Simple RMD strategy dominates 4% SWR rule

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 7:27 am

CRR brief that demonstrates that the 4% SWR is so poor that it is dominated by a simple RMD strategy that calls for withdrawing an amount equal to the RMD plus interest and dividends.
Rather than attempt the complex calculations necessary to arrive at an optimal strategy for drawing down and spending their retirement savings, retirees rely on easy-to-follow rules of thumb such as the 4-percent rule advocated by some financial planners. This brief suggests that the IRS’ Required Minimum Distribution rules may be a viable alternative. For financial and practical reasons, the effectiveness of the alternative RMD strategy compares favorably to traditional rules of thumb. And a modified RMD strategy does even better.
Link to CRR brief. http://crr.bc.edu/briefs/can-retirees-b ... ributions/

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 7:40 am

The brief mentions it is based on a longer paper (from the references) which gives the mathematical model for "optimal" and each withdrawal strategy:
http://crr.bc.edu/wp-content/uploads/20 ... 10-508.pdf

User avatar
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by LH » Sat Oct 27, 2012 7:47 am

Trouble with complex calculations is garbage in garbage out.

How many of the "complex calculations" 5 years ago assumed negative tips for instance, or zero interest rate policy?

How do you match an unknown humans financial needs 30'years hence?

By using his unknown earnings stream going forward???

It's just multiple guesses. Couple it with negative tips, if it's was not so unfortunate it would be funny.

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 7:55 am

This is hardly a complex calculation. Take your portfolio balance as of Dec 31 and divide by your life expectancy. That gives you a base withdrawal amount for the coming year. The modified strategy is to add to the base withdrawal amount the value of your interest and dividends in the prior year.

I have confidence in you LH. :wink: You can do this calculation. :happy

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
ObliviousInvestor
Posts: 3272
Joined: Tue Mar 17, 2009 9:32 am
Contact:

Re: Simple RMD strategy dominates 4% SWR rule

Post by ObliviousInvestor » Sat Oct 27, 2012 8:25 am

I read this brief recently, but I have not yet read the entire paper.

One thing that's been a stumbling block for me is the use of a utility function. In the endnotes to the brief, they explain that a 3% discount rate is used for the utility calculation for each dollar spent. I think this means that they assume a couple would be 3% happier spending a dollar today than an inflation-adjusted dollar one year from now. (Is that correct?)

It's not that I have anything against the economics concept of utility. But for me to apply this paper's findings to my own finances, I would have to be able to assess how reasonable that assumption is for me personally. Or, for me to be able to use this paper's findings to help other people make retirement spending decisions, I would have to be able to help them assess how reasonable that assumption is for them.

But I find that I have a very difficult time assigning quantitative measurements to my happiness. What does 3% more happiness feel like? Would I be twice as happy to spend a dollar today as to spend an inflation-adjusted dollar 24 years from now? I really have no idea.
Mike Piper, author/blogger

dharrythomas
Posts: 873
Joined: Tue Jun 19, 2007 4:46 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by dharrythomas » Sat Oct 27, 2012 8:47 am

Of course the RMD strategy dominates.

1. There is no RMD until age 70.5 so withdrawals are delayed.The initial RMDs are less than 4% rising to larger %s based on your changing life expectancy as you age.
2. Since the RMD is based on a percentage of the value of the account, simple math says that you can never go to zero following the strategy.

Good Luck.

Harry

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 9:21 am

dharrythomas wrote:Of course the RMD strategy dominates.

1. There is no RMD until age 70.5 so withdrawals are delayed.The initial RMDs are less than 4% rising to larger %s based on your changing life expectancy as you age.
If your goal is to get more at the beginning, 4% is more than your RMD as you state, so 4% dominates the RMD strategy. Where did you get "of course"?

The IRS has RMD tables that show a value for age 60. Where did you get the idea that there was no RMD value before age 70.5? There is even one for age 20.

Code: Select all

Age	Life Expectancy
20	      63.0
http://www.irs.gov/pub/irs-pdf/p590.pdf

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 9:45 am

I think some people are making this strategy more difficult than it is. Let's go thru a simple example.

It is January 1. Fred is single and retired. As of yesterday, December 31 of the previous year, the value of his portfolio was $800,000. Fred looks up the life expectancy of males of his age. His life expectancy is 20 years. Dividends and interest on his portfolio last year were $8,500.

To calculate this year's withdrawal amount Fred divides the value of his portfolio as of 12/31 by his life expectancy and adds to that quotient the value of his dividends and interest last year.

This year's withdrawal amount = $800,000/20 + $8,500 = $48,500

Next January 1 Fred will do the same calculation using the portfolio value at the end of this year, his life expectancy a year from now, and the amount of dividends and interest he will have received this year. This calculation is repeated at the beginning of each year going forward.

The point here is that while this very simple strategy isn't a theoretically optimal strategy or near optimal withdrawal strategy,it nonetheless dominates the 4% rule.

BobK
Last edited by bobcat2 on Sat Oct 27, 2012 9:57 am, edited 2 times in total.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

maitrina
Posts: 30
Joined: Thu Aug 02, 2012 8:20 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by maitrina » Sat Oct 27, 2012 9:56 am

How would someone extrapolate the annual withdrawal percentages following the RMD strategy to a younger age, say 55?
"I'd like to live like a poor man with lots of money." - Pablo Picasso

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 10:01 am

maitrina wrote:How would someone extrapolate the annual withdrawal percentages following the RMD strategy to a younger age, say 55?
You would look up life expectancy for a person of your gender at age 55. You will find this data in a life expectancy table.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 10:02 am

maitrina wrote:How would someone extrapolate the annual withdrawal percentages following the RMD strategy to a younger age, say 55?
Look in the IRS table? As previously posted, the age 20 life expectancy is 63. Divide.

