Are Even 3% Withdrawal Rates Excessive for Retirees Today?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Leesbro63
Posts: 10638
Joined: Mon Nov 08, 2010 3:36 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Leesbro63 »

bornloser wrote:Got one colleague in a casket for viewing tonight (age 72), one colleague in the ICU (age 69) and a partner that passed last February at 66. All the hoopla about planning to have enough retirement for age 95 is really overkill, IMHO. And if you do not plan to leave an estate, I will remind you that the poor do pretty well in this nation. If I make it to age 90, will I really care if I have to ask for a government handout?
What about surviving spouses? Do those three have them?

Asked differently: What are the odds of a 65 year old couple having one spouse survive to 85? 90? 95?
Firewood42
Posts: 125
Joined: Fri May 11, 2012 7:06 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Firewood42 »

Here is a simple question.Lets say I have $400,000 invested in Cd's that pay 2 percent. I am 70 years old. If I withdraw 3 percent per year from this portfolio, how many years will it last? I understand they are 7 year Cd's but there are ways to withdraw 3 per cent with laddering etc. How about if I withdraw 5 percent?
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

A message from the Carribbean

Post by Taylor Larimore »

Bogleheads:

Nisi quoted me correctly:
Taylor Larimore says "We simply withdrew what we needed and kept an eye on our portfolio balance," i.e. adjusted spending to portfolio performance on a fairly short-term basis.
Our portfolio balance has increased this year so Pat and I decided to enjoy a Holland-America cruise to the Carribbean.

This message was sent from an internet cafe in the Virgin Islands (the ship charges .40 cents a minute to use their internet).

Keeping life simple.

Best wishes
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
Mitchell777
Posts: 1278
Joined: Mon May 21, 2007 6:32 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Mitchell777 »

bornloser wrote:
CyberBob wrote:There is an interesting Scott Burns article out recently that looks at the issue from a totally different angle; 30-year planning horizons bump up against the reality that you may expire long before your portfolio does.
Scott Burns wrote:The bottom line, however, is that death reduces the risk of running out of money more than does portfolio management.
Life, Death and How Long Your Money Will Last.

+1. I think two very important variables are best guess at the age you are going to check out (an educated estimate based on family, lifestyle, etc) and if you want to leave an estate. Got one colleague in a casket for viewing tonight (age 72), one colleague in the ICU (age 69) and a partner that passed last February at 66. All the hoopla about planning to have enough retirement for age 95 is really overkill, IMHO. And if you do not plan to leave an estate, I will remind you that the poor do pretty well in this nation. If I make it to age 90, will I really care if I have to ask for a government handout?
This hits home with me. In the past 2 weeks I learned of the deaths of 4 former colleagues, two in their 50's. Not that I want to run out of money at any age but, personally, I may be worrying a bit too much about running out at 95 or 100
scone
Posts: 1457
Joined: Wed Jul 11, 2012 4:46 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by scone »

In a sense, people who are wondering whether to plan on a 2% or a 4% withdrawal rate have already won the game, to some extent. The very question implies they have a bag on cash to draw on, beyond Social Security. They are far better off than the millions of people who have virtually no savings, and lots of debt. When those people get to the point when they can no longer work, they will be forced to live on Social Security, and whatever other means they can devise. Apart from the social and moral issues, that lack of spending money has got to feed back into the economy and hence eventually into investments. So indirectly, the best way for we fortunate ones to get a higher SWR, is to encourage other people to save more now. :happy
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore
User avatar
Cut-Throat
Posts: 2011
Joined: Sun Oct 17, 2010 9:46 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Cut-Throat »

scone wrote: Apart from the social and moral issues, that lack of spending money has got to feed back into the economy and hence eventually into investments. So indirectly, the best way for we fortunate ones to get a higher SWR, is to encourage other people to save more now. :happy
Actually William Bernstein argues the direct opposite approach in his Retirement Calculator from Hell. http://www.efficientfrontier.com/ef/103/hell4.htm

"If you want to retire early, what matters is not how much you save, but how much more than everyone else you save. In a world where everyone saves as if they’re going to retire at fifty-five, or even at sixty-five, none can."
Last edited by Cut-Throat on Wed Nov 14, 2012 3:06 pm, edited 1 time in total.
scone
Posts: 1457
Joined: Wed Jul 11, 2012 4:46 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by scone »

Cut-Throat wrote:Actually William Bernstein argues the direct opposite approach in his Retirement Calculator from Hell.
Bernstein is using dramatic irony and satire to make you pay attention to the seriousness of the topic. His literary technique is a bit like Swift's "Modest Proposal." Not meant to be taken too literally. He's not saying we should all hope "everyone else" is poor. He thinks most people will have to work longer, because saving more will not be enough to maintain a certain level of lifestyle. That's his real point.
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore
dkturner
Posts: 1940
Joined: Sun Feb 25, 2007 6:58 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dkturner »

scone wrote:
Cut-Throat wrote:Actually William Bernstein argues the direct opposite approach in his Retirement Calculator from Hell.
Bernstein is using dramatic irony and satire to make you pay attention to the seriousness of the topic. His literary technique is a bit like Swift's "Modest Proposal." Not meant to be taken too literally. He's not saying we should all hope "everyone else" is poor. He thinks most people will have to work longer, because saving more will not be enough to maintain a certain level of lifestyle. That's his real point.
I think Bernstein was thinking along the lines that in the land of the blind the one eyed man is king. :idea:
User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Clearly_Irrational »

baw703916 wrote:The best strategy is to start retirement with a huge pile of money.
Personally, I disagree with the big pile of money strategy. I'd rather own a set of diverse income producing assets. Of course if the pile is huge enough then strategy is somewhat irrelevant.
User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Clearly_Irrational »

