Kitces - The Ratcheting Safe Withdrawal Rate

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills
Topic Author
moghopper
Posts: 143
Joined: Fri Oct 18, 2013 10:07 am

Kitces - The Ratcheting Safe Withdrawal Rate

Post by moghopper »

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by Browser »

Interesting. The 4% floor rule applies if your retirement horizon is 30 years, or to age 95 if you begin taking withdrawals at age 65. Does anyone know how this should be adjusted for shorter retirement horizons such as 25 years, 20 years?
We don't know where we are, or where we're going -- but we're making good time.
User avatar
Taylor Larimore
Advisory Board
Posts: 30178
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by Taylor Larimore »

Browser wrote:Interesting. The 4% floor rule applies if your retirement horizon is 30 years, or to age 95 if you begin taking withdrawals at age 65. Does anyone know how this should be adjusted for shorter retirement horizons such as 25 years, 20 years?
Browser:

The Trinity Study shows sustainable withdrawal rates for 15, 20, 25 and 30-year periods:

http://www.bogleheads.org/wiki/Safe_withdrawal_rates

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
gasdoc
Posts: 2000
Joined: Mon Nov 03, 2014 8:26 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by gasdoc »

I did not see in the article whether the 4% was calculated based on the starting balance, or whether it was recalculated annually. I am assuming the former, whereby there is some flexibility put in place in the paper for better than average returns. I think I would prefer a more "dynamic" withdrawal rate, such as is now recommended in Vanguard articles, where the annual amount available for spending is a percent of the previous year end balance, but with a floor and ceiling to prevent large swings from year to year.

gasdoc
Last edited by gasdoc on Wed Jun 03, 2015 7:29 pm, edited 1 time in total.
User avatar
k66
Posts: 504
Joined: Sat Oct 27, 2012 1:36 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by k66 »

Browser wrote:Interesting. The 4% floor rule applies if your retirement horizon is 30 years, or to age 95 if you begin taking withdrawals at age 65. Does anyone know how this should be adjusted for shorter retirement horizons such as 25 years, 20 years?
For example, in my own particular modelling (similar to but not exactly per Bengen's model), I have calculated Terminal Depletion Rates (the rate at which you can spend a constant inflation-adjusted percentage of initial portfolio) for several time frames:

Minimum Values
25 years - 4.03%
30 years - 3.75%
35 years - 3.61%
40 years - 3.49%

5th Percentile Failure
25 years - 4.21%
30 years - 3.90%
35 years - 3.75%
40 years - 3.63%

50th Percentile Failure
25 years - 6.29%
30 years - 5.71%
35 years - 5.30%
40 years - 4.76%

95th Percentile Failure
25 years - 9.72%
30 years - 9.09%
35 years - 7.23%
40 years - 6.81%

Maximum Values
25 years - 10.64%
30 years - 9.95%
35 years - 7.75%
40 years - 7.29%

Computed using a 60% Equities portfolio (remains constant for the time periods noted).
LOSER of the Boglehead Contest 2015 | lang may yer lum reek
ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by ourbrooks »

Seems to me that there's a much simpler approach than Kitces which will allow you to start increasing spending sooner if the market does well:

Start out with 4% of your initial balance and adjust it for inflation each year.
Every five years, calculate 4% of your current portfolio. If it's more than what you planned on spending, based on the initial 4%, increase your spending to the new 4% and continue on spending at the new rate with inflation adjustments. If it's less, stick with the old rate.

Under this scheme, you'll never have to cut back spending, although you may not increase it either. Five years is not a magic number; you could make the calculation every year if you wanted to.

This is actually a conservative rule since the 4% rate is based on a thirty year withdrawal period and the older you get, the less likely it is that you'll last that long.
User avatar
k66
Posts: 504
Joined: Sat Oct 27, 2012 1:36 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by k66 »

By far (in my opinion), the most efficient spending method is the Variable Percentage Withdrawal (VPW) or similar approach.

Although the original 4% (Bengen) and follow-up Trinity (and other) studies are very informative, I am not actually planning to follow this type of spending plan. For myself, 4% is only useful to understand at what point my accumulated savings are likely sufficient to begin retirement. As in:

If 4% x Savings > Current Spending, Then Retire.
LOSER of the Boglehead Contest 2015 | lang may yer lum reek
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by Browser »

The paper makes the excellent point that the 4% rule is the "worst possible case" rule. I've often noticed when I run portfolio returns using a 4% real withdrawal rate that it often comes up looking pretty good most of the time. In fact it's danged hard to concoct an allocation that doesn't look decent. Now I realize that's because I'm using return data that don't include the historically worst starting times. I've also pondered that the margin of safety provided by the 4% rule is likely to leave a lot of money on the table when they cart you away. Which is OK for many folks with strong bequest motives as long as it doesn't cut too deeply into their desired standard of living while they're still breathing. But the dilemma is that one's spending needs are likely to decrease during one's retirement horizon as they age and become less active, except of course if high medical/long term care expenses are incurred. Ideally, you would like to be able to spend more early in retirement rather than waiting to see how your portfolio turns out after 10 or 15 years. So, the "ratchet" strategy doesn't really address this.
We don't know where we are, or where we're going -- but we're making good time.
User avatar
gasdoc
Posts: 2000
Joined: Mon Nov 03, 2014 8:26 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by gasdoc »

k66 wrote:By far (in my opinion), the most efficient spending method is the Variable Percentage Withdrawal (VPW) or similar approach.