1 / 63 = 0.01587 = 1.587%

For age 55, the single life table shows
55 29.6


maitrina
Posts: 30
Joined: Thu Aug 02, 2012 8:20 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by maitrina » Sat Oct 27, 2012 10:16 am

sscritic wrote:
maitrina wrote:How would someone extrapolate the annual withdrawal percentages following the RMD strategy to a younger age, say 55?
Look in the IRS table? As previously posted, the age 20 life expectancy is 63. Divide.

1 / 63 = 0.01587 = 1.587%

For age 55, the single life table shows
55 29.6
I thought that was how it was done except I'm still confused. The age 55 single life expectancy is 29.6 so the RMD would be 1/29.6 = 0.0338 = 3.38% (that seems high for someone age 55). In the brief referenced by the OP, the annual withdrawal percentage for someone who is 65 is 3.13%. Older person and lower rate? Any idea what I'm missing?

BTW, I'm super appreciative of the knowledge and wisdom shared here. Thanks so much!
"I'd like to live like a poor man with lots of money." - Pablo Picasso

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 10:26 am

maitrina wrote: I thought that was how it was done except I'm still confused. The age 55 single life expectancy is 29.6 so the RMD would be 1/29.6 = 0.0338 = 3.38% (that seems high for someone age 55). In the brief referenced by the OP, the annual withdrawal percentage for someone who is 65 is 3.13%. Older person and lower rate? Any idea what I'm missing?
The example is a couple, and the paper doesn't use the IRS tables (the uniform lifetime table starts at age 70). The paper uses its own table (see the very end).
Note: Individuals are required to follow the RMD rules during the calendar year in which they reach age 70 1⁄2. The withdrawal schedule for younger ages used in this analysis is calculated based on the same life tables used for the RMD rules.
Edit: at least one part of a couple will live as long or longer than a particular one chosen from among the two. That's why the joint life tables withdraw less than the single life tables: there is a longer "life" expectancy.

User avatar
grayfox
Posts: 4885
Joined: Sat Sep 15, 2007 4:30 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by grayfox » Sat Oct 27, 2012 11:51 am

bobcat2 wrote:This is hardly a complex calculation. Take your portfolio balance as of Dec 31 and divide by your life expectancy. That gives you a base withdrawal amount for the coming year. The modified strategy is to add to the base withdrawal amount the value of your interest and dividends in the prior year.

I have confidence in you LH. :wink: You can do this calculation. :happy

BobK
OK, I plan to live forever. Balance / LifeExp = 0

So with modified strategy, I can only spend the interest and dividends.

This sound like the withdrawal strategy from c. 1960.

Rich Cape Cod
Posts: 144
Joined: Mon Feb 28, 2011 12:37 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by Rich Cape Cod » Sat Oct 27, 2012 12:23 pm

BobK, I'm confused (what else is new?). I looked at the worksheet for determining the RMD. It has five lines:

Age
Yr age was reached
Value of IRA on Dec 31 of previous year 70 1/2 reached
Distribution period (life expectancy number)
RMD determined by subtracting value of IRA from Dec 31st by dividing by life expectancy

No where does it speak of the additional calculation of any money generated by the IRA during the previous year time period.

What am I missing?

Rich Cape Cod
(hunkering down waiting for Sandy)
Rich Cape Cod/AZ

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 12:32 pm

Rich Cape Cod wrote:BobK, I'm confused (what else is new?). I looked at the worksheet for determining the RMD. It has five lines:

Age
Yr age was reached
Value of IRA on Dec 31 of previous year 70 1/2 reached
Distribution period (life expectancy number)
RMD determined by subtracting value of IRA from Dec 31st by dividing by life expectancy

No where does it speak of the additional calculation of any money generated by the IRA during the previous year time period.

What am I missing?
You aren't reading the brief. There is no worksheet.
http://crr.bc.edu/wp-content/uploads/20 ... 19-508.pdf
A potential criticism of the RMD rule is that it results in relatively low consumption early in retirement. While this outcome might be optimal for some households, particularly those fearful of rising health care costs, others might prefer greater consumption at younger ages when they are better able to enjoy it. This result could be achieved by a modification to the RMD rule, namely to consume interest and dividends (but not capital gains), plus the RMD percentage of financial assets. To illustrate, a 65-year-old couple with financial assets of $102,000 who received $2,000 of interest and dividends in the last year, would spend $5,130: the $2,000 in interest and dividends, plus 3.13 percent (the age 65 Annual Withdrawal Percentage under the RMD strategy) of $100,000. In contrast, a household following the unmodified RMD rule would spend just $3,130.
It's an RMD based withdrawal strategy, not the IRS RMD.

Rich Cape Cod
Posts: 144
Joined: Mon Feb 28, 2011 12:37 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by Rich Cape Cod » Sat Oct 27, 2012 12:57 pm

Got it now!