Mitchell777 wrote:Interesting. I have often thought about this in my planning. I think I can live with being 95% sure my money will last 40 years. But then when I add to that the chance of my actually living 40 more years, and combine the two, the odds of my running out of money before I die goes to a fraction of 1%. Of course, you still could run out of money
Maybe it's due to my IT background, but the idea of anything less than five 9s assurance of retirement funding seems like a crazy risk to me.
scone
Posts: 1457
Joined: Wed Jul 11, 2012 4:46 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by scone »

dkturner wrote:
scone wrote: I think Bernstein was thinking along the lines that in the land of the blind the one eyed man is king. :idea:
The tall poppy gets cut down. The guy with the big stash can get taxed to the max. Revolution happens. The Great Big Take Away Idea I get from Bernstein is, "don't count on it." Um, where have I heard that before? Oh, wait.... :wink:
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore
User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Clearly_Irrational »

HomerJ wrote:But that worst case scenario guy from 1966 didn't run out of money over 30 years even with a 4% withdrawal rate...
If you're running worst case scenarios you have to include the 1928-1930 data or you're fooling yourself. A good portfolio & strategy should be able to withstand picking the absolute worst time to retire we've yet experienced. Generally, I would want to simulate it through the 10 worst crises of the past 100 years to be sure it would have handled any of them because you can be sure something along those lines will happen again.
Mitchell777
Posts: 1278
Joined: Mon May 21, 2007 6:32 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Mitchell777 »

Clearly_Irrational wrote:
Mitchell777 wrote:Interesting. I have often thought about this in my planning. I think I can live with being 95% sure my money will last 40 years. But then when I add to that the chance of my actually living 40 more years, and combine the two, the odds of my running out of money before I die goes to a fraction of 1%. Of course, you still could run out of money
Maybe it's due to my IT background, but the idea of anything less than five 9s assurance of retirement funding seems like a crazy risk to me.
yes it must be your IT background because no one is going to have five 9's assurance of anything in an uncertain world
User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Clearly_Irrational »

Mitchell777 wrote:yes it must be your IT background because no one is going to have five 9's assurance of anything in an uncertain world
Well, I admit the error bars would be large due to the number of variables, but you can easily create a retirement strategy that has a higher level of certainty than just stocks + bonds. For example, adding an annuity would reduce your chance of running out of money substantially, so would rental properties. Just blythely accepting that you have a 1 in 100 chance of ending up in the poor house sounds like sloppy planning to me.

Even just taking a few minutes with Firecalc would lead you to believe that 4% is too high. A 3.5% rate for 30 years on the other hand has never failed based on all available data.
technovelist
Posts: 3611
Joined: Wed Dec 30, 2009 8:02 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by technovelist »

richard wrote:
555 wrote:A portfolio with 0% real return with a 3.33% inflation indexed withdrawal rate has a 0% chance of failing in 30 years.
If you want a totally safe inflation indexed portfolio, and have more than a nominal amount invested, you're going to need TIPS. TIPS have negative real yields for almost all of the yield curve.

You'll also have a problem under this approach if you live for 31 years.
There is nothing "totally safe" about TIPS. If you want real protection against inflation in a particular currency, don't buy it from someone who has access to a printing press for that currency.
In theory, theory and practice are identical. In practice, they often differ.
Antonio
Posts: 47
Joined: Wed Jun 25, 2008 3:47 pm

Re: A message from the Carribbean

Post by Antonio »

Hi Taylor

Are you come to Puerto Rico? If it is so, please let me to know.

Happy vacation



Taylor Larimore wrote:Bogleheads:

Nisi quoted me correctly:
Taylor Larimore says "We simply withdrew what we needed and kept an eye on our portfolio balance," i.e. adjusted spending to portfolio performance on a fairly short-term basis.
Our portfolio balance has increased this year so Pat and I decided to enjoy a Holland-America cruise to the Carribbean.

This message was sent from an internet cafe in the Virgin Islands (the ship charges .40 cents a minute to use their internet).

Keeping life simple.