Although the original 4% (Bengen) and follow-up Trinity (and other) studies are very informative, I am not actually planning to follow this type of spending plan. For myself, 4% is only useful to understand at what point my accumulated savings are likely sufficient to begin retirement. As in:

If 4% x Savings > Current Spending, Then Retire.
+1
gasdoc
ourbrooks
Posts: 1575
Joined: Fri Nov 13, 2009 4:56 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by ourbrooks »

VPW doesn't work for me at all. VPW requires either that you have a low return, nearly all bond portfolio, or that you're willing to adjust your spending to follow the gyrations of the stock market. For example, if, in 2008, you held a conservative 50/50 portfolio, the asset allocation used in the original Trinity study, you'd have had to cut your spending by around 20% if you used VPW.

Coping with a 20% cut in spending is a new experience for most people. For most of their lives, they've had a job which paid the same amount, year in year out, with perhaps the occasional raise. The only time they got a 20% cut was when they got laid off or fired. They had a constant income they could plan around. I include myself in that set, and my retirement spending plan is built around being able to cover what we'll probably need for things like housing, taxes, medical care, food, utilities, house repairs and car replacement. Even things like vacations are part of the plan; if I didn't have enough retirement income to cover the vacations we want to take, I wouldn't have retired until I had enough money to cover the vacations.

Given how variable even these planned expenses are likely to be, I've built in a large safety margin to our plan so I think we could handle even a 20% cut in income by cutting back our safety margin, but if what VPW means is periodically cutting back on our safety margin in response to market tantrums, I'm not sure what it buys me.

I suspect that VPW works well for two classes of people. One class is those who have great flexibility in their financial needs. Market not doing so hot? Move to a smaller apartment and sell the car. As long as they've got their classical music Internet radio station they're happy. (I envy them; perhaps, they were sales people on commission in their professional careers and learned to be comfortable living this way.)

The other group for who VPW probably works well are those for whom their investment portfolio provides relatively little of their retirement income. A 20% cut in 20% of your income is only 4% overall. I think I could even manage that.

For me, the five year adjustment plan I outlined above is a much better approach. I never have to cut and if I do increase, it's an increase for life.
User avatar
1210sda
Posts: 1787
Joined: Wed Feb 28, 2007 8:31 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by 1210sda »

I enjoyed the article. Gives one food for thought.

I have two concerns:
1. a 50% increase in the retirement portfolio is not an easy target, especially in the earlier years of retirement.
2. For many retirees, it means needing to use a financial planner annually or at least every few years.

The second item may not be a bad thing to do anyway, but some retirees were not planning to do it.

1210
Sconie
Posts: 932
Joined: Sun Feb 07, 2010 10:23 am
Location: Arizona

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by Sconie »

1210sda wrote:I enjoyed the article. Gives one food for thought.

I have two concerns:
1. a 50% increase in the retirement portfolio is not an easy target, especially in the earlier years of retirement.
2. For many retirees, it means needing to use a financial planner annually or at least every few years.

The second item may not be a bad thing to do anyway, but some retirees were not planning to do it.

1210
Well, you're right----however----it is also important to remember that Kitces is primarily writing for "professional" financial planners........just sayin'
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant. - Alan Greenspan
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

Everyone who writes has a take on it and a least one method to propose. Phau, Bengen, Kitces, Guyton and Klinger, Zolt... and on and on. Steady percentage, inflation adjusted dollars, averages of this or that, VPW and so forth.

At 67 and three years retired I can only see through my eyes. I see a 6 year bull market run which makes it the fourth longest on record. Trying to counter that recency bias with the lost decade or the 2000 & 2008 gyrations, who knows what's next? I see bonds at historic lows which could easily provide zero real for the next five to ten years. Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study. Anyone see those conditions existing today or as a reasonable expectation for the coming 5 to 10 years. Since the future is unknown, right down to how long you will live and how expensive your care might or might not be, and for how long, all you can do is read about withdrawal methods and find a comfort level for yourself and your situation.. a desired lifestyle, amounts to bequeath, charitable giving, etc., then be flexible with what you do. To say any plan is what you will do and mechanically go after it year after year is a bit off in my book. If things go well you can always take another trip or find ways to spend or give more, just don't paint yourself into a corner where you can't spend less if it's needed. Your human capital is gone, don't follow it with bad planning or a lack of flexibility.
User avatar
1210sda
Posts: 1787
Joined: Wed Feb 28, 2007 8:31 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by 1210sda »

midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is.
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

midareff wrote:
1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.

In Pfau's recent paper he is making a projection of future returns to arrive at the conclusion that the current "SWR" is 2.85%. That's a very different sort of analysis.
Dave
randomguy
Posts: 9208
Joined: Wed Sep 17, 2014 9:00 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by randomguy »

midareff wrote:
1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
Dr. Phau is making a bunch of assumptions on what returns will be based on current market valuations and bond returns to get that number. Those assuptions might be right (or wrong) and slight variances can give drastically different results when you are looking at the worst 5% cases. You can argue if historical sequences are better or worse than generating sequences based on historical data. There are pluses or minsus to each approach.

But lets say you believe Dr. Phau's. What do you do if over the next 5 years your portfolio returns 10%,15%,-5%,20%,9%. Your initial 5 years haven't had a major crash so you are not on the line where you need to need spend 2.8%. Should you spend 4%, 5%, or some other number? You can use a number of approaches. This ratcheting one is a pretty conservative way of going.

