cancel
Showing results for 
Search instead for 
Did you mean: 
     
Highlighted
Participant ○

Re: Bengen's 4% Distribution Method


@Mustang wrote:

I was glad that 4% worked. I would never have thought it would..  From the information I have 3% had a 100% success rate for 40 years.  To get to 50 years I would have guessed 2.5% or even lower. 


There is no need to guess when calculated results are available.  Here is a chart of the initial withdrawal rates of the Wellington fund (VWELX) and the Wellesley Income fund (VWINX) for starting years since 1971, the first full calendar year of VWINX, that left an end of 2019 balance equal to the initial balance adjusted for inflation.

 

vwelx-vwinx-2019-pwr.svg

The size of the end of 2019 balance makes it very likely that an account will support future inflation-adjusted withdrawals for years. This is truer for earlier starting years, for which the return of an additional year after 2019 has a smaller effect on the initial withdrawal rate.

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method

OK folks...while you're dancing around the 4% SWR level, here's something to consider...a real life example...my case:

--Currently,  I'm only a couple years away from living 30 years retired.  Retired 1993.  Had three kids in college.  No ROTHs then.  No SWR studies.  No personal computers.

--Initial Portfolio value approx. $650,000.

--Initial withdrawal rate: 7.8%...took for first fourteen years; then reduced takeout...portfolio GREW in size.

--During age 60's, used two substantial HELOCs, partly for living on, while converting Trad to ROTH IRAs.

--Now have a third home; and substantial ROTH.  Paying down HELOCs (3% fixed; and prime minus 3/4).  Portfolio value much larger, and at all time highs this year.

--And I got a hole-in-one on a par 3 golf hole, that I live on, in Florida.  Priceless.

-------------------------------------------------------------------------------------

Maybe I was crazy; but it all worked out very well!

R48

 

 

 

 

0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method

 

@retiredat48 Yes, the last 30 years have been very good.  Earlier in this discussion I was playing around to see what the maximum SWR would be for a 1990-2019 retirement using Wellington Fund..  It was 7 1/2%.  The problem is that the same withdrawal would have lasted only 11 years if the retirement started in 1968. We don't know what the next 30 years will be like.  We need to be dancing around the 4% level until we know more about the future.

Here are the tables.

yearBeg Balwithdrawreturn%End Balinflation%
1990$1,000,000$75,000-2.81$899,0086.1
1991$899,008$79,57523.65$1,013,2283.1
1992$1,013,228$82,0427.93$1,005,0302.9
1993$1,005,030$84,42113.52$1,045,0752.7
1994$1,045,075$86,700-0.49$953,6782.7
1995$953,678$89,04132.92$1,149,2762.5
1996$1,149,276$91,26716.19$1,229,3003.3
1997$1,229,300$94,27923.23$1,398,6861.7
1998$1,398,686$95,88212.06$1,459,9221.6
1999$1,459,922$97,4164.41$1,422,5932.7
2000$1,422,593$100,04610.4$1,460,0913.4
2001$1,460,091$103,4484.19$1,413,4871.6
2002$1,413,487$105,103-6.9$1,218,1052.4
2003$1,218,105$107,62520.75$1,340,9041.9
2004$1,340,904$109,67011.17$1,368,7633.3
2005$1,368,763$113,2896.82$1,341,0963.4
2006$1,341,096$117,14117.97$1,443,9002.5
2007$1,443,900$120,0708.37$1,434,6354.1
2008$1,434,635$124,993-22.3$1,017,5920.1
2009$1,017,592$125,11822.2$1,090,6032.7
2010$1,090,603$128,49610.94$1,067,3621.5
2011$1,067,362$130,4233.85$973,0113
2012$973,011$134,33612.57$944,0961.7
2013$944,096$136,62019.66$966,2261.5
2014$966,226$138,6699.82$908,8230.8
2015$908,823$139,7780.06$769,5060.7
2016$769,506$140,75711.01$697,9752.1
2017$697,975$143,71314.72$635,8492.1
2018$635,849$146,731-3.42$472,3911.9
2019$472,391$149,51922.51$395,5512.3
2020$395,551    

 

yearBeg Balwithdrawreturn%End Balinflation%
1968$1,000,000$75,0007.91$998,1684.7
1969$998,168$78,525-7.83$847,6346.2
1970$847,634$83,3946.4$813,1525.6
1971$813,152$88,0648.88$789,4773.3
1972$789,477$90,97010.88$774,5053.4
1973$774,505$94,063-11.83$599,9468.7
1974$599,946$102,246-17.73$409,45712.3
1975$409,457$114,82225.18$368,8246.9
1976$368,824$122,74523.36$303,5634.9
1977$303,563$128,760-4.38$167,1476.7
1978$167,147$137,3875.32$31,3449
1979$31,344$149,75113.54($134,440)13.3
1980($134,440)$169,66822.58($372,776)12.5
0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method


@PatMorgan wrote:

@Mustang wrote:

I was glad that 4% worked. I would never have thought it would..  From the information I have 3% had a 100% success rate for 40 years.  To get to 50 years I would have guessed 2.5% or even lower. 