Best wishes
Taylor
dkturner
Posts: 1940
Joined: Sun Feb 25, 2007 6:58 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dkturner »

scone wrote:
dkturner wrote:
scone wrote: I think Bernstein was thinking along the lines that in the land of the blind the one eyed man is king. :idea:
The tall poppy gets cut down. The guy with the big stash can get taxed to the max. Revolution happens. The Great Big Take Away Idea I get from Bernstein is, "don't count on it." Um, where have I heard that before? Oh, wait.... :wink:
You should explain to Bill that he doesn't understand.
User avatar
wade
Posts: 612
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by wade »

Clearly_Irrational wrote:
HomerJ wrote:But that worst case scenario guy from 1966 didn't run out of money over 30 years even with a 4% withdrawal rate...
If you're running worst case scenarios you have to include the 1928-1930 data or you're fooling yourself. A good portfolio & strategy should be able to withstand picking the absolute worst time to retire we've yet experienced. Generally, I would want to simulate it through the 10 worst crises of the past 100 years to be sure it would have handled any of them because you can be sure something along those lines will happen again.
1966 was worse than the Great Depression, which actually wasn't such a terrible time to retire (speaking only from the sense of backtesting retirement strategies on financial data) since there was deflation and the real value of bonds doubled in the 10 years after 1929.
scone
Posts: 1457
Joined: Wed Jul 11, 2012 4:46 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by scone »

dkturner wrote:You should explain to Bill that he doesn't understand.
Alternatively, you could speak for him and explain what he really, really meant. :P
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dbr »

wade wrote: dbr, I'm not sure I'm following your request: are you asking for error bands around the curves showing failure rates? I'm not sure how that would work out in practical terms, as there isn't really an error band if you just increase the sample size further. The way to introduce an error band would be to have a distribution for your asset returns assumption distributions, but that still requires assumptions and may not help matters. Would an acceptable alternative be to just show the distribution of remaining wealth in order to give a better idea about how far away from failure the strategy may put you, or the magnitude of failure if you do run out. I have considered this sort of approach.
First off, I think it is indeed interesting and useful to consider how this problem works out if we try to take knowledge of current conditions into account. So thanks for offering that result.

The essence of my question how reliable/accurate/uncertain is the estimate of SWR in any model. I believe you are correct that it is a question of sensitivity to the inputs.* The 2% presented in this thread varies from the conventional 4% as a result of selecting different inputs based on assuming knowledge of present conditions projected to the future. That could be a refinement of the answer conditional on present conditions or it could be an illustration of how sensitive the answer is to assumptions. I agree that assuming the distribution of assumptions is also subject to question. Nevertheless, we really can't accept numbers about things without some notion regarding the uncertainty of those numbers. So far I have never read an article on SWR where the author has addressed head-on the issue of the reliability of the estimate. It could be that is unfair to someone who has in fact done so. Maybe the Kitces paper on SWR after a major market decline is an example of considering one of the nuances of that problem but it isn't directly in the form of an estimate of error.

*SWR derived from historical data would seem to be immune to examining the sensitivity to assumptions as there is only one history to examine. The complaint is then made is that we have no way to know how well the next thirty years on random average is characterized by the past one hundred years. Your results do address the possible dimensions of that problem. It is also a legitimate concern that repeating the same history using markets of countries other than the US generates different results.
Exterous
Posts: 300
Joined: Mon Feb 20, 2012 12:34 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Exterous »

ResNullius wrote:What has happened to using some common sense? Continuing to strive for the perfect and absolutely certain withdrawal rate is a waste of time. The only things in life that are certain are death and taxes. If you want to start with 3% and adjust as needed, then do it. If you want to start with 4%, then do it, same for 5%. Just use some common sense. The lower the withdrawal rate, the safer you are in terms of longevity of portfolio life. Living a decent lifestyle is important, so just do it until the numbers say you should adjust upward or downward. You really can't go wrong with 4%, same for 3%, same for 2%, same for 1%. Things could get risky as you move up to 5% or 6%, but a lot depends on how much you start with, how much you need, and whether you care about leaving an inheritance. Endlessly debating the unknown seems like such a waste of time.
I don't think that a SWR is intended to be a static, unadjustable rate for those already in retirement. I think, by and large, most people trying to figure out what a SWR is will also be smart enough to know when they can or need to adjust it. I think the term Safe Withdrawl Rate and the rate of 4% serve much more as a planning role. It provides a baseline for someone in the accumulation phase and it's importance grows the farther away from retirement they are. If you are 30 now - will you have a good idea on what the economy will be doing when you are 65? What about health care costs? Taxes? How are you supposed to plan for retirement other than a vague plan of 'save as much as possible' and hope that what you consider 'as much as possible' is enough. Looking at the history behind the 4% SWR you can get a rough idea on where you should be or want to be when you retire and what you need to do to get there. As you approach retirement and questions about taxes, economy performance, health care costs become more clear you can fine tune/adjust your planning
Last edited by Exterous on Thu Nov 15, 2012 10:05 am, edited 1 time in total.
User avatar
CyberBob
Posts: 3387
Joined: Tue Feb 20, 2007 1:53 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by CyberBob »

Mitchell777 wrote:
bornloser wrote:
CyberBob wrote:There is an interesting Scott Burns article out recently that looks at the issue from a totally different angle; 30-year planning horizons bump up against the reality that you may expire long before your portfolio does.
Scott Burns wrote:The bottom line, however, is that death reduces the risk of running out of money more than does portfolio management.
Life, Death and How Long Your Money Will Last.

+1. I think two very important variables are best guess at the age you are going to check out (an educated estimate based on family, lifestyle, etc) and if you want to leave an estate. Got one colleague in a casket for viewing tonight (age 72), one colleague in the ICU (age 69) and a partner that passed last February at 66. All the hoopla about planning to have enough retirement for age 95 is really overkill, IMHO. And if you do not plan to leave an estate, I will remind you that the poor do pretty well in this nation. If I make it to age 90, will I really care if I have to ask for a government handout?
This hits home with me. In the past 2 weeks I learned of the deaths of 4 former colleagues, two in their 50's. Not that I want to run out of money at any age but, personally, I may be worrying a bit too much about running out at 95 or 100
I read an investing book years ago that suggested that the best method for making withdrawals from your portfolio was to be very conservative and assume that you would be taking out money from your portfolio until age 120 :shock: Now that's optimism!