Personally if I end up in the top 50% cases after a couple years (take that as my portfolio is growing in real terms), I am likely to be more aggressive. Why? Because I want to spend more money from 55 to 75 than from 75-95.
miles monroe
Posts: 1279
Joined: Mon Jan 20, 2014 12:14 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by miles monroe »

i don't believe the 4% rule is conservative and leaves money on the table.

many people have a 50-50 portfolio, and the days that the "safe" 50% is gonna generate a steady 6% are long gone.

its just a tool. a starting point. and too high a starting point imo.
randomguy
Posts: 9208
Joined: Wed Sep 17, 2014 9:00 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by randomguy »

miles monroe wrote:i don't believe the 4% rule is conservative and leaves money on the table.

many people have a 50-50 portfolio, and the days that the "safe" 50% is gonna generate a steady 6% are long gone.

its just a tool. a starting point. and too high a starting point imo.
You think the 30s,60s, and 70s were steady?:)
User avatar
SquawkIdent
Posts: 586
Joined: Tue Dec 23, 2008 7:14 pm
Location: Planet Earth

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by SquawkIdent »

k66 wrote:
Browser wrote:Interesting. The 4% floor rule applies if your retirement horizon is 30 years, or to age 95 if you begin taking withdrawals at age 65. Does anyone know how this should be adjusted for shorter retirement horizons such as 25 years, 20 years?
For example, in my own particular modelling (similar to but not exactly per Bengen's model), I have calculated Terminal Depletion Rates (the rate at which you can spend a constant inflation-adjusted percentage of initial portfolio) for several time frames:

Minimum Values
25 years - 4.03%
30 years - 3.75%
35 years - 3.61%
40 years - 3.49%

5th Percentile Failure
25 years - 4.21%
30 years - 3.90%
35 years - 3.75%
40 years - 3.63%

50th Percentile Failure
25 years - 6.29%
30 years - 5.71%
35 years - 5.30%
40 years - 4.76%

95th Percentile Failure
25 years - 9.72%
30 years - 9.09%
35 years - 7.23%
40 years - 6.81%

Maximum Values
25 years - 10.64%
30 years - 9.95%
35 years - 7.75%
40 years - 7.29%

Computed using a 60% Equities portfolio (remains constant for the time periods noted).
Very interesting. Do you have similar data for higher and lower stock allocations? :sharebeer
User avatar
1210sda
Posts: 1787
Joined: Wed Feb 28, 2007 8:31 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by 1210sda »

DFrank wrote: Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.
So is what midareff said about Stocks=10%, bonds= 5.1% and inflation = 3% not accurate?

1210
User avatar
k66
Posts: 504
Joined: Sat Oct 27, 2012 1:36 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by k66 »

SquawkIdent wrote:...Very interesting. Do you have similar data for higher and lower stock allocations? :sharebeer
I do. And you can introduce a bi-linear glide path too (e.g. specify Start & End Years for the Glide, after which it remains level for the remaining portion of the Timeframe).

Start: 80% Eq.
End: 100% Eq. @ 25 years

Minimum Values
25 years - 4.14%
30 years - 3.89%
35 years - 3.79%
40 years - 3.69%

Maximum Values
25 years - 12.22%
30 years - 11.09%
35 years - 10.30%
40 years - 9.88%

That data used is from Damodaran's S&P500 Total Return, and Shiller's 10-yr Treasury & CPI data (all from 1928 forward to 2014).
LOSER of the Boglehead Contest 2015 | lang may yer lum reek
User avatar
SquawkIdent
Posts: 586
Joined: Tue Dec 23, 2008 7:14 pm
Location: Planet Earth

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by SquawkIdent »

k66 wrote:
SquawkIdent wrote:...Very interesting. Do you have similar data for higher and lower stock allocations? :sharebeer
I do. And you can introduce a bi-linear glide path too (e.g. specify Start & End Years for the Glide, after which it remains level for the remaining portion of the Timeframe).

Start: 80% Eq.
End: 100% Eq. @ 25 years

Minimum Values
25 years - 4.14%
30 years - 3.89%
35 years - 3.79%
40 years - 3.69%

Maximum Values
25 years - 12.22%
30 years - 11.09%
35 years - 10.30%
40 years - 9.88%

That data used is from Damodaran's S&P500 Total Return, and Shiller's 10-yr Treasury & CPI data (all from 1928 forward to 2014).
Anyway you can post that info for 20/80, 40/60, 80/20 and 100% stock for comparison? Very interested in the 40 year numbers because that is rarely addressed in retirement spending scenarios. Thanks. :sharebeer
User avatar
k66
Posts: 504
Joined: Sat Oct 27, 2012 1:36 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by k66 »

1210sda wrote:
DFrank wrote: Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.
So is what midareff said about Stocks=10%, bonds= 5.1% and inflation = 3% not accurate?

1210
Technically no. But it could imply that those (mean) values lead to the worst-case result(s) which in turn formed Bengen's "4%" safe rate conclusion. As per Kitce's discussion (link given by the OP), the mean result has always been much (much) better... so how do you compensate for that?
LOSER of the Boglehead Contest 2015 | lang may yer lum reek
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

1210sda wrote:
DFrank wrote: Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.
So is what midareff said about Stocks=10%, bonds= 5.1% and inflation = 3% not accurate?

1210

No, it's not an accurate description of Bengen's methodology.
Dave
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

DFrank wrote:
midareff wrote:
1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.

In Pfau's recent paper he is making a projection of future returns to arrive at the conclusion that the current "SWR" is 2.85%. That's a very different sort of analysis.