There is no need to guess when calculated results are available.  Here is a chart of the initial withdrawal rates of the Wellington fund (VWELX) and the Wellesley Income fund (VWINX) for starting years since 1971, the first full calendar year of VWINX, that left an end of 2019 balance equal to the initial balance adjusted for inflation.

 

vwelx-vwinx-2019-pwr.svg

The size of the end of 2019 balance makes it very likely that an account will support future inflation-adjusted withdrawals for years. This is truer for earlier starting years, for which the return of an additional year after 2019 has a smaller effect on the initial withdrawal rate.


I like the chart.  It shows initial withdrawals for shorter and shorter retirement periods.  I’m trying to figure out how to best use the information.

I've been struggle with an idea.  My only fault with the Modified RMD method (discussed earlier) was that it didn't provide a stable income.  The retiree would have less than planned income during bad years. It would work if the retiree had enough discretionary expenses he could cut during bad years but it wouldn’t if he didn’t.  What I really liked about it is that it gives the retiree more than inflation adjusted increases if the retiree was fortunate enough to retire during a very good period. 

Two retirement periods showed this.  The good one starting in 1990 and the bad one starting in 1968.

Those two periods showed deficiencies in the 4% Distribution Method as well.  For the 1990 starting year the imaginary retiree never took advantage of the ever increasing portfolio balance.  His heirs were really fortunate.  For the 1968 starting year he never cut back on withdrawals and the portfolio failed to last the entire 30 year period.  Both show the need for periodic reviews.

Some say they do annual reviews.  I agree that annual reviews are necessary for some things but how do you see a trend with a single data point?  If I thought that was all I could do I would just accept the Modified RMD method and throw a bunch of money in a savings account to cover bad years.  This isn't Guyton's discretionary fund (discussed earlier) because the money isn't discretionary.  It’s there to cover core expenses during bad years.

I thought Guyton’s modified 4% rule was interesting but he is relying on good years for it to work.  His periodic reviews could result is serious cuts if the retiree doesn’t get them.  That is his flexible spending part.

So what is an easy way to allow greater than inflation increases to a retirement plan intended to give the retiree a stable income (no cuts)?

0 Kudos
Highlighted
Explorer ○

Re: Bengen's 4% Distribution Method


@Mustang wrote:

@PatMorgan wrote:

@Mustang wrote:

I was glad that 4% worked. I would never have thought it would..  From the information I have 3% had a 100% success rate for 40 years.  To get to 50 years I would have guessed 2.5% or even lower. 


There is no need to guess when calculated results are available.  Here is a chart of the initial withdrawal rates of the Wellington fund (VWELX) and the Wellesley Income fund (VWINX) for starting years since 1971, the first full calendar year of VWINX, that left an end of 2019 balance equal to the initial balance adjusted for inflation.

 

vwelx-vwinx-2019-pwr.svg

The size of the end of 2019 balance makes it very likely that an account will support future inflation-adjusted withdrawals for years. This is truer for earlier starting years, for which the return of an additional year after 2019 has a smaller effect on the initial withdrawal rate.


I like the chart.  It shows initial withdrawals for shorter and shorter retirement periods.  I’m trying to figure out how to best use the information.

I've been struggle with an idea.  My only fault with the Modified RMD method (discussed earlier) was that it didn't provide a stable income.  The retiree would have less than planned income during bad years. It would work if the retiree had enough discretionary expenses he could cut during bad years but it wouldn’t if he didn’t.  What I really liked about it is that it gives the retiree more than inflation adjusted increases if the retiree was fortunate enough to retire during a very good period. 

Two retirement periods showed this.  The good one starting in 1990 and the bad one starting in 1968.

Those two periods showed deficiencies in the 4% Distribution Method as well.  For the 1990 starting year the imaginary retiree never took advantage of the ever increasing portfolio balance.  His heirs were really fortunate.  For the 1968 starting year he never cut back on withdrawals and the portfolio failed to last the entire 30 year period.  Both show the need for periodic reviews.