Thanks,

Rich Cape Cod
Rich Cape Cod/AZ

User avatar
zaboomafoozarg
Posts: 1930
Joined: Sun Jun 12, 2011 12:34 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by zaboomafoozarg » Sat Oct 27, 2012 2:51 pm

grayfox wrote:OK, I plan to live forever.
good luck w/ that :D

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by 555 » Sat Oct 27, 2012 4:01 pm

bobcat2 wrote:"CRR brief that demonstrates that the 4% SWR is so poor that it is dominated by a simple RMD strategy that calls for withdrawing an amount equal to the RMD plus interest and dividends. "
Rather than attempt the complex calculations necessary to arrive at an optimal strategy for drawing down and spending their retirement savings, retirees rely on easy-to-follow rules of thumb such as the 4-percent rule advocated by some financial planners. This brief suggests that the IRS’ Required Minimum Distribution rules may be a viable alternative. For financial and practical reasons, the effectiveness of the alternative RMD strategy compares favorably to traditional rules of thumb. And a modified RMD strategy does even better.
Link to CRR brief. http://crr.bc.edu/briefs/can-retirees-b ... ributions/

BobK
(1) What is the precise mathematical meaning of "dominated" in this context.

(2) Who are these financial planners who actually "advocate" the 4-percent rule? I've never seen any evidence that such a person exists.

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 4:23 pm

555 wrote: (1) What is the precise mathematical meaning of "dominated" in this context.
There is a score. Low scores are good. The score for either RMD strategy is lower than the score for the 4% rule. Details on how the score is constructed are in the paper, so if you aren't watching football the way I am, you can read the paper.
This comparison uses a measure called Strategy Equivalent Wealth (SEW). The number for each strategy is the factor by which the dollar value of the household’s wealth, at age 65, must be multiplied so that the couple is as well off as a household that follows the optimal strategy. The optimal strategy has an SEW of 1, and the SEWs for the suboptimal strategies are, by definition, greater than 1.
...
For the rules of thumb, the SEW factors range from 1.29 for the life expectancy strategy – the best – to 1.49 for the 4-percent rule – the worst. Interestingly, the RMD approach, with an SEW of 1.39, performs better than the 4-percent rule.
...
the SEW of the modified RMD strategy... At 1.03, it outperforms all the alternatives, including the unmodified RMD rule.
Man, there I go again. I took the time to read the brief for you, and I missed a touchdown!

redhounds
Posts: 160
Joined: Thu Sep 20, 2012 10:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by redhounds » Sat Oct 27, 2012 4:27 pm

maitrina wrote:3.38% (that seems high for someone age 55).
When we talk about the 4% SWR, the usual assumption is that in the first year of withdrawals, you withdraw 4% of your portfolio. Let's pretend you have $1,000,000, so on January 2nd, 2012, you withdrew $40,000. That's the one & only time that you use the "4%" value. In 2013, you will withdraw $40,000 + an inflation adjustment based on the previous year's actual inflation numbers. Let's pretend it was 3% for 2012, so you will withdraw $40,000 + (3% x 40,000) = $41,200. The next year you would do the same, so if inflation was 2.1% for 2013, you would withdraw $41,200 + (2.1% x 41,200) = $42,065, and so on, never adjusting your numbers except for the inflation adjustment, regardless of whether your account balances have shrank or grown along the way.

The strategy discussed based on RMD's bases each year's percentage withdrawal on the current balance and the current life expectancy, so the dollar amount of the withdrawal can vary a great deal if the portfolio does particularly bad or particularly good. For example, if the portfolio was $1,000,000 and your "number" based on your life expectancy is 3.38%. Therefore, you would withdrawn $33,800 at the beginning of the year. Assuming that next year's life expectancy number didn't change from this year's, the new number next year would be 1/28.7, or 3.48%, so you would multiply the period end balance by this number. If stocks were on a tear and your portfolio was up to $1,200,000, you would withdraw $41,760. If it was a tough year and your portfolio had dropped to $800,000, you'd only withdraw $27,840. Therefore, even though it seems like too high of a starting number, the fact that you are readjusting your withdrawal amount based on your actual balance, you hopefully will not be any more likely to run out of money early. Of course, if the 3.xx% isn't enough real dollars to pay your bills, you've got a problem.

Which way is better? Who knows--we can't know for sure until you've either ran out of money or died with a surplus.

Full disclosure--I didn't read the linked article yet, so someone please correct me if I'm wrong on anything here. Thanks!
--Red

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sat Oct 27, 2012 4:37 pm

redhounds wrote: Which way is better? Who knows--we can't know for sure until you've either ran out of money or died with a surplus.
The same is true when comparing the 4% rule to the 8% rule, but very few people advocate the 8% rule. Why? Because while you won't know which one works out best for you as an individual "until you've either [run] out of money or died with a surplus," we have a pretty good idea that the 8% rule won't work out very well for a large number of people out of a population of retirees just starting to withdraw at age 65.

I could flip a fair coin 20 times and get 20 heads, and I won't know if I can until I either get 20 heads or get a tail, but I do know before I start that the odds aren't good. That's all I am looking for in my withdrawal strategy, good odds.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by 555 » Sat Oct 27, 2012 4:42 pm

sscritic wrote:
555 wrote: (1) What is the precise mathematical meaning of "dominated" in this context.
There is a score. Low scores are good. The score for either RMD strategy is lower than the score for the 4% rule. Details on how the score is constructed are in the paper, so if you aren't watching football the way I am, you can read the paper.
This comparison uses a measure called Strategy Equivalent Wealth (SEW).....
Well, that's certainly not the kind of thing that is usually meant by "dominated". Saying that Strategy A dominates Strategy B should be mean that in no circumstance could Strategy A ever do worse than Strategy B. So the it seems OP misuses this term.

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 5:04 pm

red writes.
Full disclosure--I didn't read the linked article yet, so someone please correct me if I'm wrong on anything here. Thanks!
You don't have the recommended strategy of the linked brief quite right. The recommended strategy is

To calculate this year's withdrawal amount you divide the value of your portfolio as of 12/31 by your life expectancy and add to that quotient the value of your dividends and interest last year.