To quote from the link to the paper I provided starting with the section title;

"The Averages

To begin with, let's see how our hypothetical planner got into trouble. By referring to the Ibbotson data (which we assume had not changed significantly since 2004), our planner learned that common stocks had returned 10.3 percent compounded over the years, and intermediate-term Treasuries had returned 5.1 percent. Inflation averaged 3 percent for the same period."

That's the use of averages.
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

randomguy wrote:
midareff wrote:
1210sda wrote:
midareff wrote: Bengen used a 10% equity return with 5.1% on 5 year treasury bonds vs. a 3% CPI in his 4% withdrawal rate study.
I thought he used history (stocks, bonds and inflation) so that he could capture the "worst case" such as the Great Depression. I thought he had followed a similar approach to that of the Trinity Study.

Maybe I just mis-read.

1210

That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
Dr. Phau is making a bunch of assumptions on what returns will be based on current market valuations and bond returns to get that number. Those assuptions might be right (or wrong) and slight variances can give drastically different results when you are looking at the worst 5% cases. You can argue if historical sequences are better or worse than generating sequences based on historical data. There are pluses or minsus to each approach.

But lets say you believe Dr. Phau's. What do you do if over the next 5 years your portfolio returns 10%,15%,-5%,20%,9%. Your initial 5 years haven't had a major crash so you are not on the line where you need to need spend 2.8%. Should you spend 4%, 5%, or some other number? You can use a number of approaches. This ratcheting one is a pretty conservative way of going.

Personally if I end up in the top 50% cases after a couple years (take that as my portfolio is growing in real terms), I am likely to be more aggressive. Why? Because I want to spend more money from 55 to 75 than from 75-95.
That's really part of the point. No one knows exactly what the market's returns will be going forward, or even if there will be positive returns. All there are looking forward from how long we will live to the continuance of social security, pensions, market returns, inflation rates, taxing structures and so forth is assumptions. The road is only clear through the rear view mirror and there are lots of papers and methods to try and see through the windshield but nothing is truly finite about the future of markets and life. As Taylor would say .. be flexible.
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

DFrank wrote:
1210sda wrote:
DFrank wrote: Bengen used year by year historical data for returns and inflation, not a single average figure. So, for example, he looked at what would have happened if retirement began in 1926 with a 4% withdrawal, and used the actual annual stock and bond returns and inflation over the next period of time. Repeat for a retirement start date in 1927, etc. He calculated the probability of success based on what percentage of all the starting years he analyzed produced a successful result.
So is what midareff said about Stocks=10%, bonds= 5.1% and inflation = 3% not accurate?

1210

No, it's not an accurate description of Bengen's methodology.

You might try opening the link I provided and reading the language under the sub-title "The Averages" which is near the bottom of the first page for yourself. It is exactly what Bengen says.
Mitchell777
Posts: 1154
Joined: Mon May 21, 2007 6:32 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by Mitchell777 »

I'm having trouble downloading Dr. Pfau's paper but in one he published a week later in 2013 he and his co-authors used "1% as a proxy for the asset management fees that are likely to be paid by an investor". Made me think I can go a bit higher in withdrawal if I wanted to use his logic
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

midareff wrote:You might try opening the link I provided and reading the language under the sub-title "The Averages" which is near the bottom of the first page for yourself. It is exactly what Bengen says.
But that section of the paper is not a description of how Bengen derived the 4% SWR rule. It's actually a hypothetical description of how using a forecasting method based on long term averages can lead you astray because that method can not account for what we now call sequence of returns risk. It's his example of how and why that method is faulty that sets up his alternative, which uses a (relatively) large number of sequences of historical returns and inflation.
Dave
randomguy
Posts: 9208
Joined: Wed Sep 17, 2014 9:00 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by randomguy »

midareff wrote:
That's really part of the point. No one knows exactly what the market's returns will be going forward, or even if there will be positive returns. All there are looking forward from how long we will live to the continuance of social security, pensions, market returns, inflation rates, taxing structures and so forth is assumptions. The road is only clear through the rear view mirror and there are lots of papers and methods to try and see through the windshield but nothing is truly finite about the future of markets and life. As Taylor would say .. be flexible.
NAh Taylor says retire at the start of a 18 year bull market in bonds and stocks and life is good:) Yes you need to be flexible but what these studies give you is a guidelines of how flexible you should be. If the market drops 50%, do you need to cut spending by 50% or is 20% enough? If the market is up 50%, how much should you save for the rainy day versus spending on booze, woman and cars. How low can you let your balance drop before you really start to worry. And so on. Anyone who looks at things like the 4% rule, VPW, or anything as something to be followed blindly for 30 years is insane. But that doesn't mean there isn't value in looking it how things have worked in the past (or might in the future if your a modeling person) for hints on likely outcomes. Odds are they will be wrong.
User avatar
k66
Posts: 504
Joined: Sat Oct 27, 2012 1:36 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by k66 »

SquawkIdent wrote:...Anyway you can post that info for 20/80, 40/60, 80/20 and 100% stock for comparison? Very interested in the 40 year numbers because that is rarely addressed in retirement spending scenarios. Thanks. :sharebeer
The terminal deplation rates (TDR) for 25-, 30-, 35-, and 40-year timeframes using portfolios built from constant equity portions (the S&P500 total return) as noted with the remainder being composed of Fixed Income (10-year Treasuries). The percentile figure represents the relative number of portfolios having less than the TDR tabulated. Data is from 1928 to 2014 (63 25-year periods, 58 30-year periods, etc). The TDR is a constant inflation-adjusted withdrawal rate required to deplete the portfolio at the end of the given timeframe.

