (Social Security).  That's why when your annual withdrawal rate gets down to 4%, even if you're only in your 30s you're probably still good to go.

Spouse and I were agonizing over this issue in 2000, back before Bob's book came out.  If I retired at age 41 in 2002 as planned, my 4% SWR would have to last for a lot longer than 30 years.  However the end of the NASDAQ bull market injected a new note of reality into this angst, and we didn't really have a good handle on the probabilities. 

Then the markets closed for the week after 9/11.  When they re-opened, we watched our portfolio melt like an ice cube on a concrete sidewalk in August.  At the end of the day I ran the "new" numbers through FIRECalc and Financial Engines.  Both said that the portfolio would survive for 30 years (over 95%), and neither one took Social Security into consideration.  By mid-2002 our portfolio was well on the way to recovery and I retired.

2008-09 was a pretty intense gut check.  (At one point we were 56% off the 2007 highs, which were stupidly high.)  The interesting aspect of our 2009 portfolio low was that it was still higher (adjusted for inflation) than our 9/11 lows.  FIRECalc and Financial Engines still gave us a >95% success rate, but this was starting from 2009 instead of 2001.  So we stopped worrying about it.

I wouldn't recommend this double-recession gut-check method, but if your portfolio survives the first 10 years on a 4% SWR system (let alone Bob's 4%/95% system) then you're probably going to be fine.  The key is the very human tendency to reduce spending during recessions, no matter what FIRECalc or any other calculator claims you can do.

Or you could continue to work until you reach some of the ridiculously low withdrawal rates in that ERE thread.  Unless you absolutely love your occupation, though, you pay a high price to guarantee a 100% retirement portfolio success rate.

It could be called the "Groucho Marx" portfolio:
Interviewer:  "Mr. Marx, what are you going to live on during retirement?"
Marx:  "I've invested all my money in U.S. Treasuries."
Interviewer:  "Nobody can live on just the income from U.S. Treasuries!"
Marx:  "You can if you have enough of them."

Joe Dominguez (of Your Money or Your Life) made 100% Treasuries work from the 1960s up until 1997, but it did not keep up with inflation.  Joe's spending for the last few years of his life made Jacob's ERE budget look like a wild-eyed hard-partyin' Paris Hilton lifestyle.