This year's withdrawal amount = [(value of portfolio as of 12/31 of last year)/(your life expectancy)] + (last year's dividends + interest)
or
This year's withdrawal amount = RMD +(last year's dividends + interest)

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
nisiprius
Advisory Board
Posts: 36016
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Simple RMD strategy dominates 4% SWR rule

Post by nisiprius » Sat Oct 27, 2012 5:06 pm

bobcat2, several years ago I fiddled around with RMD and life-expectancy-based strategies and did some spreadsheet exercises and it didn't work at all. It sucked, big-time. Now perhaps I don't understand their strategy.

(A side note: I don't understand the IRS life expectancy table. It not even close to an authentic actuarial table. It seems very optimistic. The IRS says life expectancy at age 65 is 21.0 years, unisex. But the Social Security agency says it is only 19.89 years for women, only 17.19 years for men. What's with that?)

Here's the problem I ran into. The problem is that the longer you live, the longer your expected total years in retirement becomes.

At birth, if you are male, your (SSA table) life expectancy is 75.38 years = 10.38 total years in retirement.
At age 65, it is 17.19 total years in retirement.
At age 75, 75 + 10.65 - 65 = 20.65 years.
At age 80, 80 + 7.90 - 65 = 22.90 years.
At age 85, 85 + 5.69 - 65 = 25.69 years.

So, if you follow a withdrawal plan that is gauged to your remaining life expectancy, you are constantly coming up short, because the longer you live, the greater the number of total years that your portfolio needs to provide for. At age 65, your best estimate is that your money needs to last 17.19 years, but when you reach age 80, you discover that your remaining life expectancy is 7.90 years and in retrospect it now appears that your best estimate is that your money needed to last not 17.19 years but 22.90 years.

It's just a smoothed-out, softer form of the question "If you follow a 4% SWR and it works perfectly, leaving you with precisely $0 at age 95--if you don't die on your 95th birthday, what do you do then?"

Surely anyone can see the danger of a man reaching age 90 and believing he can now safely spend 1/4th of his portfolio every years... or a woman believing she can safely spend 1/5th of it every year? If the "RMD" strategy is workable at all, it must be due to the IRS tables being so much more optimistic than authentic life tables.

At any rate, this just seems like another one of those meaningless exercises in false precision. It completely ignores the Trinity study's comment that the 4% rule is "the selection of a withdrawal rate is not a matter of contract but rather a matter of planning." The Trinity study tried to answer the question of whether a sensible planning guide was in the neighborhood of 7%, as people like Peter Lynch had been saying, and concluded that the planning number should be more like 4%. I just don't think you can pin it down any closer than that.

I think the best thing I've read on the topic of SWR is Taylor Larimore's post. His "rule" is:
"We simply withdrew what we needed and kept an eye on our portfolio balance. Most years our balance went up and we spent the money on vacations, luxuries and charity. When our balance went down we tightened our belt and economized. This is what most people do and it works."
You can expand this to say, "Oh, and by the way, that withdrawal number will be somewhere in the general ballpark of 4%."
Last edited by nisiprius on Sat Oct 27, 2012 5:22 pm, edited 5 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 5:09 pm

Hi 555,

You are right that technically a dominate strategy is always better. In this case we have a strategy that is typically much better rather than always better. So a wordier but technically more accurate title would have been, "Simple RMD strategy typically far outperforms 4% SWR rule".

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

S&L1940
Posts: 1576
Joined: Fri Nov 02, 2007 11:19 pm
Location: South Florida

Re: Simple RMD strategy dominates 4% SWR rule

Post by S&L1940 » Sat Oct 27, 2012 6:12 pm

Quote by nisiprius: "I think the best thing I've read on the topic of SWR is Taylor Larimore's post. His "rule" is "We simply withdrew what we needed and kept an eye on our portfolio balance. Most years our balance went up and we spent the money on vacations, luxuries and charity. When our balance went down we tightened our belt and economized. This is what most people do and it works." You can expand this to say, "Oh, and by the way, that withdrawal number will be somewhere in the general ballpark of 4%."

This is exactly what we do. At 73 & 72 we have a good idea what we need for the lifestyle we created. SS being our only cash stream, we simply draw down the cash (which is larger than the RMD amount) we need to reach our budgeted cost of living.
When pulling out too much based on our asset projections, we dial back on our spend. A consulting project comes along, we stop the auto withdrawals. Future challenges, health or whatever, can throw us off track but as mentioned here many times, 'it is what it is'. We found we can not go back and change paths. Next time we plan to find Bogleheads.org much earlier in life...
Rich
Don't it always seem to go * That you don't know what you've got * Till it's gone

redhounds
Posts: 160
Joined: Thu Sep 20, 2012 10:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by redhounds » Sat Oct 27, 2012 8:23 pm

sscritic wrote:That's all I am looking for in my withdrawal strategy, good odds.
+1000
bobcat2 wrote:You don't have the recommended strategy of the linked brief quite right. The recommended strategy is

To calculate this year's withdrawal amount you divide the value of your portfolio as of 12/31 by your life expectancy and add to that quotient the value of your dividends and interest last year.

This year's withdrawal amount = [(value of portfolio as of 12/31 of last year)/(your life expectancy)] + (last year's dividends + interest)
or
This year's withdrawal amount = RMD +(last year's dividends + interest)

BobK
Thanks for straightening that out!
--Red

rmark1
Posts: 357
Joined: Sat Jun 16, 2007 7:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by rmark1 » Sat Oct 27, 2012 8:35 pm

'So, if you follow a withdrawal plan that is gauged to your remaining life expectancy, you are constantly coming up short, because the longer you live, the greater the number of total years that your portfolio needs to provide for. At age 65, your best estimate is that your money needs to last 17.19 years, but when you reach age 80, you discover that your remaining life expectancy is 7.90 years and in retrospect it now appears that your best estimate is that your money needed to last not 17.19 years but 22.90 years.'