Bob
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by azanon »

Agreeing with the original poster (but in a different way), YES simply deciding to take a certain percentage from your portfolio at retirement - even 3% - , inflation adjusting it, and then refusing to reconsider that withdrawal amount ever again despite potentially significant chances to your portfolio amount is "excessive". But that's not really the right word, is it? I'd go with ignorant or simply unwise, instead of excessive.

Did anyone ever really consider implementing a plan that involves you to not think anymore as a viable one, other than perhaps one that is strictly annuity/pension based? In practice, I don't actually believe anyone would do that. And if they would, I'd prefer not to know.
BruceA
Posts: 80
Joined: Thu Feb 23, 2012 3:34 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by BruceA »

A lot of very good comments regarding use of a SWR in planning. For reasons given, many give up on SWR and switch to Monte Carlo Simulations. In my way of thinking MCS is even worse, no matter how sophisticated the MCS program is. What does one do with the knowledge of an 86.43% chance of success??

I gave up on both of these and created a personal spreadsheet that computes the amount that I will have available to spend each year for the rest of my life. The number updates continually, and I have found it to be very helpful in my own planning. My personal model contains many variables that are set very conservatively, but are easily changeable. What-if disaster scenarios or major expenditures can easily be plugged in. It works.

I am not going to submit my personal spreadsheet to his forum, but I would be happy to work with someone, who is knowledgeable in Excel, who would be willing to review and help adapt this spreadsheet for broad use. I am not selling anything.
floydtime
Posts: 413
Joined: Sat Feb 26, 2011 9:09 pm
Location: A book

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by floydtime »

Imo, 5-6% (only as a rule of thumb of course) is quite reasonable if you aren't counting Social Security as part of your portfolio.

We're ~20 years from taking SS, but it will cover most of our budget (our budget today no less) when the time comes. 6% SWR as bridge income until we get there will still leave a pretty good nest egg. Heck, even 8% would.

Again, only as a rule of thumb - of course everyone should adjust to markets/life changes/reality as they go.
"Do not value money for any more nor any less than its worth; it is a good servant but a bad master" - Alexandre Dumas
YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by YDNAL »

Simplegift wrote:What's interesting is that Wade then goes on to run Monte Carlo simulations using these expected returns to project 30-year failure rates for various stock/bond portfolios at selected withdrawal rates (chart below). Accepting that "safe withdrawal rates" are no longer the latest word in retirement planning (e.g., see Milevsky or Bernstein on "guaranteed floor/upside potential"), it's still rather shocking to see that, given today's low expected returns, a 50% stock/50% bond portfolio with a 4% withdrawal rate is forecast to fail 47% of the time. This same portfolio with a 3% withdrawal rate has a 17% chance of failure. Yikes. Any thoughts?
It is fascinating that this thread is up to 73 replies and 4,300+ views when I saw it.

My thoughts, since you asked....
  • The US Census says 1.8% of the population in 2010 is over 85 years of age (table 2, pg 9).
    http://www.census.gov/prod/cen2010/brie ... 0br-09.pdf
  • Some people enjoy toying-around with calculators and numbers, yet NO ONE knows what awaits a retiring 65yo in the next 30 years.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dbr »

floydtime wrote:Imo, 5-6% (only as a rule of thumb of course) is quite reasonable if you aren't counting Social Security as part of your portfolio.

We're ~20 years from taking SS, but it will cover most of our budget (our budget today no less) when the time comes. 6% SWR as bridge income until we get there will still leave a pretty good nest egg. Heck, even 8% would.

Again, only as a rule of thumb - of course everyone should adjust to markets/life changes/reality as they go.
That is correct, but bridge income to an annuity is not the model Wade is discussing.

This is an excellent example of the fact that SWR models do not describe what happens in real world retirement. Of course, one can construct SWR methodology models in which income profiles and spending profiles are more complex and custom built. Even FireCalc does that. Doing so, one would find that a model that typically supports 4% SWR would support 6% in bridge mode as being safe or more than safe.
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by grayfox »

nisiprius wrote:In 1998, the Wall Street Journal Guide To Planning Your Financial Future: The Easy-To-Read Guide to Planning for Retirement said to use these as planning numbers.

Image

In 2012, Wade Pfau says to use these as planning numbers.

Image
The big difference here is that whoever reported that forecast at the New YorK Times in 1998 were a bunch dummies, while Wade actually uses his head to think.

In 1998, 11% per year was the past annualized return since the beginning of stocks. Anyone who has looked at forecasting stock returns knows that the past return is about of the worst way to forecast future returns. In 1998, valuations, which do forecast future returns, were at record highs due to stock market bubble. The dividend yield on S&P 500 was only about 1.3% and P/Es were about 33. The expected real return was about 0% real, which I think is about what we'e seen since 1999.