Some say they do annual reviews.  I agree that annual reviews are necessary for some things but how do you see a trend with a single data point?  If I thought that was all I could do I would just accept the Modified RMD method and throw a bunch of money in a savings account to cover bad years.  This isn't Guyton's discretionary fund (discussed earlier) because the money isn't discretionary.  It’s there to cover core expenses during bad years.

I thought Guyton’s modified 4% rule was interesting but he is relying on good years for it to work.  His periodic reviews could result is serious cuts if the retiree doesn’t get them.  That is his flexible spending part.

So what is an easy way to allow greater than inflation increases to a retirement plan intended to give the retiree a stable income (no cuts)?


As I understand it, Bengen established the 4% withdrawal limit as a means of insuring against worst case scenarios, i.e. potential shortfalls during bad times.  Part of the reason it works is because the excess money generated during good times stays in the portfolio to grow additional value that is available during any potential bad times.  None the less, even at 4%, a bad sequence of returns early on is a challenge that could require the retiree to take drastic measures early on to manage the shortfall risk and stay on track.  Winding up with way too much money at the end - well, that is another risk. 

Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method

For me, I would not be withdrawing each month to live from either fund.  They would be bucket 3 funds to be left alone on reinvestment with rebalancing opportunistically.  I would be drawing from bond funds.  My bond fund money would be sized to my comfort level and what's left would be bucket 3 investments.

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method


@retiredat48 wrote:

OK folks...while you're dancing around the 4% SWR level, here's something to consider...a real life example...my case:

--Currently,  I'm only a couple years away from living 30 years retired.  Retired 1993.  Had three kids in college.  No ROTHs then.  No SWR studies.  No personal computers.

--Initial Portfolio value approx. $650,000.

--Initial withdrawal rate: 7.8%...took for first fourteen years; then reduced takeout...portfolio GREW in size.

--During age 60's, used two substantial HELOCs, partly for living on, while converting Trad to ROTH IRAs.

--Now have a third home; and substantial ROTH.  Paying down HELOCs (3% fixed; and prime minus 3/4).  Portfolio value much larger, and at all time highs this year.

--And I got a hole-in-one on a par 3 golf hole, that I live on, in Florida.  Priceless.

-------------------------------------------------------------------------------------

Maybe I was crazy; but it all worked out very well!

R48

 

 

 

 


You were crazy and lucky  :-)

I would have never done it your way because I retired when I was absolutely certain it will work.  I did it only after 23 years of investing with a portfolio size that covers our annual expenses in the mid 20" times and now after 1.5 years, it's at 34 times.  We don't have pensions and will never inherit anything.  It all depends on our portfolio + SS.

You retired at a great time when stocks were exploded and you probably held mostly stocks. That was pure luck. If you retired at the end of 1999, the results would be a lot worse since the SP500 was negative for 10 years.  From 1/1/2000 to 1/1/2010 the SP500 total performance was -9%. 

How did you get a few more houses? did you inherit anything?

In the past, you mentioned that if things would not work you would go back to work and/or call your kids.

 

 

0 Kudos
Highlighted
Participant ○

Re: Bengen's 4% Distribution Method

Mustang
I have not followed all posts in this thread.   But I think it would be interesting to test the periodic review versions of Bengen's  4% rule for periods  k=1,2,3,4,5  and compare them with the original Bengen's rule and some others such as Kitces' ratcheting rule or a hybrid endowment formula.   At the time of review, compare the withdrawal given by the CURRENT rule with the withdrawal given by a new safe withdrawal rate with a shorter horizon and choose rule that gives the higher withdrawal.

Pfau has a table of SAFEMAX by yearly retirement duration and asset allocation (How Much Can I Spend in Retirement?, p. 147)

 
 
 
0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method

After I am 75 or so I plan to look into a reverse mortgage.  I want to have access to all my home equity if I need it.  I don't know if reverse mortgage allows me to pick and choose the cadence of my drawdown.  But I have no kids and my needs come before my niece's.

0 Kudos
Highlighted
Explorer ○

Re: Bengen's 4% Distribution Method

@Mustang 

Mustang: "So what is an easy way to allow greater than inflation increases to a retirement plan intended to give the retiree a stable income (no cuts)?"

Tibbles: The Kitces Ratcheting Rule. Much earlier I reported Pfau's data comparing it with basic Bengen and with the Modified RMD. Here's a link to Kitces' explanation of the rule:

https://www.kitces.com/blog/the-ratcheting-safe-withdrawal-rate-a-more-dominant-version-of-the-4-rul...

 

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method


@PaulR888 wrote:

After I am 75 or so I plan to look into a reverse mortgage.  I want to have access to all my home equity if I need it.  I don't know if reverse mortgage allows me to pick and choose the cadence of my drawdown.  But I have no kids and my needs come before my niece's.