TDR for 0% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	2.82%	2.35%	2.01%	1.71%
5%	3.17%	2.63%	2.19%	1.86%
10%	3.26%	2.71%	2.29%	1.94%
25%	3.52%	3.06%	2.55%	2.20%
50%	3.89%	3.22%	2.86%	2.61%
75%	5.18%	4.20%	3.45%	3.00%
90%	6.80%	5.57%	4.02%	3.40%
95%	7.90%	7.23%	4.40%	3.51%
100%	8.91%	8.18%	5.74%	3.81%
TDR for 20% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	3.72%	3.36%	2.99%	2.65%
5%	3.85%	3.40%	3.11%	2.79%
10%	3.92%	3.50%	3.19%	2.87%
25%	4.21%	3.70%	3.37%	3.07%
50%	4.59%	4.00%	3.60%	3.30%
75%	5.73%	4.92%	4.06%	3.60%
90%	7.49%	6.32%	4.78%	4.09%
95%	8.51%	7.92%	5.04%	4.29%
100%	9.56%	8.85%	6.46%	4.62%
TDR for 40% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	3.90%	3.60%	3.43%	3.30%
5%	4.10%	3.82%	3.62%	3.45%
10%	4.32%	3.90%	3.67%	3.51%
25%	4.78%	4.33%	4.00%	3.78%
50%	5.70%	5.00%	4.53%	4.15%
75%	6.81%	5.80%	5.15%	4.55%
90%	8.15%	7.04%	5.65%	4.98%
95%	9.16%	8.55%	5.95%	5.34%
100%	10.13%	9.44%	7.14%	5.72%
TDR for 60% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	4.03%	3.75%	3.61%	3.49%
5%	4.21%	3.90%	3.75%	3.63%
10%	4.50%	4.19%	3.98%	3.84%
25%	5.21%	4.69%	4.41%	4.24%
50%	6.29%	5.71%	5.30%	4.76%
75%	8.05%	7.14%	6.34%	5.90%
90%	8.93%	8.19%	7.05%	6.42%
95%	9.72%	9.09%	7.23%	6.81%
100%	10.64%	9.95%	7.75%	7.29%
TDR for 80% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	4.11%	3.84%	3.73%	3.62%
5%	4.33%	4.03%	3.88%	3.75%
10%	4.70%	4.36%	4.14%	3.95%
25%	5.49%	4.97%	4.62%	4.32%
50%	7.21%	6.59%	6.09%	5.30%
75%	9.09%	8.49%	7.63%	7.25%
90%	10.13%	9.50%	8.29%	7.92%
95%	10.72%	10.03%	8.56%	8.35%
100%	11.83%	10.70%	9.88%	9.43%
TDR for 100% Equities:

Code: Select all

%ile	25 yr	30 yr	35 yr	40 yr
0%	3.64%	3.50%	3.42%	3.37%
5%	4.14%	3.92%	3.82%	3.72%
10%	4.50%	4.21%	4.05%	3.89%
25%	5.55%	5.00%	4.75%	4.38%
50%	8.07%	7.38%	6.48%	5.72%
75%	9.91%	9.64%	8.81%	8.66%
90%	11.31%	10.59%	9.89%	9.56%
95%	11.92%	10.84%	10.37%	10.14%
100%	14.12%	12.97%	12.16%	11.71%
Disclaimer: Calculations have not been verified.
LOSER of the Boglehead Contest 2015 | lang may yer lum reek
trueblueky
Posts: 1921
Joined: Tue May 27, 2014 3:50 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by trueblueky »

k66 wrote:
Browser wrote:Interesting. The 4% floor rule applies if your retirement horizon is 30 years, or to age 95 if you begin taking withdrawals at age 65. Does anyone know how this should be adjusted for shorter retirement horizons such as 25 years, 20 years?
For example, in my own particular modelling (similar to but not exactly per Bengen's model), I have calculated Terminal Depletion Rates (the rate at which you can spend a constant inflation-adjusted percentage of initial portfolio) for several time frames:

Minimum Values
25 years - 4.03%
30 years - 3.75%
35 years - 3.61%
40 years - 3.49%

5th Percentile Failure
25 years - 4.21%
30 years - 3.90%
35 years - 3.75%
40 years - 3.63%

50th Percentile Failure
25 years - 6.29%
30 years - 5.71%
35 years - 5.30%
40 years - 4.76%

95th Percentile Failure
25 years - 9.72%
30 years - 9.09%
35 years - 7.23%
40 years - 6.81%

Maximum Values
25 years - 10.64%
30 years - 9.95%
35 years - 7.75%
40 years - 7.29%

Computed using a 60% Equities portfolio (remains constant for the time periods noted).
Minimum means never fails. 5% means fails one time in 20. But how do you compute the maximum (always fails) value? And to four significant digits for 25 years -- I'm impressed.

Why isn't the maximum risk-free failure rate 100%, regardless of the time frame?

When going for the maximum safe failure rate, when should you zero out your EF?

Rather than 60/40, do you recommend some other AA as best to ensure failure? Maybe 50% trifectas and 50% day trading penny stocks? If I hit a trifecta, should I rebalance into Lotto tickets?

Isn't a 95% failure rate a good enough plan for most people? If the market doesn't cooperate, you can always ratchet up your withdrawal rate to improve your odds of failing.