Except you're not waiting until 80 to discover you might live longer, you're recalculating at 66, 67, 68, etc. Plus your chance of not living to 80 is likely going up.

Retirees drawing from retirement accounts are likely more affluent and longer lived than the population used in Social Security life expectancies.

rmark1
Posts: 357
Joined: Sat Jun 16, 2007 7:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by rmark1 » Sat Oct 27, 2012 10:20 pm

Having read the brief and attempted to read the paper, two things strike me a odd. One, they treat life expectancy based withdrawals as life expectancy at retirement with a 50% failure rate, as if life expectancy doesn't change as you age, while using changing life expectancy in the RMD calculations. Two, the modified RMD withdrawal combines withdrawals from a portfolio with cash flow from within that portfolio, in essence double counting. Withdrawal + cash flow is just a higher overall withdrawal rate.

User avatar
joe8d
Posts: 4277
Joined: Tue Feb 20, 2007 8:27 pm
Location: Buffalo,NY

Re: Simple RMD strategy dominates 4% SWR rule

Post by joe8d » Sat Oct 27, 2012 11:12 pm

Quote by nisiprius: "I think the best thing I've read on the topic of SWR is Taylor Larimore's post. His "rule" is "We simply withdrew what we needed and kept an eye on our portfolio balance."
:thumbsup
All the Best, | Joe

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sat Oct 27, 2012 11:35 pm

rmark1 wrote:Having read the brief and attempted to read the paper, two things strike me a odd. One, they treat life expectancy based withdrawals as life expectancy at retirement with a 50% failure rate, as if life expectancy doesn't change as you age, while using changing life expectancy in the RMD calculations. Two, the modified RMD withdrawal combines withdrawals from a portfolio with cash flow from within that portfolio, in essence double counting. Withdrawal + cash flow is just a higher overall withdrawal rate.
The modified RMD rule takes at least (3) things explicitly into account that the 4% rule does not account for.

1) the present value of your portfolio each year
The 4% rule OTOH assumes you are on target from the present value of the portfolio in the first year of the retirement plan, which is a glaring weakness of the 4% rule. The value of your portfolio assets could be far off target several years into the plan. This cannot happen when each year's spending is based on the present value of the portfolio at the beginning of that year.

2) life expectancy each year you are alive
If the 4% rule was correct 15-20 years ago for a 65 year old, it cannot possibly be correct this year, because life expectancy at age 65 is nearly two years greater now than it was 20 years ago.

3) the current level of interest rates
If interest rates are high in a given year you can safely withdraw more from your portfolio than if interest rates are low. The modified RMD rule takes current interest rates explicitly into account by having you withdraw the amount of interest and dividends.* The 4% rule takes no account of prevailing interest rates, which is one of the many weaknesses of the 4% rule.

In a nutshell what is odd is the 4% rule, which violates just about every principle of basic finance. :D

BobK

* The percentage of interest and dividends you withdraw can range from zero percent to 100% in the RMD strategy. For many people 100% will be fine, but for those conservative retirees worrying about running low on income should they survive for 25 years or more, a percentage close to zero or zero will be more appropriate.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by 555 » Sat Oct 27, 2012 11:42 pm

Yes, but the 4% rule was never intended as a decumulation strategy. It's an accumulation strategy, designed to give you a ballpark figure for how much you need to save.

User avatar
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by LH » Sun Oct 28, 2012 2:27 am

Rather than attempt the complex calculations necessary to arrive at an optimal strategy for drawing down and spending their retirement savings, retirees rely on easy-to-follow rules of thumb such as the 4-percent rule advocated by some financial planners. This brief suggests that the IRS’ Required Minimum Distribution rules may be a viable alternative. For financial and practical reasons, the effectiveness of the alternative RMD strategy compares favorably to traditional rules of thumb. And a modified RMD strategy does even better.
bobcat2 wrote:This is hardly a complex calculation. Take your portfolio balance as of Dec 31 and divide by your life expectancy. That gives you a base withdrawal amount for the coming year. The modified strategy is to add to the base withdrawal amount the value of your interest and dividends in the prior year.

I have confidence in you LH. :wink: You can do this calculation. :happy

BobK
BobK,

The "complex calculations" is a direct reference to, well, the "complex calculations" including in the original post....... If I refer to "complex calculations" and you put "complex calculations" in your post....... Well..... To be explicit, Its not the noncomplex calculations I am refering to..... Its the "complex calculations" you referenced I am taking about.

Garbage in, garbage out.

Pure and simple.

I have faith in you that you will see you have clearly made a logical error, which I assume is not purposeful, in assuming, that I am not directly addressing the "complex calculations" you referenced in the post, when I say "complex calculations"............. Right?

Again. Any attempt at "complex calculations", to achieve an "optimal" solution, is near nonsensical, and conflates, at minimum, the ability to average out groups of people, to achieve some sort of half assed ok guess for large groups of people, with that of the ability to achieve meaningful inputs for an individual.....

For an individual, is all a wild ass guess:

1)what will the individual earn in his lifetime (for "income smoothing")
2)what will the individual need 30 years hence
3) TIPS are the risk free investment...... really.... how is that working out now with negative yields tens years out?