??My understanding is a reverse mortgage requires one to actually take the money up-front.  I don't think you can just tap it on demand.  With a HELOC, you can.  You do not ever need to take the money.

Others have reverse mtge experience, can comment?

R48

 

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method


@FD1001 wrote:

@retiredat48 wrote:

OK folks...while you're dancing around the 4% SWR level, here's something to consider...a real life example...my case:

--Currently,  I'm only a couple years away from living 30 years retired.  Retired 1993.  Had three kids in college.  No ROTHs then.  No SWR studies.  No personal computers.

--Initial Portfolio value approx. $650,000.

--Initial withdrawal rate: 7.8%...took for first fourteen years; then reduced takeout...portfolio GREW in size.

--During age 60's, used two substantial HELOCs, partly for living on, while converting Trad to ROTH IRAs.

--Now have a third home; and substantial ROTH.  Paying down HELOCs (3% fixed; and prime minus 3/4).  Portfolio value much larger, and at all time highs this year.

--And I got a hole-in-one on a par 3 golf hole, that I live on, in Florida.  Priceless.

-------------------------------------------------------------------------------------

Maybe I was crazy; but it all worked out very well!

R48

 

 

 

 


You were crazy and lucky  :-)

I would have never done it your way because I retired when I was absolutely certain it will work.  I did it only after 23 years of investing with a portfolio size that covers our annual expenses in the mid 20" times and now after 1.5 years, it's at 34 times.  We don't have pensions and will never inherit anything.  It all depends on our portfolio + SS.

R48 in blue...Ah, but you hit the nail on the head.  I had a high withdrawal rate, but it didn;t need to last 30 years.  Like most folks, I started getting social security at age 62 (reduced amount-took it);  I also had a pension coming.  So I did not need my portfolio to grow, to survive.  It's just that it did grow even to SS payout time.

You retired at a great time when stocks were exploded and you probably held mostly stocks. That was pure luck. If you retired at the end of 1999, the results would be a lot worse since the SP500 was negative for 10 years.  From 1/1/2000 to 1/1/2010 the SP500 total performance was -9%. 

Fallback was that I was employable, and could go back to work if an adverse situation developed.  So if I was age 48 in year 2000 and retired, then bear market, I could always stop, go back to work, and simply resume.  Also, year 2000 bear market was MOSTLY a high tech bubble that was broken.  I held very little high tech then, and doubt I would not have experienced this -9% you refer to.

How did you get a few more houses? did you inherit anything?

No inheritance.  I spent some of the market gains on down payment.  I lived well-no miser me.  I now have over a million in real estate.  I did have major HELOCs in use for past three decades.  No mortgages.

In the past, you mentioned that if things would not work you would go back to work and/or call your kids.

Yes, but kinda a humorous viewpoint.  Obviously one has Social Security and Pension forever.  I also took a lump sum instead of GE pension.  So that money went to IRA.  As GE tanked over the  years, that lump sum decision looked better and better.

BTW, something never discussed about retiring, and the risk of running adversely on performance and SWRs, is the usefulness of just one month of employment on retirement analyses.  If you have to, you can take temp jobs, such as December/XMAS, allowing you to take out less in down years.  Such jobs abound in Florida...especially during peak-season times.  Many seniors do such work, simply to balance their life.  Find a fun job for a few weeks.  Put all the earned income into IRAs.

R48

 

 

 

 


 

0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method


@Tibbles wrote:

@Mustang 

Mustang: "So what is an easy way to allow greater than inflation increases to a retirement plan intended to give the retiree a stable income (no cuts)?"

Tibbles: The Kitces Ratcheting Rule. Much earlier I reported Pfau's data comparing it with basic Bengen and with the Modified RMD. Here's a link to Kitces' explanation of the rule:

https://www.kitces.com/blog/the-ratcheting-safe-withdrawal-rate-a-more-dominant-version-of-the-4-rul...

 


In our previous discussion I must have skipped over it.  Thank you for posting it again.  It is exactly what I was looking for. I was looking for a way to ratchet up withdrawals without cuts.  I thought that pretending to start retirement with the new portfolio balance every five years might work but this is far better.  Kitces tested this in all historic 30 year retirement periods.  He said:

“…it turns out that any time the account balance ever grows 50% above of its starting value (after withdrawals), the portfolio is already far enough ahead that it won’t be depleted in a 30-year time horizon and there will be extra money left over.’

