95% rule and the 4% SWR

[intercst]

I've been reading Bob Clyatt's book "Work Less, Live More" with great interest -- especially where he describes the work done by Keith Marbach on the "95% rule and the 4% SWR".

One item that's unclear to me is the table on page 196 where he compares a "Standard" withdrawal with the "95% rule". For withdrawal rates above 4%, the "Standard" method shows higher survivability than the "95% rule". For example, for a 30-year pay out and 4.5% withdrawal, the standard method had a 93.9% success rate, while the "95% rule" only shows a 91.8% success rate. Since the 95% rule dictates withdrawing less money in years where the market is down, it doesn't make sense that it would result in diminshed success.

Has anyone else noticed this, or perhaps the question has already been answered?

intercst

 

[SaveSome]

I've read the book as well. I understood the 95% rule to actually permit greater withdrawals than the standard 4% withdrawal rate would allow in down years. In a down year you can take 95% of your prior year's withdrawal even if the amount exceeds the 4% rate. Page 187 lays out the 95% rule.

 

[intercst]

I've read the book as well.  I understood the 95% rule to actually permit greater withdrawals than the standard 4% withdrawal rate would allow in down years.  In a down year you can take 95% of your prior year's withdrawal even if the amount exceeds the 4% rate.  Page 187 lays out the 95% rule.

Thanks. I see my mistake. The 4% rule  they use is "4% of the year-end balance, each year" rather than the "4% of initial balance, adjusted for inflation" used by the Trinity study, Bengen, etc.

intercst

 

[retire@40]

Thanks. I see my mistake. The 4% rule  they use is "4% of the year-end balance, each year" rather than the "4% of initial balance, adjusted for inflation" used by the Trinity study, Bengen, etc.

An even more conservative approach may be to take the lower of 4% of the initial balance adjusted for inflation or 4% of the year-end balance each year.

And come to think of it, an even more conservative approach is to take the lower of the above using 0.3333 (4% annualized) of the previous month balance instead of 4% of the previous year-end balance.

 

[brewer12345]

An even more conservative approach may be to take the lower of 4% of the initial balance adjusted for inflation or 4% of the year-end balance each year.

And come to think of it, an even more conservative approach is to take the lower of the above using 0.3333 (4% annualized) of the previous month balance instead of 4% of the previous year-end balance.

Or to take no withdrawal at all.

 

[SteveR]

Or to take no withdrawal at all.

Now that would be conservative.!

It would be nice to have a FIRE calc.-like spreadsheet where you can make more specific adjustements to income than are provided. I could write one but I don't have the time or the energy to do it. I know I will need a high cashflow at the start of ER but a much lower one later on so I am adjusting my income streams to account for this and as a way to lower taxes in later life by having smaller RMDs due to reducted balances before age 70.5.

 

[Jarhead*]

Now that would be conservative.!

I know I will need a high cashflow at the start of ER but a much lower one later on so I am adjusting my income streams to account for this and as a way to lower taxes in later life by having smaller RMDs due to reducted balances before age 70.5.   

SteveR:  I have had more than a passing interest in RMDs, since I will have to begin in a year or so.  (I'd still be young if I were a Pine Tree.)

Anyway, as I'm sure you're aware, the formula for RMD has been stretched, (about a year and a half ago), and is no longer as punitive re: minimum withdrawals as they were prior to the change.

On a $1,000,000 IRA, the prev. RMD for first year was $62,500.00.

Current rules require first year withdrawal of
$36,496.00

(27.4 versus prev. 16).

 

[Nords]

It would be nice to have a FIRE calc.-like spreadsheet where you can make more specific adjustements to income than are provided.

Dory, that's an interesting question.

Now that FIRECalc has the code to allow for three changes to withdrawals, how much pain is it to add modules for five, 10, or even 30 changes to withdrawals?  Could TH happily program in all his roof/appliance/vehicle/college/wedding/vacation purchases for the next five decades?

I feel it's especially appropriate to ask the question as we approach the traditional time of year to make a six-pack contribution to the Dory server fund...

SteveR: I have had more than a passing interest in RMDs, since I will have to begin in a year or so. (I'd still be young if I were a Pine Tree.)

Jarhead, I've been meaning to ask you about that. A couple weeks ago I stumbled across some DoD military-retiree statistics from 2004, including one retiree who (at that time) was still receiving his pension at the ripe ol' age of 110. Is there something that you're trying to tell us?

 

[Jarhead*]

Jarhead, I've been meaning to ask you about that.  A couple weeks ago I stumbled across some DoD military-retiree statistics from 2004, including one retiree who (at that time) was still receiving his pension at the ripe ol' age of 110.  Is there something that you're trying to tell us?
[/quot)

Nords: A lottery winner!

No, my goals are not that ambitious.

I'd like to shoot my age and be able to play tournament golf into my 80's. (I have to really suck it up anymore to work out enough to stay flexible).