The lesson here is to not believe that the stuff you read in newspapers is anywhere close to correct. Especially if it has anything to do with numbers. People I know that majored in journalism couldn't handle pre-algebra.
YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by YDNAL »

grayfox wrote:In 1998.....

The lesson here is to not believe that the stuff you read in newspapers is anywhere close to correct. Especially if it has anything to do with numbers. People I know that majored in journalism couldn't handle pre-algebra.
One important lesson is that nobuddy nose nutting about the future! *

* I can handle pre-algebra, but spelling is something else. :)
Landy | Be yourself, everyone else is already taken -- Oscar Wilde
john94549
Posts: 4638
Joined: Tue Jul 26, 2011 8:50 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by john94549 »

The academic debate over withdrawal strategies is interesting, as I am a 65 year old retiree (my wife is still working, same age) and -- eventually -- we will begin to tap our "nest egg".

I am, however, amazed that so many folks can come to what appear to be widely divergent opinions. Everyone does take delight in piling on the "4% rule" (which I would agree was never a "rule" to begin with), but then one finds:

1. The "Modified RMD" approach (recent thread). For our ages, were we to start now, the math suggests a withdrawal of right around 5.3% - 6%, since our "interest plus dividends" amount for last year was comfortably north of 2.5%;

2. The Rick Ferri "2% plus interest and dividends" approach. This would translate to roughly 4.5% this year;

3. The "4% for starters" strategy, then re-evaluate in five years (my favorite) and its ancestor, the classic "4% plus annual inflation adjustments" non-rule;

4. The more conservative schools, which suggest something south (perhaps way south) of 4%; and

5. The Taylor approach, which I will paraphrase as "use your common sense" (good advice, but not terribly helpful for a newb like me).

To those I have not mentioned, or to the extent I have inadvertantly mis-characterized the various viewpoints, my apologies. However, since my math skills have diminished over time, and I chose Stat over Calculus to fulfill my math requirement at Stanford (I lasted a week, I was totally flummoxed by calc), I tend to stick with what's easy to understand.

So until otherwise persuaded, I think we'll start with 4%, then take a look in five years. Might "inflation-adjust" at that time. I'm still finding a reasonable choice of 2% long-term CDs out there for my ladder, even under ZIRP, so I'm using 2% for "interest plus dividends" for medium-term (i.e., five year) planning.
Leesbro63
Posts: 10638
Joined: Mon Nov 08, 2010 3:36 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Leesbro63 »

john94549 wrote: 2. The Rick Ferri "2% plus interest and dividends" approach. This would translate to roughly 4.5% this year;
Did I misread Rick? I don't remember that 2% base. Just "interest and dividends" is what I thought he recently said he advises his clients to spend not more than.
Prudence
Posts: 920
Joined: Fri Mar 09, 2012 3:55 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Prudence »

If Rick recommends spending only interest plus dividends, is he recommending don't spend any prinicpal throughout retirement?
User avatar
archbish99
Posts: 1649
Joined: Fri Jun 10, 2011 6:02 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by archbish99 »

Leesbro63 wrote:
john94549 wrote: 2. The Rick Ferri "2% plus interest and dividends" approach. This would translate to roughly 4.5% this year;
Did I misread Rick? I don't remember that 2% base. Just "interest and dividends" is what I thought he recently said he advises his clients to spend not more than.
Interest + dividends, plus 2% of principal if not trying to leave an inheritance, IIRC.
I'm not a financial advisor, I just play one on the Internet.
YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by YDNAL »

Leesbro63 wrote:
john94549 wrote: 2. The Rick Ferri "2% plus interest and dividends" approach. This would translate to roughly 4.5% this year;
Did I misread Rick? I don't remember that 2% base. Just "interest and dividends" is what I thought he recently said he advises his clients to spend not more than.
This is what he posted, actually in response to you Leesbro63, but I suggest for everyone to make their own determination.
Rick Ferri wrote:The SWR I tend to use is the cash flow from your portfolio in the form of dividend and interest income. This, along with your Social Security, possible pensions, annuities, and rental property if you have any should cover your basic living expenses. Cash flow from a 50.50 portfolio is currently about 3% if your portfolio has a position in high yield bonds.

That being said, none of us are going to live forever. Let's assume you're going to leave your heirs what's left and not worry about leaving them today's principal. Then you could take another 2% in principal each year as well without requiring capital gains over your lifetime. This would put SWR at 5%.

http://www.bogleheads.org/forum/viewtopic.php?t=81005
Landy | Be yourself, everyone else is already taken -- Oscar Wilde
User avatar
Abe
Posts: 2572
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Abe »

Leesbro63 wrote:I believe that I asked Rick Ferri, on here a while back, what HE does with his clients. And his answer, I think, was that he tries to have them "withdraw" or live on the income of their portfolio...interest and dividends. Which, as mentioned on this thread, was the "old school" way of doing things before we had enough computer power to play with and figure out "SWR". Currently if you do that with any sort of somewhat balanced portfolio you'd get a 2ish to 3ish percent income rate. And if you look at current concerns that 4% may be too high in the current environment, this "spend only the income" method sounds to me to be a reasonable common-sense way to plan for retirement spending. Plus it doesn't require knowing when to sell and is fairly automatic...garner the income then spend it!
I agree. I always thought in terms of "spend only the income" until the last few years when I got exposed to the Boglehead philosophy. The old way was spend only the income and never touch the principal. The Cardinal sin was spending the principal. I still don't think this is a bad way to look at it necessarily, assuming one has enough money. It is interesting that Rick Ferri does this with his clients. If he does, I'm sure he does both SWR as a percentage of total portfolio and spend only the income, depending on the situation. One thing about spending only the income, you don't have to worry so much about running out of money. Having said that, all of this can be talked about and analyzed until doomsday, but there is no concrete answer because no one knows what the future holds. For most of the questions asked on here, the answer is -- it depends.
Slow and steady wins the race.
peppers
Posts: 1650
Joined: Tue Oct 25, 2011 7:05 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by peppers »

YDNAL wrote:
grayfox wrote:In 1998.....