“Accordingly, a simple way to establish a new, higher income floor is simply to commit that spending will only be increased (above and beyond annual inflation adjustments) once the account balance grows 50% about its initial amount. So for a retiree with $100,000, there’s no extra spending increase until the account value is over $150,000. For a retiree starting with $1,000,000, the target is $1.5M.’

“Thus, for instance, the “rule” might be that any time the account balance is up 50% over the original value, spending is increased by 10% (over and above any ongoing inflation adjustments), but such spending bumps can only occur once every 3 years at most (to avoid having spending ratchet too high too quickly).’

“In fact, the ratchet approach increases spending at least once (if not multiple times) in almost all historical scenarios; the only times a ratchet does not occur are the scenarios that had adverse sequence-of-return events, where the portfolio barely had enough money to last for 30 years in the first place.

“None of the scenarios run out of money, even with the spending increases, nor does spending ever need to be cut to ensure the sustainability for 30 years.”

Thank you.  This is the missing piece.  I’m going to test it myself using Wellington and Wellesley funds with starting points of 1990 and 1968. Thanks again.

0 Kudos
Highlighted
Participant ○

Re: Bengen's 4% Distribution Method

BTW, something never discussed about retiring, and the risk of running adversely on performance and SWRs, is the usefulness of just one month of employment on retirement analyses.  If you have to, you can take temp jobs, such as December/XMAS, allowing you to take out less in down years.  Such jobs abound in Florida...especially during peak-season times.  Many seniors do such work, simply to balance their life.  Find a fun job for a few weeks.  Put all the earned income into IRAs.

R48

 

A typical temporary work for a retired professor is a part time instructor.   Alas, it often pays only about $3000 per course.   On the plus side, complaining students can keep you occupied for a whole semester.
 

Saratoga

 

 

0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method

I tested Kitces modified 4% withdrawal rule.  First a review of the portfolio is still required every year to apply the inflation increase.  But, every third year there is an additional step before submitting the withdrawal.  If after subtracting the inflation adjusted withdrawal the portfolio’s balance is still 50% higher than its starting balance then the retiree can increase that third year withdrawal again by adding 10%.

The 4% rule is designed for the worst case scenario.  It is overly conservative.  Kitces tested his modification over all  of Bengen’s 30 year retirement periods.  The 10% raise is over and above inflation but it is also conservative usually resulting in an ending balance that is higher than the starting investment.

Using Wellington the 1990-2019 retirement period didn’t see the additional raise until 1995 (the second 3-year review).  After that the retiree never failed to get the raise.  Even after the 10% raises the portfolio’s ending balance grew 412% to $5,124,000.

There wasn’t any sense of testing the 1968 starting point.  The portfolio went below its starting balance in the second year and never recovered.  But, I had another spreadsheet that I used to compare to Wellesley Income Fund that started in 1971.  Starting there the portfolio’s gain was 60% by 1986.  It dropped back to 52% in 1987 then jumped back up to 63% in 1988, the 6th three year review.  Even with the additional 10% raises the portfolio stayed above 50% and ended the 30 year period 72% higher with a balance of $1,717,000.

I intend to include this in my succession plan.  Thanks to all who have contributed to this thread.  The only killer now is long term care.  Here is the 1971 spreadsheet.

 