Its garbage in, garbage out on all fronts. The inputs are garbage. You could make up near anything for a given person, and the end result is gonna have HUGE error bars expectantly. Its like trying to use the Gordon equation to run a mutual fund (tracked=fail ). Its product to be sold.

Then BobK, even for large groups of people you can statistically average..... How are defined benefit plans doing????????????????????????? Oh, yeah, all sorts of "complex calculations" there. Call in the actuary! But same problem, garbage in, garbage out.

Failure.

Rules of thumb are used for good reason. We are limited by the uncertainty of inputs to "Complex equations" such that they are near meaningless ex ante.


Now if you are saying your quote about "complex equations" is false, they are actually noncomplex, then fine I guess. Complex=noncomplex would go along with the whole approach of "matching" highly imprecise/inaccurate needs with highly imprecise/inaccurate income streams over long periods of time, and pretending you have a precise meaningful answer. Complex=non-complex. Precise=non precise. That would fit.

I would just love to see some matching "math" EXAMPLE applied pre 2007, written in 2007, using "riskless" TIPS and such to achieve some sort of "safe" outcome, just in the current short medium term..... with negative TIPS yields coming in the near unknown future to screw up the projections, not to say anything of the uncertain longer term. It would almost certainly be failing, TIPS at negative yields would blow any pre 2007 TIPS model out of the water (should we assume 3 percent real TIPS return? one could laugh or cry), just like falling home prices did subprime.

Complex equations are a mirage, they have near zero utility, or as much meaningful utility as the 1959 Gordon equation did in predicting much of anything since 1959 ex ante. I do not recollect the Gordon equation referenced in the in hindsight rosy IMF 2007 report for instance. http://www.imf.org/external/pubs/ft/weo ... f/text.pdf

If it was referenced, and the inputs they chose made it come out bad, its was not given any weight.

The trouble with "complex" financial calculations applied to achieve an "optimal" result is the garbage in garbage out input problem. The inputs are just expectantly wild inaccurate guesses. Which is why people use rules of thumb.


4 percent, RMD are both rule of thumb approaches as I read the your referenced paper. The paper uses a "complex calculation" as the benchmark, as I understand the phraseology, but does not appear to validate it, its simply made true by definition, which is very convenient, but quite likely does not describe any meaningful reality.

Anyway, as always, I enjoy reading your posts BobK. Its all good theory, but I cannot see how much of it is ready for prime time application? There just seems to be a schism between real world practicality, and theorectical models used in finance papers. LTCM, 100 percent TIPS ladders, and portfolio insurance were fine on paper, and this stuff looks great too.... but real world is not true by definition, its messy.


LH

User avatar
bobcat2
Posts: 5201
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Simple RMD strategy dominates 4% SWR rule

Post by bobcat2 » Sun Oct 28, 2012 7:45 am

555 wrote:Yes, but the 4% rule was never intended as a decumulation strategy. It's an accumulation strategy, designed to give you a ballpark figure for how much you need to save.
The 4% rule is also a failure as an accumulation strategy.

The goal of retirement planning is to meet an aspirational level of retirement income. But the amount of financial assets needed at retirement to fund a given level of retirement income is conditional on the level of real interest rates. For instance, if real interest rates are close to average it will take $1 million at retirement to fund a given level of real retirement income. But if interest rates are significantly higher than average, it might take less than $900,000 to fund that same level of retirement income. Conversely, if interest rates are significantly lower than average, it might take more than $1,100,000 to fund that same level of retirement income. The 4% rule completely ignores the level of real interest rates.

Because life expectancy at age 65 is increasing at roughly one month per year, the 4% rule cannot be correct over time as an accumulation rule. If it was correct for a 65 year old in 1994, it will not be anywhere close to correct for someone today planning to retire at age 65 in 2034.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

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

Re: Simple RMD strategy dominates 4% SWR rule

Post by Browser » Sun Oct 28, 2012 7:52 am

The study authors are advocating a withdrawal strategy based on the IRS RMD schedule in which the portfolio withdrawal rate increases with age, which seems to make sense based on declining life expectancy. But unless I'm missing something, they're referencing nominal withdrawal rates. By comparison, the 4% rule is based on maintaining a relatively constant real withdrawal rate; however, unless there is no inflation, the nominal withdrawal rate under the 4% rule will increase with age as well. So, what is the difference between these two withdrawal strategies, other than the trajectories of their effective nominal withdrawal rates? The 4% rule seems to have the virtue of trying to maintain a fixed, constant level of spending power. The RMD strategy would result in a variable and unpredictable real spending rate. Is this an apples vs. oranges comparison?
Last edited by Browser on Sun Oct 28, 2012 8:16 am, edited 1 time in total.
We don't know where we are, or where we're going -- but we're making good time.

john94549
Posts: 4638
Joined: Tue Jul 26, 2011 8:50 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by john94549 » Sun Oct 28, 2012 8:14 am

Two observations from an actual retiree:

(1) the 4% "guideline" was a very handy planning tool during the accumulation phase* (and still is, as my wife is still working); and

(2) the "modified" RMD formula is not only elegant, it is practical, as many retirees still tilt to FI. It stands to reason you should tighten up spending a tad when those CDs aren't doing so well.

As with any withdrawal strategy, the "modified" RMD path requires a dose of flexibility and common sense.

Thanks for this discussion. I'm jotting down the "modified" RMD for inclusion in our retirement binder.

