Author Topic: Safe Withdrawal Rates: Safer Than You'd Think  (Read 9421 times)

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Safe Withdrawal Rates: Safer Than You'd Think
« on: August 20, 2012, 08:52:26 PM »
So I've always been in the camp that one planning for a sufficiently long retirement should shoot for a 3% SWR.  The traditional advice of 4% is based on the assumption of a 65 year-old shooting for a 30 year retirement.  (My personal SWR will be a bit wonky, as a lot will be based on rental income, rather than a traditional dividend or total returns portfolio, but in general, I'm supportive of a 3% SWR for a traditional portfolio).

Someone who is ERing thirty years younger will almost certainly need it to last quite a bit longer, and will almost certainly hit stretches of bad returns during that retirement.

MMM posted an article on the 4% SWR being good enough (as there are potential fallbacks / safety layers ), but I'd rather play it a bit safer, and not have to rely on any other income, or expenses not changing, or anything like that.

However, this brilliant article was just posted that argues that a 4% SWR may in fact actually be conservative, because it's based on a worst case scenario.  That is, unless we hit the worst that we have, we'll likely be okay with a higher SWR, or 4% should certainly be safe enough.  In fact, as the article points out, since the 4% SWR is based on a worst-case type scenario, "the safe withdrawal rate actually has a 96% probability of leaving more than 100% of the original starting principal!"  That is, almost all of the time (96 out of 100 times, historically), you will end those 30 years with more money than you started.

Okay, so that helps understand why the current 4% SWR figure feels pretty safe.  But what about today?  This article puts current "worst return" scenarios that generate that 4% SWR figure into current context (i.e. what would have to happen for the next 15 years for a 4% SWR to not be safe)..  So for those bears who think we're in for a new normal of low returns, that's okay, and may even allow for a higher SWR than you're thinking!

Quote
Given this reality, it's notable that merely getting 'new normal' returns of low single digits would actually represent not a risk to historical safe withdrawal rates; it would actually be an upside surprise that would result in materially higher lifetime spending!

If you think equities can even return a 1% real return over the next 15 years (i.e. slightly better than flat), then a 4 to 4.5% SWR is fully appropriate! 

I immensely enjoyed the whole article, and figure you all might also:
http://www.kitces.com/blog/archives/387-What-Returns-Are-Safe-Withdrawal-Rates-REALLY-Based-Upon.html
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.

Lars

  • Stubble
  • **
  • Posts: 105
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #1 on: August 20, 2012, 11:08:01 PM »
Thanks for the link to an interesting article.

As I read it two questions leaped out at me. What does your portfolio look like at the end of 15 years of 1% real returns? What returns does he have in mind after 15 years? Best case with consistent returns and withdrawals of 4% of remaining balance per years would yields a balance and yearly spending down to two thirds of the starting balance in real dollars (or (1.01-0.04)^15) With withdrawals of 4% of initial balance adjusted for inflation and a monte carlo analysis using the reasonable value for the historical variation in returns (10%), the average portfolio value is half the starting value after adjusting for inflation. Continue for thirty years and you run out of money about half the time. In order for the author to not be concerned with the drops in portfolio values, he must be assuming returns going forward after 15 years return to historical averages. Either his time frame for retirement is short (<20 years) or he forgot to mention a big assumption.

Personally I find the studies on international safe withdrawal rates and the higher than average growth of US compared to the world in past that are very unlikely to continue to be the most convincing arguments for 3% SWR and this article didn't change that. Unless withdrawals are done like MMM, then 4% makes perfect sense.   

tooqk4u22

  • Magnum Stache
  • ******
  • Posts: 2832
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #2 on: August 22, 2012, 08:05:53 AM »
Interesting article and perspective on SWR, but I disagree with it in general but especially for FI. 

The main issue is that the conclusions are based on average initial SWR and a 30year period (i.e. assumes that you deplete principal fully after 30 years). One, it ignores the change in asset prices/returns in subsequent years and two FI people will need the money/income longer than 30 years.

Also think about this assuming the following:
- $1,000,000 initial investment.
- 1% real return in your investments (i.e. 4% nominal return).
- 30 years like in article

This equals $38,748 payment in real terms over 30 years - very close to 4% SWR but you burn through principal - so great if you only need to live 30 years.  If you had to live 50 years it would be $25k or a 2.5% SWR.  Some will say that 4% return is two low but in retirement even if you are living longer you should still have a more conservative bias and if the expected stock market returns are in the 5-7% range right now (for the near and medium term anyway) then a more conservative mix will yield lower. 

If you don't want to deplete principle you need a 4% SWR in real terms (or 7% in nominal), which is what MMM assumes.   

Although the closer you get to your life expectancy the more you can use principle to live on.

Also you shouldn't bank on upside surprise like the article implies.....this is basically why every pension plan in the US is bankrupt now.  Assume a reasonable rate and adjusted when things are better but if corresepondingly you will have to adjust when things get worse too.
« Last Edit: August 22, 2012, 08:09:46 AM by tooqk4u22 »

timstobbs

  • 5 O'Clock Shadow
  • *
  • Posts: 5
  • Location: SK, Canada
    • Canadian Dream: Free at 45
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #3 on: August 28, 2012, 01:00:26 PM »
I can't recall the name, but an engineer in Canada did some excellent work looking at those stats for the 4% SWR and came to the conclusion that while 4% is 'safe', 3% is basically bullet proof.  There are as always a few particulars to consider in saying things like that. 