The lesson here is to not believe that the stuff you read in newspapers is anywhere close to correct. Especially if it has anything to do with numbers. People I know that majored in journalism couldn't handle pre-algebra.
One important lesson is that nobuddy nose nutting about the future! *

* I can handle pre-algebra, but spelling is something else. :)
I think YDNAL summed it up pretty well. :wink:
"..the cavalry ain't comin' kid, you're on your own..."
Leesbro63
Posts: 10638
Joined: Mon Nov 08, 2010 3:36 pm

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Leesbro63 »

Abe wrote:I agree. I always thought in terms of "spend only the income" until the last few years when I got exposed to the Boglehead philosophy. The old way was spend only the income and never touch the principal. The Cardinal sin was spending the principal. I still don't think this is a bad way to look at it necessarily, assuming one has enough money. It is interesting that Rick Ferri does this with his clients. If he does, I'm sure he does both SWR as a percentage of total portfolio and spend only the income, depending on the situation. One thing about spending only the income, you don't have to worry so much about running out of money. Having said that, all of this can be talked about and analyzed until doomsday, but there is no concrete answer because no one knows what the future holds. For most of the questions asked on here, the answer is -- it depends.
Acutally, Rick talks about spending the cash flow (interest and dividends) and UP TO another 2%, recognizing that the extra 2% is probably a draw down of your future estate. Here is what he said:


http://www.bogleheads.org/forum/viewtop ... st=1532645



"by Rick Ferri » Fri Aug 19, 2011 8:57 pm

SWR = sustainable withdrawal rate

The SWR I tend to use is the cash flow from your portfolio in the form of dividend and interest income. This, along with your Social Security, possible pensions, annuities, and rental property if you have any should cover your basic living expenses. Cash flow from a 50.50 portfolio is currently about 3% if your portfolio has a position in high yield bonds.

That being said, none of us are going to live forever. Let's assume you're going to leave your heirs what's left and not worry about leaving them today's principal. Then you could take another 2% in principal each year as well without requiring capital gains over your lifetime. This would put SWR at 5%.

These numbers are not inflation adjusted. If you use 3% as inflation as a long-term inflation number, then you're back to Bill's 2%. However, I believe 3% inflation is too high. IMO, giving the global economy as it is, inflation is likely to be 2% or less. This gets you back to a 3% real SWR.

That's my belief at this time, subject to change. Thanks for asking!

Rick Ferri"
User avatar
wade
Posts: 612
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by wade »

dbr wrote: First off, I think it is indeed interesting and useful to consider how this problem works out if we try to take knowledge of current conditions into account. So thanks for offering that result.

The essence of my question how reliable/accurate/uncertain is the estimate of SWR in any model. I believe you are correct that it is a question of sensitivity to the inputs.* The 2% presented in this thread varies from the conventional 4% as a result of selecting different inputs based on assuming knowledge of present conditions projected to the future. That could be a refinement of the answer conditional on present conditions or it could be an illustration of how sensitive the answer is to assumptions. I agree that assuming the distribution of assumptions is also subject to question. Nevertheless, we really can't accept numbers about things without some notion regarding the uncertainty of those numbers. So far I have never read an article on SWR where the author has addressed head-on the issue of the reliability of the estimate. It could be that is unfair to someone who has in fact done so. Maybe the Kitces paper on SWR after a major market decline is an example of considering one of the nuances of that problem but it isn't directly in the form of an estimate of error.

*SWR derived from historical data would seem to be immune to examining the sensitivity to assumptions as there is only one history to examine. The complaint is then made is that we have no way to know how well the next thirty years on random average is characterized by the past one hundred years. Your results do address the possible dimensions of that problem. It is also a legitimate concern that repeating the same history using markets of countries other than the US generates different results.
I'm not sure if I'm entirely clear yet on what you are looking for. Indeed, the SWR estimate does depend quite a bit on the asset market assumptions. Does the following figure help? It shows how, for a 30 year retirement and 10% failure rate, the SWR depends heavily on the assumptions about portfolio returns and standard deviations. Those lines in the figure represent the levels of sustainability withdrawal rates for the corresponding return/risk assumption.

Image
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dbr »

Wade, thanks for the above. I probably would have seen that if I had been following your publications exhaustively, but I don't.

Yes, this is pertinent to the question, indeed essential information required to interpret the results of any given MC simulation.