year

Beg Bal

withdraw

return%

End Bal

inflation%

Old Withdrawal

1971

$1,000,000

$40,000

8.88

$1,045,248

3.3

$40,000

1972

$1,045,248

$41,320

10.88

$1,113,155

3.4

$41,320

1973

$1,113,155

$42,725

-11.83

$943,799

8.7

$42,725

1974

$943,799

$46,442

-17.73

$738,255

12.3

$46,442

1975

$738,255

$52,154

25.18

$858,861

6.9

$52,154

1976

$858,861

$55,753

23.36

$990,714

4.9

$55,753

1977

$990,714

$58,485

-4.38

$891,398

6.7

$58,485

1978

$891,398

$62,403

5.32

$873,097

9

$62,403

1979

$873,097

$68,020

13.54

$914,085

13.3

$68,020

1980

$914,085

$77,066

22.58

$1,026,017

12.5

$77,066

1981

$1,026,017

$86,700

2.91

$966,652

8.9

$86,700

1982

$966,652

$94,416

24.55

$1,086,370

3.8

$94,416

1983

$1,086,370

$98,004

23.57

$1,221,325

3.8

$98,004

1984

$1,221,325

$101,728

10.7

$1,239,394

3.9

$101,728

1985

$1,239,394

$105,695

28.53

$1,457,143

3.8

$105,695

1986

$1,457,143

$109,712

18.4

$1,595,359

1.1

$109,712

1987

$1,595,359

$110,918

2.28

$1,518,286

4.4

$110,918

1988

$1,518,286

$127,379

16.11

$1,614,983

4.4

$115,799

1989

$1,614,983

$132,983

21.6

$1,802,111

4.6

$120,894

1990

$1,802,111

$139,101

-2.81

$1,616,280

6.1

$126,455

1991

$1,616,280

$162,344

23.65

$1,797,792

3.1

$134,169

1992

$1,797,792

$167,377

7.93

$1,759,707

2.9

$138,328

1993

$1,759,707

$172,231

13.52

$1,802,103

2.7

$142,340

1994

$1,802,103

$194,569

-0.49

$1,599,657

2.7

$146,183

1995

$1,599,657

$199,822

32.92

$1,860,659

2.5

$150,130

1996

$1,860,659

$204,818

16.19

$1,923,922

3.3

$153,883

1997

$1,923,922

$232,735

23.23

$2,084,050

1.7

$158,961

1998

$2,084,050

$236,691

11.84

$2,066,086

1.6

$161,663

1999

$2,066,086

$240,478

-4.14

$1,750,028

2.7

$164,250

2000

$1,750,028

$271,668

16.17

$1,717,410

3.4

$168,685

2001

$1,717,410

     
0 Kudos
Highlighted
Explorer ○

Re: Bengen's 4% Distribution Method

@Mustang 

Interesting results. To add further perspective to the Kitces rule, I post below an augmented version of what I posted earlier: It’s from a Pfau table apparently no longer online. Pfau assumes a 4% initial withdrawal for each strategy from a $100,000 accumulation, using a 50-50 allocation. The withdrawal is then adjusted annually for 30 years in accord with each strategy. (The results are based on rolling 30-year periods from 1926 to 2015.)

1. Bengen (“Constant Inflation-Adjusted Spending, Baseline”)
Real spending in 10/20/30 years, 90th percentile: $4000/$4000/$4000
Real spending in 10/20/30 years, 50th percentile: $4,000/$4,000/$4000
Real spending in 10/20/30 years, 10th percentile: $4,000/$4,000/$4000
Remaining wealth after 30 years, 90th//50th/10th percentile: $327,400/$124,740/$27,730

2. Kitces Ratcheting Rule
Real spending in 10/20/30 years, 90th percentile: $4,840/$6,440/$9,430
Real spending in 10/20/30 years, 50th percentile: $4,000/$5,320/$7,090
Real spending in 10/20/30 years, 10th percentile: $4,000/$4000/$4000
Remaining wealth after 30 years, 90th/50th/10th percentile: $236,960/$79,500/$25,630

3. Modified RMD Rule (1.24 times the RMD)
Real spending in 10/20/30 years, 90th percentile: $7,510/$11,690/$10,180
Real spending in 10/20/30 years, 50th percentile: $5,080/$6,650/$6,220
Real spending in 10/20/30 years, 10th percentile: $3,200/$3,620/$4,250
Remaining wealth after 30 years, 90th/50th/10th percentile: $70,160/$41,570/$30,250

I certainly prefer rule 2 to rule 1. But being single, childless, and able to cut spending a lot if I had to, I feel the draw of rule 3. And for any of us, rule 3 is commended by these points: Money we enjoy now is money we’re sure to enjoy. In 10 years, (a) we might be dead; (b) we might not be healthy enough to be able to travel or otherwise enjoy spending our money; and/or (c) war, pandemics, environmental catastrophe, political revolution, or zombie apocalypse might have brought an end to the world as we know it.

On the other hand, as misers know, it’s enjoyable just having money, regardless of whether we ever get to enjoy spending it. And we will be glad to have followed the more conservative rule 2 if the 10 years ahead are bad—perhaps because we lose a lot of the real, after-tax value of our portfolios to (d) severe inflation (we might not have another Paul Volker); and/or (e) greatly increased taxes to finance ever greater government spending.

0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method

It has been a long discussion.  I wish to thank everyone who participated.  It really helped me think things through.

To summarize, my wife’s income from other sources will be 25% short of meeting her expenses. I need a simple easy to use withdrawal method that will provide her a stable income with no loss of purchasing power and no years of big cutbacks.

Bengen’s 4% Distribution method seemed to fit the bill but sample testing using our Vanguard Wellington Fund  identified some flaws.  The first was a sequence of return failure.  Starting the 30-year retirement in 1968 resulted in a failure.  The portfolio ran out of money after 25 years.  The investment went negative in year 2 and never recovered.  The only thing that saved it was taking one percentage point less than full inflation increases every year.