*As in, "if I retired today, could I pay the bills with a 4% withdrawal".
Last edited by john94549 on Sun Oct 28, 2012 8:34 am, edited 1 time in total.

rmark1
Posts: 357
Joined: Sat Jun 16, 2007 7:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by rmark1 » Sun Oct 28, 2012 8:26 am

3) the current level of interest rates
If interest rates are high in a given year you can safely withdraw more from your portfolio than if interest rates are low. The modified RMD rule takes current interest rates explicitly into account by having you withdraw the amount of interest and dividends.* The 4% rule takes no account of prevailing interest rates, which is one of the many weaknesses of the 4% rule.'

Did you see a recommended or model portfolio in the study? I didn't find one in the brief. Maybe I shouldn't be trying to read retirement studies after spending a day at a football game :)

The 4% rule has no upside potential nor downside warning, and safe rates have varied considerably
RMD based tends to under withdraw, unless your spouse is in fact 10 years younger than you are
'Buckets' are mostly just spending from bonds first with some trigger to refill from the growth side
Interest and dividends may limit current spending while leaving a larger estate (fine if you want that)
And most research uses only one data set, so you can't come up with too much that's very different

Personally, my planning uses Heblers autopilot, as it uses life expectancy and incorporates a pension easily.

User avatar
magellan
Posts: 3469
Joined: Fri Mar 09, 2007 4:12 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by magellan » Sun Oct 28, 2012 10:15 am

ObliviousInvestor wrote:But I find that I have a very difficult time assigning quantitative measurements to my happiness. What does 3% more happiness feel like? Would I be twice as happy to spend a dollar today as to spend an inflation-adjusted dollar 24 years from now? I really have no idea.
IMO, Mike has identified the core problem with the brief. The original paper that's referenced in the brief relies on something called the CRRA (constant relative risk aversion) utility model that's described on page 533 of this paper.

A retired couple's 'inter temporal elasticity of substitution' (eg preference for spending now vs later) is tied to their overall level of risk-aversion and is assumed to be a constant throughout retirement. It's also assumed that the couple has no bequest motive.

Even if the authors could accurately estimate something as fuzzy as a couple's inter-temporal elasticity of substitution, it seems unlikely that it would be constant throughout retirement. Given these challenges, the SEW values that the authors report out to two decimal places probably have an error of a hundred percent or more.

This is another case of well-meaning academics doing useful and interesting theoretical work, then sloppily serving up their results into half-baked practical applications.

Jim
Last edited by magellan on Sun Oct 28, 2012 10:21 am, edited 2 times in total.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by dbr » Sun Oct 28, 2012 10:16 am

Two comments:

The 4% rule is a great strawman. Arguing against it is a truly magnificent contribution to the literature.

The single best utility function that exists is the ability to spend what one thinks one needs at any point in time. I side with Taylor on this one.

User avatar
nisiprius
Advisory Board
Posts: 36016
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Simple RMD strategy dominates 4% SWR rule

Post by nisiprius » Sun Oct 28, 2012 10:30 am

There's another issue here. The big advantage of the framing of the 4% rule is that you have a clear figure of merit: did you run out of money or not? Now, it is utterly unrealistic to suppose that someone is going to follow a withdrawal rule inflexibly into the ground. But at least everyone can agree that if inflexible strategy A fails 3% of the time and inflexible strategy B fails 10% of the time, then A is a better unrealistic inflexible strategy than B.

The problem is that almost any strategy that results in adjusting withdrawals can reduce or even eliminate outright failure--and that there's no obviously right way to compare the merits of two variable streams of income. For example, the simple rule of withdrawal 4% of the current portfolio value at the start of the year will never fail. Neither will the simple rule of withdrawing 50% of the current portfolio! Obviously, a strategy that leaves you to get through the year on a withdrawal of thirteen cents isn't a very good one. But it's not so clear how you judge between different strategies that give you different fluctuating levels of income. (For example, one that gives you balanced swings between feast and famine, versus one that gives you lower average income, but skewed swings of generous feast and mild famine).
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
k66
Posts: 458
Joined: Sat Oct 27, 2012 1:36 pm

Re: Simple RMD strategy dominates 4% SWR rule

Post by k66 » Sun Oct 28, 2012 11:07 am

Browser wrote:The study authors are advocating a withdrawal strategy based on the IRS RMD schedule in which the portfolio withdrawal rate increases with age, which seems to make sense based on declining life expectancy. But unless I'm missing something, they're referencing nominal withdrawal rates. By comparison, the 4% rule is based on maintaining a relatively constant real withdrawal rate; however, unless there is no inflation, the nominal withdrawal rate under the 4% rule will increase with age as well. So, what is the difference between these two withdrawal strategies, other than the trajectories of their effective nominal withdrawal rates? The 4% rule seems to have the virtue of trying to maintain a fixed, constant level of spending power. The RMD strategy would result in a variable and unpredictable real spending rate. Is this an apples vs. oranges comparison?
While it is true that any percentage based program includes a mechanism for adjusting for inflation, I think that there is also an inherent and subtle means of accounting for inflation in a RMD-type program as well. If we assume that a retirement portfolio can maintain a certain rate of return above inflation, then pulling off the prescribed distribution from that portfolio (regardless of the particular distribution value) includes compensation for inflation as well.

I think you are also correct in that an advantage could also be a disadvantage -depending on how you choose to view it. The annual distribution will necessarily vary. As some have noted here earlier, perhaps that is the best way to manage your portfolio anyway: allow for more spending in good years and economize during the lean.