The first is the fact that your first five years of withdrawals are the most risky.  Hitting a bad market for that period can terminate just about any portfolio before someone dies.  The rule of thumb here is if you start to use your principle at all in the first five years you have a risk of running out of money.

The solution around this is really about risk management and your personal degree of tolerence on risk.  Obvious solutions are keep working part time for your first 2 to 3 year of FI to ensure you don't touch your principle.  Don't include your house equity in your FI plans and keep it as a back up.  Or just keep a slush fund of 2 to 3 years expenses beyond your FI plan to ensure you don't take too much out during the low years.

In the end, find what works for you and plan roughly for 4% SWR.  You can end up playing with what numbers work for you based on the rest of your risk management plan (ie: is you have no backup plans plan for 3%SWR, if you have five backup plans you might be able to handle a 5% SWR).

Tim

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #4 on: August 28, 2012, 09:10:05 PM »
Did you guys fully read the article in the OP?

I'll reply more when I have some time (first week of school, pretty busy), but I think you guys really missed the point.

In a nutshell: 4% SWR is a worst case scenario, not the standard that we'be all been thinking.  It would have you end up with more money than you started almost every time, and almost ALWAYS leave you with some money (I.e. not fail).  Unless you're projecting the next 20 years to be the worst years ever (even after the last 12 years?), 4% is much safer than thought.
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.

velocistar237

  • Handlebar Stache
  • *****
  • Posts: 1424
  • Location: Metro Boston
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #5 on: August 29, 2012, 09:09:37 AM »
On FIRECalc, a 4% withdrawal rate (i.e., 4% of initial portfolio, inflation adjusted) gave a 95% success rate after 30 years with a 60/40 stock/bond split. In that case, 78% of cases ended up with final balances higher than the initial balance. With the more MPT-type portfolio option, the success rate was 100%, and about 95% of cases resulted in a higher final balance.

I don't know the details of Kitces's analysis, but there are some common portfolios that don't quite live up to it. In the long-term view, for a 60/40 lazy portfolio, the highest withdrawal rate to yield 100% success after 60 years was about 3.3%.

If someone wanted out at 5%, though, it would probably work with no problem. If they hit one of the rarer low-SWR scenarios, they could still make it work. MMM followers are prepared in so many different ways that a 100%-success SWR doesn't have to be one of them.

I'm interested in this correlation between SWR and the returns on the first 15 years. That seems to suggest that 15 years into my retirement, I'll be able to gauge the financial health of the rest of my retirement, or more actively, if I make sure that my portfolio balance at 15 years is high, then I'll be set. Then again, Kitces uses this 15-year return to gauge success over 30 years, and the correlation might go away for longer retirements.

Nords

  • Magnum Stache
  • ******
  • Posts: 3421
  • Age: 63
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #6 on: August 29, 2012, 11:01:26 AM »
Did you guys fully read the article in the OP?
I'll reply more when I have some time (first week of school, pretty busy), but I think you guys really missed the point.
That problem tends to be self-correcting...

People either read the article and take the steps to achieve FI earlier and have more flexibility to "quit the job" sooner.

Or they stay at work longer, pay more into Social Security/Medicare, and eventually achieve FI anyway.

In the meantime we can just keep leading people to the research and pointing out that there are better ways to eliminate that nagging 5% failure possibility.

timstobbs

  • 5 O'Clock Shadow
  • *
  • Posts: 5
  • Location: SK, Canada
    • Canadian Dream: Free at 45
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #7 on: August 29, 2012, 12:27:21 PM »
Did you guys fully read the article in the OP?

Yes I read it.  I just have a problem taking a mathematical conclusion based on 30 year periods and expanding it to 50 years for my retirement plan and assuming the conclusion is still valid, which is why I take a fair bit of caution with articles like this.   

I understand the 4% SWR takes in the worst of the past, but as velocistar237 points out to get to 100% for a 60 year timeline that rate still drops to 3.3%. 

In the end, I agree with Nords...getting that last 5% of risk out of the way can be done several different ways, which was the point of my post.  Find what works for you and go with it, but a good default SWR is 4%.

Lars

  • Stubble
  • **
  • Posts: 105
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #8 on: August 31, 2012, 10:07:16 PM »
Did you guys fully read the article in the OP?

I'll reply more when I have some time (first week of school, pretty busy), but I think you guys really missed the point.