What about mortality risk (the risk you'll die before you manage to spend it all)?
:wink:
miles monroe
Posts: 1279
Joined: Mon Jan 20, 2014 12:14 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by miles monroe »

midareff wrote: That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
thank you for posting the link to that article.

anybody who believes 4% is a swr in todays environment needs to read that article. the sooner the better.
User avatar
NoRoboGuy
Posts: 736
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by NoRoboGuy »

miles monroe wrote:
midareff wrote: That was the historical performance data at the time he did the study. http://www.retailinvestor.org/pdf/Bengen1.pdf here it is. Additionally, a recent article by Dr. Phau determined that the maximum today would be 2.85%. Let me see if I can find that. Here that paper is and downloadable. http://papers.ssrn.com/sol3/papers.cfm? ... id=2201323 If you go to Table 3 Dr. Phau provides WR and AA guidelines for today's market and interest rates with his calculated fail percentages.
thank you for posting the link to that article.

anybody who believes 4% is a swr in todays environment needs to read that article. the sooner the better.
Note this statement at the end of the article:
Few clients will be satisfied spending such a small amount in retirement. It is possible to boost optimal withdrawal rates by incorporating assets that provide a mortality credit and longevity protection. Pfau (2013), for example, estimates that combining stocks with single premium immediate annuities, rather than bonds, provides an opportunity for clients to jointly achieve goals related both to meeting desired lifestyle spending, and to preserving a larger reserve of financial assets. In the absence of some added income protection, there is a high likelihood that low yields will require planners to rethink the safety of a traditional investment based retirement income plan.
This above point is close to the one I emphasize. Ideally, one should save enough to provide a minimum inflation-indexed base standard of living using pensions, social security, SPIAs, and TIPS/Bond ladders, before taking the retirement plunge. Then, hopefully you have planned and saved additionally for discretionary spending that can be withdrawn using a variable withdrawal method. For that, I would consider Guyton/Klinger floor and ceiling.

By the way, kudos to Mr. Kitces for an outstanding article.
There is no free lunch.
heyyou
Posts: 3842
Joined: Tue Feb 20, 2007 4:58 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by heyyou »

Pfau quote
Few clients will be satisfied spending such a small amount in retirement. It is possible to boost optimal withdrawal rates by incorporating assets that provide a mortality credit and longevity protection. Pfau (2013), for example, estimates that combining stocks with single premium immediate annuities, rather than bonds, provides an opportunity for clients to jointly achieve goals related both to meeting desired lifestyle spending, and to preserving a larger reserve of financial assets.
Pfau saw that, but when he delved into it in a later study, the effect of the annuity was spending bonds first, delaying the start of spending from equities. Letting the equities grow longer was the key, not anything intrinsic within the annuity. He did write that target retirement funds with decreasing equity exposures needed empirical justification.

The problem is behavioral, not numerical. All those spreadsheet contortions are for using a SWR based on initial assets, which has sequence of return (SOR) risk. The solution to SOR risk is future adapting of WDs to market returns. Smoothing is enough to work well there, but even that is unattractive to many retirees.

Fresh retirees think someone can calculate a plan (to a hundredth of a percent) to pull a steady, optimal paycheck for several decades from assets invested in the stock and bond markets. The search for certainty about the future continues in spite of the 100% failure rate of previous results. "This time is different" is alive and well for those seekers.
User avatar
NoRoboGuy
Posts: 736
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by NoRoboGuy »

heyyou wrote:Pfau quote
Few clients will be satisfied spending such a small amount in retirement. It is possible to boost optimal withdrawal rates by incorporating assets that provide a mortality credit and longevity protection. Pfau (2013), for example, estimates that combining stocks with single premium immediate annuities, rather than bonds, provides an opportunity for clients to jointly achieve goals related both to meeting desired lifestyle spending, and to preserving a larger reserve of financial assets.
Pfau saw that, but when he delved into it in a later study, the effect of the annuity was spending bonds first, delaying the start of spending from equities. Letting the equities grow longer was the key, not anything intrinsic within the annuity. He did write that target retirement funds with decreasing equity exposures needed empirical justification.

The problem is behavioral, not numerical. All those spreadsheet contortions are for using a SWR based on initial assets, which has sequence of return (SOR) risk. The solution to SOR risk is future adapting of WDs to market returns. Smoothing is enough to work well there, but even that is unattractive to many retirees.

Fresh retirees think someone can calculate a plan (to a hundredth of a percent) to pull a steady, optimal paycheck for several decades from assets invested in the stock and bond markets. The search for certainty about the future continues in spite of the 100% failure rate of previous results. "This time is different" is alive and well for those seekers.
Agreed, and particularly that behavioral mistakes are a significant issue. To combat this, I advocate "annuitizing" (this is in quotes because it means not just SPIAs, but pensions, SS, or other streams of income) the costs needed to maintain a "basic lifestyle" while separately establishing a risk portfolio and using it for variable withdrawals of discretionary spending. This has the added benefit of making behavioral mistakes less likely. If you have flexibility to reduce spending if market conditions are bad, then you have more confidence to stay the course.
There is no free lunch.
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by BahamaMan »

ourbrooks wrote:Given how variable even these planned expenses are likely to be, I've built in a large safety margin to our plan so I think we could handle even a 20% cut in income by cutting back our safety margin, but if what VPW means is periodically cutting back on our safety margin in response to market tantrums, I'm not sure what it buys me.
Don't confuse Withdrawal Amounts with Spending Amounts. two different things. If you're getting Social Security and or Pensions these are not part of your withdrawal amount.

There are two portions of spending, Discretionary and non-discretionary. IMHO, your Discretionary Spending amount should be at least equal or more to be able to retire in the first place. That way if you have to cut back, it will very easy. IOW- If you are living "Paycheck to Paycheck" when retired, you should probably get a job or cut back your standard of living.