Another point that occurs to me, is that there is no requirement to spend the entirety of a given year's worth of distributions. One could always choose to keep "over-distributions" in a tax sheltered account until it may be needed later.
LOSER of the Boglehead Contest 2015 | lang may yer lum reek

rmark1
Posts: 357
Joined: Sat Jun 16, 2007 7:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by rmark1 » Sun Oct 28, 2012 11:31 am

Divide portfolio annually over remaining IRS life expectancy, spending up to the lesser of the new division or the previous yearly division plus inflation.

williamg
Posts: 225
Joined: Thu Mar 01, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by williamg » Sun Oct 28, 2012 11:34 am

Any one know how the IRS comes up with the RMD table numbers? 27.4 is not the projected life expectancy of a 70 1/2 year old.

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sun Oct 28, 2012 11:47 am

williamg wrote:Any one know how the IRS comes up with the RMD table numbers? 27.4 is not the projected life expectancy of a 70 1/2 year old.
But was it when the tables were developed? How often does the IRS change the tables? Have they ever changed the tables?

There, I gave you three questions for one of yours. :)

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sun Oct 28, 2012 11:50 am

sscritic wrote:
williamg wrote:Any one know how the IRS comes up with the RMD table numbers? 27.4 is not the projected life expectancy of a 70 1/2 year old.
But was it when the tables were developed? How often does the IRS change the tables? Have they ever changed the tables?

There, I gave you three questions for one of yours. :)
Let me help you.
Several years ago, the IRS revised distribution rules and the various life expectancy tables used to make the RMD calculations. The reformulation means that taxpayers now have to take out less.
So the life expectancies used to be even lower (higher RMDs).

rmark1
Posts: 357
Joined: Sat Jun 16, 2007 7:43 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by rmark1 » Sun Oct 28, 2012 12:21 pm

IRS 590, Table 3 is a minimum distribution table, not a life expectancy table, and assumes a spouse 10 years younger. 27.4 years is the joint life expectancy of a 70 year old with a 60 year old spouse, it can be located in table 2.

sscritic
Posts: 21858
Joined: Thu Sep 06, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by sscritic » Sun Oct 28, 2012 12:24 pm

This isn't that easy to track down. There were proposed regulations in 2001 that used 26.2, but the final regs issued April 17, 2002, used 27.4.

http://www.irs.gov/pub/irs-regs/td8987.pdf

Proposed:

Code: Select all

 Age of employee Distribution period 
        70               26.2       
http://www.gpo.gov/fdsys/pkg/FR-2001-01 ... 01-304.pdf

The proposed regs and the final regs both use the terms Age of the employee and Distribution period. Note that Life Expectancy isn't the issue, but the period over which the distributions are to be made (which while looking fixed, isn't).
Q-2. What is the applicable distribution period for an individual account for purposes of determining required minimum distributions during an employee’s lifetime under section 401(a)(9)?
So how did we get the 27.4?
New Mortality Tables
The 2001 proposed regulations provided that the life expectancies for purposes of section 401(a)(9) would be determined using the expected return multiples set forth in the regulations under section 72 that are used for other purposes under the Code. These tables, based upon the experience reflected in the 1983 individual annuity mortality table (without load), were adopted for purposes of section 72 in 1986 and had been used in both the 1987 proposed regulations and the 2001 proposed regulations under section 401(a)(9).

Section 634 of EGTRRA instructed the Secretary of Treasury to modify the life expectancy tables used for purposes of the minimum distribution rules to reflect current life expectancy. In accordance with that instruction, the final regulations adopt new tables of life expectancies to be used for determining required minimum distributions. The new tables were derived by starting with the basic 2000 individual annuity mortality table and projecting mortality improvement for the period 2000 through 2003 using the assumed mortality improvement factors that were adopted in developing the Annuity 2000 mortality table. The resulting mortality rates were blended using a fixed 50% male 50% female blend. The uniform lifetime table provided in these final regulations has also been adjusted to reflect these new mortality tables.
Edit: plus what rmark1 said about the joint table with the 10 year younger spouse. This is also in footnote 12 of the full paper.
12 They are the Table 2 values for spouses exactly 10 years younger than the account holder.

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

Re: Simple RMD strategy dominates 4% SWR rule

Post by Browser » Sun Oct 28, 2012 12:51 pm

Neither one (RMD or life expectancy tables) are good mortality risk tables. What I want to do is to incorporate declining mortality risk as accurately as possible into my portfolio withdrawal strategy. But what I really want to do is to figure out how to maximize spending at the front end, while controlling the risk of going broke too soon at the back end. A strategy that increases withdrawal rate with age is backwards, and actually seems behaviorally counter-intuitive. As a geezer, I won't be ramping up my spending unless for healthcare expenses or some similar dire circumstance. Won't be needing those Harleys or trips to Vegas anymore. Frankly, I'd rather take my portfolio survival risk earlier rather than later. With any luck my longevity risk will show up and everything will work out anyway.

But the real deal is that I just don't see how the heck any systematic withdrawal strategy is going to come close to optimality whether it's based on a constant real SWR rule, RMD rule, or the Rule of All Rules. As has been pointed out elsewhere, one should have two retirement investment portfolios: a liability-matching portfolio and (if there's anything left over) a systematic withdrawal portfolio. As for the SWP, I'm not sure it makes all that much difference how you schedule your withdrawals, since you don't know ex-ante what is optimal anyway. If you picked a crappy strategy, you still have the liability-matching portfolio anyway, so Life Is Good.
We don't know where we are, or where we're going -- but we're making good time.

williamg
Posts: 225
Joined: Thu Mar 01, 2007 8:36 am

Re: Simple RMD strategy dominates 4% SWR rule

Post by williamg » Sun Oct 28, 2012 1:11 pm

Thanks for the feedback on RMD table entries.
Numbers is hard.

Post Reply