Author Topic: Instead of SWR how about a SSR?  (Read 3864 times)

matchewed

  • Magnum Stache
  • ******
  • Posts: 4313
  • Location: CT
Instead of SWR how about a SSR?
« on: August 07, 2013, 02:09:45 PM »
Great article by Pfau. Basically a more technical version of the "your savings rate is the best determination of FIRE."

http://www.fpanet.org/journal/currentissue/tableofcontents/safesavingsrates/

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 27552
  • Age: -999
  • Location: Traveling the World
Re: Instead of SWR how about a SSR?
« Reply #1 on: August 07, 2013, 07:06:33 PM »
Yeah, that's a good article for people planning to work a long time or with lower savings rates.  For Mustachians who save 70%+, not as useful.

Table 1 is a good example:


It also boils down to "you need to save this much (to reach a desired SWR)" so it doesn't eliminate the SWR problems.

Overall a good article, but more useful for a "regular" early retirement (age 55-60, rather than 65) than what many here are planning.

I think it's good for one starting out.  Later on one has to muck into SWRs, because just knowing "okay, I saved X% for 20 years, I should be good?" isn't enough, especially since savings rates fluctuate.
We are two former teachers who accumulated a bunch of real estate, retired at 29, and now travel the world full time with two kids.
If you want to know more about me, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8428
  • Registered member
Re: Instead of SWR how about a SSR?
« Reply #2 on: August 07, 2013, 09:57:50 PM »
Isn't this just "shockingly simple math" in reverse?  You plug in years and it gives you a savings rate, instead of the other way around.

yahui168

  • 5 O'Clock Shadow
  • *
  • Posts: 92
  • Age: 42
  • Location: CA
Re: Instead of SWR how about a SSR?
« Reply #3 on: August 07, 2013, 10:39:46 PM »
I don't quite like advice that says "save x% for 30 years" because income can range drastically over such a long period of time. And as arebelspy pointed out, savings rate fluctuate as well. This makes the math way more complicated.

dragoncar

  • Walrus Stache
  • *******
  • Posts: 8428
  • Registered member
Re: Instead of SWR how about a SSR?
« Reply #4 on: August 07, 2013, 11:46:25 PM »
I don't quite like advice that says "save x% for 30 years" because income can range drastically over such a long period of time. And as arebelspy pointed out, savings rate fluctuate as well. This makes the math way more complicated.

Absolutely.  I once ran the numbers with these rough parameters:  50% initial savings rate, 5% (real) annual raise, 5% (real) annual returns on savings. 

If you maintained the 50% savings rate (i.e., you inflated your lifestyle with salary), it takes 25 years to reach a 4% SWR.

If you kept your original expense rate (i.e., no lifestyle inflation), it only took 15 years to reach the 4% SWR.  By that time you were saving 71%.

Obviously, the parameters matter.  But this lifestyle inflation has a large impact no matter how you slice it.

Edit:  Even at a higher rate, say 80% initial savings rate (which is my personal savings rate), lifestyle inflation can increase your working years from 5 to 7 -- a 40% increase. 

Unfortunately, I had been inflating my lifestyle as my salary increased (i.e., increased my spending by $20 for every extra $100 I earned).
« Last Edit: August 07, 2013, 11:49:36 PM by dragoncar »

skyrefuge

  • Handlebar Stache
  • *****
  • Posts: 1007
  • Location: Suburban Chicago, IL
Re: Instead of SWR how about a SSR?
« Reply #5 on: August 08, 2013, 08:23:51 AM »
Isn't this just "shockingly simple math" in reverse?  You plug in years and it gives you a savings rate, instead of the other way around.

Sort of, but the main result of this research is to reveal information about long-term market effects, and I see it only peripherally connected to savings rates and withdrawal rates.

The main conclusion is that over a long enough period, there is a "reversion to mean" on investment returns. It's saying that someone who has relatively shitty returns over their 30 year working career is still likely to be ok in their 30 year retirement (even though their withdrawal rate might be much higher than 4%), because those second 30 years are likely to produce better-than-normal returns.

If an addendum was made to "The Shockingly Simple Math", it would say something like this: "After you've saved X% for the Y years prescribed by the Shockingly Simple Math, your stash may be much smaller or much larger than you initially expected it to be, due to an investment environment that was different from the assumed average. But don't worry about that final number, or working another year. Just retire and withdraw the dollar amount you originally expected to, and it'll all work out!

If you retire with a smaller-than-expected stash, your withdrawal rate will be greater than 4%, but that's fine because you're gonna get kickass investment returns throughout retirement. If you retire with a bigger-than-expected stash, your withdrawal rate will be less than 4%, but don't get greedy and increase it to 4%, because you're going to get some shitty returns going forward."

Of course it seems a bit risky to wholly rely on that "reversion to mean" (especially in the case of early retirement where the working period covers a rather short amount of time), but it's interesting to see the data show that it basically would have worked in the past for normal retirement timelines.

It's why I don't actually feel much more "FI" now than I did a year ago, even though the numerical size of my stash is some 30% larger. A "reversion to mean" could quickly return my stash to the size it was a year ago, and if it does, I similarly won't feel much less "FI" (I hope!)

It also helps explain that people who have such specific FI numbers/dates aren't quite as misled as I thought they were. It used to baffle me how someone can declare FI in, say, June of 2013, given how much market volatility sloshes a stash around (especially when it's near FI-size). But this study essentially says that the sloshing can be ignored, and as long as you did the prescribed savings it'll all work out.
« Last Edit: August 08, 2013, 08:32:57 AM by skyrefuge »

mpbaker22

  • Handlebar Stache
  • *****
  • Posts: 1095
Re: Instead of SWR how about a SSR?
« Reply #6 on: August 08, 2013, 10:26:34 AM »
Even with the shockingly simple math, I don't plan out my x% savings rate to retire in y years.  I have figured out what I need to retire on a 3-4% SWR, and I'm just waiting to get there.  Yes, I have an idea of how long it would take, but there are so many factors, it doesn't make sense to pretend that will be the actual time.

Skyrefuge - good summation.  I view it as a 4% SWR is almost guaranteed to last 30 years.  But if you take it a step further and look at the end of 30 year periods where investment returns were relatively poor, the SWR would actually be higher.  The 4% SWR is based on all investment time periods ... IE it treats starting your retirement on Jan 1. 2008 the same as starting on Jan 2010.

Nords

  • Magnum Stache
  • ******
  • Posts: 3167
  • Age: 58
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Instead of SWR how about a SSR?
« Reply #7 on: August 10, 2013, 10:47:33 PM »
I don't quite like advice that says "save x% for 30 years" because income can range drastically over such a long period of time. And as arebelspy pointed out, savings rate fluctuate as well. This makes the math way more complicated.
We've been hearing the exact same complaints about the 4% SWR for years, and yet people manage to ER anyway. 

The advantage of the "save x%" or "spend x%" advice is that there's a margin of safety for variable savings rates, variable investment returns, and variable spending.