The second problem was excess accumulated wealth.  Starting the 30-year retirement in 1990 Wellington’s value increased steadily until the ending balance was 3.75 times its initial balance.  A method was needed to give the retiree more money during his lifetime.

Here is what I’m putting in my succession plan for my wife to use should something happen to me.  For our taxable accounts we are using a modified Bengen 4% Distribution method.  After depositing the insurance she is to take the portfolio’s balance and multiple it times 4%.  That is her initial withdrawal.  For example, a starting balance of $500,000 would result in an initial withdrawal of $20,000.

 At the beginning of the next year, she answers a question. Is the portfolio’s ending balance bigger than the initial balance?  If so then she multiplies the previous withdrawal in dollars by inflation.  If it is smaller then she multiplies it by inflation minus one percentage point.  That is her next withdrawal.  For example, if inflation is 3% and the end of year balance is $515,000 then the next withdrawal would be $20,600 ($20,000 x 1.03).  If the ending balance was $480,000 then the next withdrawal would be $20,400 ($20,000 x 1.02).

Every three years there is an opportunity to ratchet up the withdrawal and take a bonus.  After calculating the next withdrawal she would subtract that from the portfolio’s end of year balance.  If the balance is still 50% higher than the initial balance then she would increase that withdrawal by 10%.  It may take a little time for this to happen.  For example, it’s the sixth year.  The next inflation adjusted withdrawal is calculated to be $23,185.  The portfolio’s end of year balance has grown to $780,000.  $780,000 minus $23,185 is $756,815.  That is 51% higher than the portfolio's initial balance ($756,815/$500,000 = 1.514) so she would give herself a 10% raise.  The next withdrawal would be $25,504 ($23,185 x 1.1).

The following year would be a normal end of year review.  Is the portfolio's ending balance bigger than the initial balance?  The next opportunity for a bonus would in the ninth year.  Is the ending balance 50% bigger than the initial balance after the next withdrawal?  The answers to these two questions sets up the withdrawals to possibly avoid the somewhat rare sequence of return failure and to take advantage of portfolio's long term growth.

I think this approach is simple, easy to use and unless something extraordinary happens it should provide her with a somewhat stable income. (MRDs from the T-IRA will still vary from year to year.)

What do you think?

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method


@Mustang wrote:

It has been a long discussion.  I wish to thank everyone who participated.  It really helped me think things through.

To summarize, my wife’s income from other sources will be 25% short of meeting her expenses. I need a simple easy to use withdrawal method that will provide her a stable income with no loss of purchasing power and no years of big cutbacks.

Bengen’s 4% Distribution method seems to fit the bill but sample testing using our Vanguard Wellington Fund  identified some flaws.  The first was a sequence of return failure.  Starting the 30-year retirement in 1968 resulted in a failure.  The portfolio ran out of money after 25 years.  The investment went negative in year 2 and never recovered.  The only thing that saved it was taking one percentage point less than full inflation increases every year.

The second problem was excess accumulated wealth.  Starting the 30-year retirement in 1990 Wellington’s value increased steadily until the ending balance was 3.75 times its initial balance.  A method was needed to give the retiree more money during his lifetime.

Here is what I’m putting in my succession plan for my wife to use should something happen to me.  We are using a modified Bengen 4% Distribution method.  After depositing the insurance she is to take the portfolio’s balance and multiple it times 4%.  That is her initial withdrawal.  For example, a starting balance of $500,000 would result in an initial withdrawal of $20,000.

 At the beginning of the next year, she answers a question. Is the portfolio’s ending balance bigger than the initial balance?  If so then she multiplies the previous withdrawal in dollars by inflation.  If it is smaller then she multiplies it by inflation minus one percentage point.  That is her next withdrawal.  For example, if inflation is 3% and the end of year balance is $515,000 (a 7.3% return) then the next withdrawal would be $20,600 ($20,000 x 1.03).  If the ending balance was $480,000 (zero return) then the next withdrawal would be $20,400 ($20,000 x 1.02).

Every three years there is an opportunity to ratchet up the withdrawal and take a bonus.  After calculating the next withdrawal she would subtract that from the portfolio’s end of year balance.  If the balance is still 50% higher than the initial balance then she would increase that withdrawal by 10%.  It may take a little time for this to happen.  For example, it’s the sixth year.  The next inflation adjusted withdrawal is calculated to be $23,185.  The portfolio’s end of year balance has grown to $780,000.  $780,000 minus $23,185 is $756,815.  That is 51% higher than the portfolio's initial balance ($756,815/$500,000 = 1.514) so she would give herself a 10% raise.  The next withdrawal would be $25,504 ($23,185 x 1.1).