I think you should be able to cut spending easily by 50%, if need be or you should not retire.
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

BahamaMan wrote:I think you should be able to cut spending easily by 50%, if need be or you should not retire.
But what if, for instance, 90% of your spending is covered by SSA, pensions and the like? In that case this seems overly conservative.
Dave
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by BahamaMan »

DFrank wrote:
BahamaMan wrote:I think you should be able to cut spending easily by 50%, if need be or you should not retire.
But what if, for instance, 90% of your spending is covered by SSA, pensions and the like? In that case this seems overly conservative.
If I had a Pension like CEOs are getting these days, it would be a piece of Cake!

But, think of it this way. If 90% of your spending is Non-Discretionary, then retirement is not going to be that fun anyway.
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

DFrank wrote:
BahamaMan wrote:I think you should be able to cut spending easily by 50%, if need be or you should not retire.
But what if, for instance, 90% of your spending is covered by SSA, pensions and the like? In that case this seems overly conservative.

There is always a way to find a financial statement in this area that is either too conservative or optimistic, or just not fitting an individual's situation. Of course if 90% of your income is covered by pension(s) and SSA a 50% cut in the 10% remaining could be overly conservative ... but of you Otar'd @ 33X it would probably never have to happen. If you were 40% pension and SSA it would be a brutal, extreme and perhaps horribly life changing reduction.

The key points accomplished by a withdrawal method would be to leave nothing remaining upon your death (if you knew when that was going to be and that includes spousal, partner, etc.), or leaving inheritance amounts matched predetermined amounts afterwards. End result, your resources were fully consumed to support your life(s) and life style. If you live to the preselected age, VPW does that perfectly. Moving past perfect we are going to try and match moving targets of needed withdrawals vs. market performance and increasing WR%, or amounts withdrawn in most scenarios. OK, the doctor just poked me in the eye with a needle and my crystal ball was cloudy anyway so I have no idea what market performance is going to be for the next decade, only what it has been. Will it be another lost decade? Will it become the #1 bull market duration on record? I have no idea and most probably neither does anyone else so lets call it 2% dividends, which is conservative but close to the last 12 months output of a mixed US/International equities portfolio. Let's look at the bond side and for the purpose of this post assume you have 50% 5 year Treasuries yielding 1.6% (I know they are a bit less right now.). If we want to take them to Bengen's 5.1% yield would have to rise 3.5%, dropping NAV on a 5 year treasury 17.5% ( "rule of thumb theoretically"). That's roughly 3.5 years to recover the NAV drop, again theoretically, and perhaps the full 5 years to recover the CPI-U costs (again, 2% theoretical) and retake 0% real change. Might be longer, might be shorter, it's another who knows situation with lots of unknown moving pieces.

Let's get back to the VPW solution, which is a perfect drain if you live the specified duration. .. but nothing is 100% perfect and the VPW does permit meaningful swings in drawdown annual dollars depending on market variation induced portfolio NAV dollar performance changes and allowed other factors. Well, goal #1 is to fully drain at the moment of death (of both, married or partners) of course. We have to add a second goal, which is to smooth withdrawal variations as much as practical, to life needs. This is where Kitces, Phau, Bergen and quite a few others have written papers and done simulations studying methods to do this to maximum effect. It's tough to hit a bulls eye if you don't know where it's going to be when you shoot the arrow. .. goal #3 say's you must be flexible (obviously) since your guess of your needs in 20 years is no better than your guess of the market in 20 years, other than it should be higher and you will be older.

It should be apparent nothing is perfect since the market and our needs don't fit a cookie cutter. 60's and 70's.. at least early 70's .. go-go, let's see what we haven't.. Europe, Africa, Asia.. yadda, yadda... travel, party, have fun.. we are retired. Later 70's and early 80's... slo-go with maybe some more leisurely travel, more medical needs and helping your kids go-go by watching the grand kids (more cards, less carnival). Later 80's on... more medical than anything and it's called no-go. Anyone here have a clue what these expenses are going to be for them next 25-30 years? .. only so much gets covered by insurance despite their large print. Private assistants 24/7 are a different expense.

To sum it all up... we know a perfect drain mechanism (VPW) if we live to the date specified and can tolerate income variations that may be meaningful. After that we want to smooth income to match needs as closely as possible, with maximum draw vs. portfolio depletion vs. needs matching being the name of the game. .. or .. unknown vs. un-guessable vs. un-predictable. If Kitces has a method, it may be a good one. I use something similar on a small part of my portfolio (Roth). I use other drawdown smoothing methods and strategies elsewhere. If you have a clearer crystal ball than mine, and know what equities and bonds are going to do in the next 1, 3, 5, 10 and longer yearly time periods and have crafted a plan to match withdrawals to market performance and your expected personal needs please wire us into your crystal ball. Otherwise, stay flexible.

EDIT NOTE: It's far easier to over accumulate and take a longer trip, buy a nicer car, a better sofa, etc., than it is to cut back.
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

BahamaMan wrote:
DFrank wrote:
BahamaMan wrote:I think you should be able to cut spending easily by 50%, if need be or you should not retire.
But what if, for instance, 90% of your spending is covered by SSA, pensions and the like? In that case this seems overly conservative.
If I had a Pension like CEOs are getting these days, it would be a piece of Cake!

But, think of it this way. If 90% of your spending is Non-Discretionary, then retirement is not going to be that fun anyway.
I didn't say what if 90% of your spending is non-discretionary, I said what if 90% of your spending is covered by SSA, pensions etc. We see quite a few people come to the forum with situations where their pension and SSA covers a large portion if not all of their spending, and I don't think most of them are CEOs.