The next step is the difficult one, which is to assess where on that map is the region of probable portfolio performance that we expect. Suppose for a very simple illustration we accept Rick Ferri's 30-year estimate and buy a portfolio that is 100% US large cap stocks at 5.5% return and 19% SD. (It is not clear that Rick means SD of REAL returns, which might be a different number, also whether his numbers are meant to be arithmetic or geometric). Then we just read off an SWR of a little under 3%.

But the problem is not solved at all. Now we have to rail at Rick over the issue that he has put no error bars on his numbers so that we can draw a circle of 95% confidence around that 5.5 x 19 point on your chart. The pertinent question is how does one forecast the returns distribution. One might note that in your chart, for a portfolio around 6% real return it takes a drop of about 4.5% in return to reduce a 4% SWR to a 2% SWR. Hypothesizing a real return of no more than 1.5% real for a practical portfolio for the US investor today seems extreme at the least. I note that your comment about that is the following:

"I’m now trying to finalize a set of assumptions to use in my new research. Table 3 shows my current planned assumptions, though I do encourage and request any feedback you may have about these as I don’t fancy myself to be much of a market forecaster:"

I would be the last one to claim to know what the best estimates are and what the uncertainty in those estimates is. I think this leaves us with useful information as an exercise in understanding the theoretical behavior of portfolios. I would be less sure there is something here that the actual retiree can apply as helpful advice in financial planning.
User avatar
wade
Posts: 612
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by wade »

dbr wrote:Wade, thanks for the above. I probably would have seen that if I had been following your publications exhaustively, but I don't.

Yes, this is pertinent to the question, indeed essential information required to interpret the results of any given MC simulation.

The next step is the difficult one, which is to assess where on that map is the region of probable portfolio performance that we expect. Suppose for a very simple illustration we accept Rick Ferri's 30-year estimate and buy a portfolio that is 100% US large cap stocks at 5.5% return and 19% SD. (It is not clear that Rick means SD of REAL returns, which might be a different number, also whether his numbers are meant to be arithmetic or geometric). Then we just read off an SWR of a little under 3%.

But the problem is not solved at all. Now we have to rail at Rick over the issue that he has put no error bars on his numbers so that we can draw a circle of 95% confidence around that 5.5 x 19 point on your chart. The pertinent question is how does one forecast the returns distribution. One might note that in your chart, for a portfolio around 6% real return it takes a drop of about 4.5% in return to reduce a 4% SWR to a 2% SWR. Hypothesizing a real return of no more than 1.5% real for a practical portfolio for the US investor today seems extreme at the least. I note that your comment about that is the following:

"I’m now trying to finalize a set of assumptions to use in my new research. Table 3 shows my current planned assumptions, though I do encourage and request any feedback you may have about these as I don’t fancy myself to be much of a market forecaster:"

I would be the last one to claim to know what the best estimates are and what the uncertainty in those estimates is. I think this leaves us with useful information as an exercise in understanding the theoretical behavior of portfolios. I would be less sure there is something here that the actual retiree can apply as helpful advice in financial planning.
dbr: I think I see better what you are getting at now.

About a 95% confidence band, that could be constructed for the case in which you accept a return and volatility as given. In each simulation under these assumptions, you can calculate the maximum sustainable withdrawal rate. Then you order them and identify the 97.5 and 2.5 percentiles, and viola, there is your confidence interval. That is possible because one of your assumptions is the volatility. It is essentially what I do to make this figure... I'm identifying the sustainable withdrawal rate from the 10th percentile of the distribution.

More broadly, a clear uncertainty is that we do not know if the assumption about the return is any good. If you relax that assumption, then making the confidence interval becomes harder, unless you are going to identify a secondary distribution for the mean return to be used in each Monte Carlo simulation.

Perhaps the lesson from all of this, though, is that a safe withdrawal rate can not be identified with any sort of real confidence. This suggests alternative approaches such as first making sure that basic needs can be met no matter what happens in the markets.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by dbr »

Thanks.

I think a good example of a parallel problem is the propagation of error problem in experimental science. In that case one wants to know the uncertainty in a calculated result given the uncertainty in the numbers entered into the calculation. In that sense the calculation is the "easy" part while determining the numbers to enter into the calculation and their uncertainty is the "hard" part, meaning the outcome of having to do an experiment as opposed to dropping the experimental measurements into a formula. Note that the MC calculation of the SWR actually does not produce a result that has any stochastic variability once we run enough iterations of the simulation, a point that you made in your first reply to my question. What you have is the derivation of a formula that outputs a definite SWR for a given specification of the acceptable failure rate, the time duration, and the mean and standard deviation of the (normal) return distribution. Given that, the uncertainty or reliability problem reduces to estimating the error in the mean and standard deviation that we choose to use. That is a problem outside the MC simulation itself just as making an experimental measurement is a problem outside the selection of the formula that we will use to calculate some result from the experiment.

As an example, we could assume that when we estimate returns and standard deviations of returns of portfolios that the 95% confidence interval on those estimates is +/-2% and +/-5%. Laying those ranges on my hypothesized 5.5 x 19 center point produces an "oval" of confidence in SWR that ranges from 2% to 4% and has a value around 3% at the center. So we might say that the SWR is 3% +/- 1%. Technically I think you have to use some formulas to get the actual 95% interval in the SWR, but the picture would be approximately right.
YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by YDNAL »

dbr wrote:
wade wrote:Perhaps the lesson from all of this, though, is that a safe withdrawal rate can not be identified with any sort of real confidence. This suggests alternative approaches such as first making sure that basic needs can be met no matter what happens in the markets.
I think a good example of a parallel problem is the propagation of error problem in experimental science. In that case one wants to know the uncertainty in a calculated result given the uncertainty in the numbers entered into the calculation. In that sense the calculation is the "easy" part while determining the numbers to enter into the calculation and their uncertainty is the "hard" part, meaning the outcome of having to do an experiment as opposed to dropping the experimental measurements into a formula.
Bingo!