The following year would be a normal end of year review.  Is the portfolio's ending balance bigger than the initial balance?  The next opportunity for a bonus would in the ninth year.  Is the ending balance 50% bigger than the initial balance after the next withdrawal?  The answers to these two questions sets up the withdrawals to possibly avoid the somewhat rare sequence of return failure and to take advantage of portfolio's long term growth.

I think this approach is simple, easy to use and unless something extraordinary happens it should provide her with a somewhat stable income. (MRDs from the T-IRA will still vary from year to year.)

What do you think?


Sounds way too complicated to me!

My suggestion:  Start retirement with 1 year of withdrawals in your money market account ($20,000, for an initial $500,000 portfolio.)  Each month, take out her $1,666.

During the year, take all diveys/distributions/interest as cash (don't reinvest 'em).  Since you are taking cash out AND putting cash in, your money market account will either go up or down in value (depends upon whether or not your total portfolio distribution yield is greater than, or less than, your withdrawal.)

At the end of the year, rebalance back to your allocation (selling either your stock or you bond fund).  If your portfolio value is higher at the end of the year compared to the beginning of the year, then bring NEXT years withdrawal up to 4% of the current value.  Ifn the value is less than at the beginning (negative portfolio return), then bring next years withdrawal up to last years value ($20,000) using rebalancing cash.

Your 30 year withdrawal will end up being ever increasing, since, in those years whenever the portfolio return is negative, you withdraw the same amount as the previous year.  You won't need to do any inflation adjusting.

ElLobo, de la casa de la toro caca grande
0 Kudos
Highlighted
Contributor ○○○

Re: Bengen's 4% Distribution Method


@ElLobo wrote:

Sounds way too complicated to me!

Way too complicated?  You're joking?  Answering two questions with a yes or no answer is too complicated? Or is it the multiplication and division that you find to be too complicated?  I assure you.  My wife can handle the math. (I also know as a retired engineer you can too.  Probably far better than I can.)

But, perhaps you aren't joking.  You did say in a previous post that you thought anything other than your own approach was too complicated.  And your comments have always been the same.  They are basically do it your way.  Sorry, but I don't want to.  I think this is a better way.

Thank you for taking the time to read my post.  Does anyone else have any comments.

0 Kudos
Highlighted
Frequent Contributor

Re: Bengen's 4% Distribution Method


@Mustang wrote:

@ElLobo wrote:

Sounds way too complicated to me!

Way too complicated?  You're joking?  Answering two questions with a yes or no answer is too complicated? Or is it the multiplication and division that you find to be too complicated?  I assure you.  My wife can handle the math. (I also know as a retired engineer you can too.  Probably far better than I can.)

But, perhaps you aren't joking.  You did say in a previous post that you thought anything other than your own approach was too complicated.  And your comments have always been the same.  They are basically do it your way.  Sorry, but I don't want to.  I think this is a better way.

Thank you for taking the time to read my post.  Does anyone else have any comments.


What's complicated in your yearly inflation adjustment, and all your 'rules' for cutting back whenever returns are negative.

In simple terms, my 'method' is to take a 4% withdrawal of the yearly value of the portfolio while never taking less out than what you took out the previous year.  The 'problem' you, and others, have is that you can't, conveniently, backtest that against past historical return performance!

I've spent a lot of time listening and reading what you, and others, have written, and I have commented whenever I had something to offer, given I've been doing this retirement withdawal stuff for 20 years now.  So don't patronize me, newbie!  8-))

"We are using a modified Bengen 4% Distribution method."

"At the beginning of the next year, she answers a question. Is the portfolio’s ending balance bigger than the initial balance? If so then she multiplies the previous withdrawal in dollars by inflation. If it is smaller then she multiplies it by inflation minus one percentage point."

"Every three years there is an opportunity to ratchet up the withdrawal and take a bonus. After calculating the next withdrawal she would subtract that from the portfolio’s end of year balance. If the balance is still 50% higher than the initial balance then she would increase that withdrawal by 10%. It may take a little time for this to happen."

"The following year would be a normal end of year review. Is the portfolio's ending balance bigger than the initial balance? The next opportunity for a bonus would in the ninth year. Is the ending balance 50% bigger than the initial balance after the next withdrawal? The answers to these two questions sets up the withdrawals to possibly avoid the somewhat rare sequence of return failure and to take advantage of portfolio's long term growth."

Take 4% of the end of the year value of the portfolio and withdraw the greater of that value and last year's withdrawal.  Easy peesy!

ElLobo, de la casa de la toro caca grande
0 Kudos
Announcements