I'll be more direct: I think your earlier statement that you shouldn't retire unless you can easily cut back on your spending by 50% is way too general, too broad, and probably not sound advice for many people.
Dave
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

midareff wrote:
DFrank wrote:
BahamaMan wrote:I think you should be able to cut spending easily by 50%, if need be or you should not retire.
But what if, for instance, 90% of your spending is covered by SSA, pensions and the like? In that case this seems overly conservative.

There is always a way to find a financial statement in this area that is either too conservative or optimistic, or just not fitting an individual's situation. Of course if 90% of your income is covered by pension(s) and SSA a 50% cut in the 10% remaining could be overly conservative ... but of you Otar'd @ 33X it would probably never have to happen. If you were 40% pension and SSA it would be a brutal, extreme and perhaps horribly life changing reduction.
My point is there are many, many people for whom the advice that "I think you should be able to cut spending easily by 50%, if need be or you should not retire." is inappropriate.
Dave
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by BahamaMan »

DFrank wrote:I'll be more direct: I think your earlier statement that you shouldn't retire unless you can easily cut back on your spending by 50% is way too general, too broad, and probably not sound advice for many people.
And I'll also be more Direct. If you cannot cut back on your spending by 50% (Which means that you have a lot more Non-Discretionary spending than Discretionary) Your Retirement probably won't be much fun! You are basically living Paycheck to Paycheck !

So, I think it is perfectly sound advice.
longinvest
Posts: 4534
Joined: Sat Aug 11, 2012 8:44 am

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by longinvest »

BahamaMan wrote: If you cannot cut back on your spending by 50% (Which means that you have a lot more Non-Discretionary spending than Discretionary) Your Retirement probably won't be much fun! You are basically living Paycheck to Paycheck !
It is refreshing to see someone promoting a comfortable retirement. People are usually told that they can retire the day their portfolio and pensions (e.g. Social Security, etc.) can generate enough income to match their planned spending (much of it not really discretionary) with a 95% confidence level, not leaving much room for error.

I really like the idea of having lots of money for discretionary spending in retirement.
Bogleheads investment philosophy | One-ETF global balanced index portfolio | VPW
User avatar
jimmyq
Posts: 191
Joined: Wed Apr 17, 2013 6:34 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by jimmyq »

While a ratcheting method may be better than a flat 4% method, I'm still not a fan. Retirees may not like the idea of having withdrawals that can decrease, but a variable withdrawal method is so much more efficient that I think it is worth it. I wouldn't go 100% variable, though. I think a retiree will want a certain amount of funds that he/she can count on (SPIA, Pension, SS), and then the rest would be variable. This could be something like 50% annuitized and 50% variable, or some other reasonable percentage that meets their needs.
User avatar
midareff
Posts: 7324
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by midareff »

longinvest wrote:
BahamaMan wrote: If you cannot cut back on your spending by 50% (Which means that you have a lot more Non-Discretionary spending than Discretionary) Your Retirement probably won't be much fun! You are basically living Paycheck to Paycheck !
It is refreshing to see someone promoting a comfortable retirement. People are usually told that they can retire the day their portfolio and pensions (e.g. Social Security, etc.) can generate enough income to match their planned spending (much of it not really discretionary) with a 95% confidence level, not leaving much room for error.

I really like the idea of having lots of money for discretionary spending in retirement.

Full agreement here. .. to borrow an A Team line.. "I love it when a plan comes together." ... and when a six year bull gives it turbo boost, better yet.
DFrank
Posts: 474
Joined: Thu Mar 20, 2014 10:44 pm
Location: North Idaho

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by DFrank »

BahamaMan wrote:
DFrank wrote:I'll be more direct: I think your earlier statement that you shouldn't retire unless you can easily cut back on your spending by 50% is way too general, too broad, and probably not sound advice for many people.
And I'll also be more Direct. If you cannot cut back on your spending by 50% (Which means that you have a lot more Non-Discretionary spending than Discretionary) Your Retirement probably won't be much fun! You are basically living Paycheck to Paycheck !

So, I think it is perfectly sound advice.
Maybe this hinges on one's definition of discretionary. I protect a lot of our spending in the "non-discretionary" portion of my budget which truth is we could live without. Our non-discretionary budget is about 5X the average per capita income where we'll be living, so it's much more than a hand-to-mouth existence given the COL in that area. This drives me to put a plan in place that protects that spending with a very high probability of success.

Could we cut our spending by 50% without compromising truly essential spending? Yes, we could, but that spending level would be about equal to our two SSA benefits plus my small pension, so that would represent a scenario where we lost 100% of our savings. Short of some black swan event that isn't going to happen.

So back to my original point, I don't think about it terms of "we could easily cut our spending by 50%," because while we could do that, that would be too large a sacrifice in our desired quality of life. To use your words, retirement wouldn't be very fun at that point. Instead, we think about a ~25% spending reduction as one that still enables us to enjoy the minimum quality of life we want in retirement, and believe our plan avoids any significant risk of having to cut more than that.
Dave
BahamaMan
Posts: 896
Joined: Wed Oct 01, 2014 5:52 pm

Re: Kitces - The Ratcheting Safe Withdrawal Rate

Post by BahamaMan »

DFrank wrote:Could we cut our spending by 50% without compromising truly essential spending? Yes, we could, but that spending level would be about equal to our two SSA benefits plus my small pension, so that would represent a scenario where we lost 100% of our savings.
Congratulations! Sounds to me like you proved my Point !
Post Reply