In a nutshell: 4% SWR is a worst case scenario, not the standard that we'be all been thinking.  It would have you end up with more money than you started almost every time, and almost ALWAYS leave you with some money (I.e. not fail).  Unless you're projecting the next 20 years to be the worst years ever (even after the last 12 years?), 4% is much safer than thought.
The 4% SWR is a worst case scenario but it is a worst case scenario primarily based on historical data for a single county, the US, over a time period where its share of the world capital markets more than doubled (therefore ~1% higher than average returns) and its volatility/variations in returns were lower than average. With the US already around half of total world market capitalization, I don't think we should count on continued above average returns. This paper (http://ideas.repec.org/p/ngi/dpaper/10-12.html) estimates SWR for 17 developed country. The worst case SWR for 30 years vary from 4.4% for Canada to 0.5% for Japan. However, 4% did work at least 95% of the time in half the countries. However, 4% SWR failed in 3 years  in Japan starting in 1940. So 1% real returns is down right rosy compared to sometimes and places. That and a related paper really hammered home the value of redundancy in a retirement plan.

The

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 28444
  • Age: -997
  • Location: Seattle, WA
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #9 on: October 30, 2012, 09:03:19 AM »
This recent thread from the early retirement forums has a good SWR discussion:
http://www.early-retirement.org/forums/f28/article-on-safe-withdrawal-rates-63477.html

Two posts that help explain the OP (of this thread):
Quote
most folks dont have a clue what the 4% rule even represents. on another forum you and i frequent they are so ignorant of these things that they argue all the time with me about how risky equities are.

i cant get through to them that the 4% rule doesnt represent big gains or even average gains. it represents the worst real returns history ever threw at retirees .

if anything like you said its toooo conservative and to much money usually goes unspent if you follow it to the letter of the law.

the answer the mis-informed give you all the time is "oh im not gambling in equities in retirement " and so they sit in the bank and get zero and complain.

they dont realize they are picking the riskiest investment there is as cash has failed more ofton then anything else.

as michael kitces wrote ,things would have to get to the point that you got under a 2% real return for 15 years to do damage and get that return for 30 years to totaly fail before the 4% rule is in danger.

thats a whole lot worse then now thats for sure.

about the worst group now may be those who retired in 2000 depending on allocations as they are facing the perfect storm.

they retired at a time of low rates,low dividends ,poor market performance and a poor outlook based on valuations.

its still to early though to even pass judgement on them as the danger point is around 15 years in.

i havent retired yet so im in better shape than those in 2000 as there is already just about 13 years of history behind me of crappy years so perhaps i will be closer to an improvement in 2 years when i retire. .

as it looks now ill only need about a 2% withdrawal rate because we have other income so im going to be one of those who break the rule and maintain very low equities allocations because i can meet goal .

but its all going to be dynamic and fine tuned under battlefield conditions as time goes in retirement.

And the counter argument about other countries, what Lars was espousing above:
Quote
Wade Pfau wrote this article on safe withdrawal rates in other countries:
An International Perspective on Safe Withdrawal Rates: The Demise of the 4 Percent Rule?

The safe withdrawal rate for Japan (table 3) was 0.47%!!!!

Note that the only safe conclusion from this depressive study is: work until you die!

I'd argue the title of this thread is correct.  Unless you think we'll see 15 more years of no real growth, the 4% rule is quite safe.  I grant that is possible, but I'd say not more likely than not. 

Anyways, that was a thread with some good discussion I wanted to cross reference here for anyone interested.
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.

Nords

  • Magnum Stache
  • ******
  • Posts: 3421
  • Age: 63
  • Location: Oahu
    • Military Retirement & Financial Independence blog
Re: Safe Withdrawal Rates: Safer Than You'd Think
« Reply #10 on: October 30, 2012, 09:32:09 AM »
In the end, I agree with Nords...getting that last 5% of risk out of the way can be done several different ways, which was the point of my post.  Find what works for you and go with it, but a good default SWR is 4%.
Let me further clarify that "getting that last 5% risk out of the way" can also be done in ER.  You don't have to stay employed waiting for a rotating green light to start flashing.

The 4% SWR has its issues that keep it from fully modeling reality-- no argument there.  But it works well enough, and it works great as a tripwire.  When your assets (and your spending) reach the conditions of the 4% SWR then you're good to go.  Put in your resignation.  Even if you're only 25 years old.  Seriously.

You have plenty of time during ER to optimize your plan.  If you're worried about longevity then annuitize a portion of your portfolio (in addition to U.S. Social Security, if applicable).  Unlike the 4% SWR plan, most humans will be psychologically unable to blissfully raise their spending every year at the rate of inflation-- especially when the market drops.  Vary your spending by cutting it during bad years (Bob Clyatt's 4%/95% rule is a good guideline here) and raise it when the economy recovers.  Diversify your portfolio for some dividend income or run a rental property (or buy REIT shares).  If you have a budget surplus then either reinvest it or add it to your cash stash for a new roof or a luxury vacation.  Turn a hobby in to a revenue stream or take a part time job if it interests you. 

The key is to watch your portfolio withdrawals during that first 5-10 years.  One or two years may be 5%-6% instead of the 4% SWR, but if at the end of 10 years you can still hack it on 4-5% then you're almost certainly going to be fine.  After that point your "1% failure scenario" probably involves nuclear weapons, hyperinflation, or alien invasions.  Or all three. 

The key points are that the 4% SWR is a tripwire, and during the first 5-10 years you'll be able to address any nagging concerns to raise success from 95% to 99%.  By that point you've optimized the problem-- stop fretting and go enjoy your life.