It is easy to develop a financial model based on assumptions. It is impossible to know if the assumptions will materialize going forward. That said, in science we know that water will boil at 212° F (99.98° C) at 1 atmosphere of pressure (sea level). In finance we have no certainty whether the U.S. Stock Market returns 8% annually or 0% annually over the next decade or two...... assumptions, assumptions.
Landy | Be yourself, everyone else is already taken -- Oscar Wilde
User avatar
Abe
Posts: 2572
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Abe »

YDNAL wrote:Bingo!
It is easy to develop a financial model based on assumptions. It is impossible to know if the assumptions will materialize going forward. That said, in science we know that water will boil at 212° F (99.98° C) at 1 atmosphere of pressure (sea level). In finance we have no certainty whether the U.S. Stock Market returns 8% annually or 0% annually over the next decade or two...... assumptions, assumptions.
That's the reason I like "spend only the income".
Slow and steady wins the race.
User avatar
Cut-Throat
Posts: 2011
Joined: Sun Oct 17, 2010 9:46 am

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Cut-Throat »

Abe wrote:
YDNAL wrote:Bingo!
It is easy to develop a financial model based on assumptions. It is impossible to know if the assumptions will materialize going forward. That said, in science we know that water will boil at 212° F (99.98° C) at 1 atmosphere of pressure (sea level). In finance we have no certainty whether the U.S. Stock Market returns 8% annually or 0% annually over the next decade or two...... assumptions, assumptions.
That's the reason I like "spend only the income".
What if there is no Income ?.......You don't know that either.
User avatar
Abe
Posts: 2572
Joined: Fri Sep 18, 2009 5:24 pm
Location: Earth in the Milky Way Galaxy

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Abe »

Cut-Throat wrote:
Abe wrote:
YDNAL wrote:Bingo!
It is easy to develop a financial model based on assumptions. It is impossible to know if the assumptions will materialize going forward. That said, in science we know that water will boil at 212° F (99.98° C) at 1 atmosphere of pressure (sea level). In finance we have no certainty whether the U.S. Stock Market returns 8% annually or 0% annually over the next decade or two...... assumptions, assumptions.
That's the reason I like "spend only the income".
What if there is no Income ?.......You don't know that either.
Well, no I don't absolutely know anything that's in the future. If there is no income, I guess I'll have to try something else. I didn't say it was chiseled in concrete.
Slow and steady wins the race.
User avatar
Jerilynn
Posts: 1929
Joined: Tue Sep 06, 2011 12:49 pm
Location: USA, Earth

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by Jerilynn »

CyberBob wrote: I read an investing book years ago that suggested that the best method for making withdrawals from your portfolio was to be very conservative and assume that you would be taking out money from your portfolio until age 120 :shock: Now that's optimism!

Bob
I keep telling y'all that there will some medical breakthrough(s) in the near future that will extend the human lifespan significantly. Remember, you heard it here first.
Cordially, Jeri . . . 100% all natural asset allocation. (no supernatural methods used)
User avatar
nisiprius
Advisory Board
Posts: 52211
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Are Even 3% Withdrawal Rates Excessive for Retirees Toda

Post by nisiprius »

Jerilynn wrote:I keep telling y'all that there will some medical breakthrough(s) in the near future that will extend the human lifespan significantly. Remember, you heard it here first.
I think George Bernard Shaw beat you to it, in Back to Methuselah: (A Metabiological Pentateuch), which was first performed in 1922. He is a little bit vague on the details, though. His theory was that the problems of the human condition were entirely due to our limited life span, and that it was absolutely necessary for the survival of the human race that we extend our life span to 300 years, and therefore once we were all sufficiently convinced of that, simply by our willing it to happen, it would happen. Through the action of the Life Force, if I remember correctly.

However, in his play(s) it doesn't happen until 2170 AD. Dang it. Too late for me.

This is the play in which the Serpent says to Eve, "When you and Adam talk, I hear you say 'Why?' Always 'Why?' You see things, and you say 'Why?' But I dream things that never were, and I ask 'Why not?'" Borrowed by Robert F. Kennedy.

I see no evidence at all of any progress on life extension--more and more of us our reaching our allotment of years and reaching it in better and better health. But it wasn't rare for men to live into the 80s and 90s two centuries ago--Benjamin Franklin lived to age 84, John Adams to 90--and it isn't common for men to live past 90 today. Anyway, I thought for a long time that we were never really going to see "flat TV you can hang on a wall," which had been continually promised since the 1950s, but it is finally there. So perhaps if we finally have flat TV, we can get a breakthrough in life extension. If so, would be very convenient if it could happen within the next decade or two. If I can't get that, then, please, can I at least get the helicars and moon colonies? I'll pass on the "complete meal in a single pill."
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Post Reply