The RMD stretch-out (I hope) will allow my wife and I to break tradition with my ancestors, and leave something for my kids other than un-paid bills.

By the way, you missed your calling! I could see you as the World Champion Trivial Pursuit player, or the League Statistician for MLB. (Doubt if you'd have much competition).

 

[SteveR]

SteveR:  I have had more than a passing interest in RMDs, since I will have to begin in a year or so.  (I'd still be young if I were a Pine Tree.)

Anyway, as I'm sure you're aware, the formula for RMD has been stretched, (about a year and a half ago), and is no longer as punitive re: minimum withdrawals as they were prior to the change.

On a $1,000,000 IRA, the prev. RMD for first year was $62,500.00.

Current rules require first year withdrawal of
$36,496.00

(27.4 versus prev. 16).

Thanks JarHead.  I have run the numbers several differnet ways and still need to reduce my overall IRA balances by the time I get to 70.5 to keep the taxes down over the life of the IRA withdrawl.  The first few years are not the problem....the last 10 (assuming I live that long) do create a nasty tax hit and more income that I could possibly need.  

Of course, it is all theoretical at this point but with two current roll over IRAs and two more in a couple of years....after a while it gets to be some real money.  

 

[ESRBob]

I've read the book as well. I understood the 95% rule to actually permit greater withdrawals than the standard 4% withdrawal rate would allow in down years. In a down year you can take 95% of your prior year's withdrawal even if the amount exceeds the 4% rate. Page 187 lays out the 95% rule.

Thanks, Savesome-- exactly right;
Intercst did that answer your question? If not let me know and I'll answer here. The 95% Rule was designed to smooth out the withdrawals in a system which, while giving great long-run survivability, does have ups and downs in annual withdrawals. Having some sort of part-time paid work you can crank up and down to fill in the gaps fits well into this system.

Here's what I'm doing these days to earn spare cash -- people in our neighborhood are starting to pay me to make scuptures of their kids -- tons of fun and so far steady demand. http://clyattsculpture.com/customportraits.html This has been my real FIRE dividend-- something I never knew I could do until stumbling into it a few years ago.

 

[intercst]

Thanks, Savesome-- exactly right;
Intercst did that answer your question?  If not let me know and I'll answer here.  The 95% Rule was designed to smooth out the withdrawals in a system which, while giving great long-run survivability, does have ups and downs in annual withdrawals.  Having some sort of part-time paid work you can crank up and down to fill in the gaps fits well into this system. 

Here's what I'm doing these days to earn spare cash --   people in  our neighborhood are starting to pay me to make scuptures of their kids -- tons of fun and so far steady demand.  http://clyattsculpture.com/customportraits.html  This has been my real FIRE dividend-- something I never knew I could do until stumbling into it a few years ago.

Yes. I'm writing a review that I'll post on the REHP web site on Jan 1st. Two thumbs up!

Your sculptures are very impressive. Will you eventually move on to life-size works?

intercst

 

[obryanjf]

Nords,

I'm writing a FIRE calc like spreadsheet with multiple extras. what are you looking for on the salary side? Currently I use a proxy salary (inflation adjusted salary throughout life) to guage success, warnings and failures). I currently have a decay feature built into the variable, fixed and/or salary based draws. I'm interested in modelling different withdrawl strategies, asset allocation strategies, etc. thoughts

 

[Nords]

I'm writing a FIRE calc like spreadsheet with multiple extras. what are you looking for on the salary side? Currently I use a proxy salary (inflation adjusted salary throughout life) to guage success, warnings and failures).   I currently have a decay feature built into the variable, fixed and/or salary based draws.  I'm interested in modelling different withdrawl strategies, asset allocation strategies, etc. thoughts

"Salary"?!? Bleagh.

I guess you'd want the ability to adjust pensions for COLAs, delayed/lagging COLAs, or no COLAs at all.

I'm just wondering if FIRECalc could gain the ability to vary the withdrawals yearly. Six or eight years ago I remember it being a colossal PITA to find a retirement calculator that would let me retire in 2002 (one COLA pension starts up), spouse retire in 2021 (a second COLA pension starts up), and the mortgage payoff in 2032 (more cashflow freed up). It was hard enough to find one that accepted a retirement age younger than 62, let alone a robust one with variable-pension/cashflow features.

Now I'm wondering if we can add code to evaluate the wisdom of spending $25K for a major landscaping renovation sometime in the next 10 years, tripling our healthcare premiums, buying an investment condo, paying for our kid's wedding, renovating the kitchen, or splurging on other life changes. But the whole idea of variable withdrawals is just an added convenience. Even with those planned splurges I think our overall withdrawal rate would stay below 4%.

 

[ESRBob]

Intercst, asked: "Will you eventually move on to life-size works?" (and I can't figure out how to just quote that part...)

The heads are lifesize or bigger than lifesize now, but to do a full body lifesize is still a dream -- it will be quite challenging and fun but beyond my current 'serious hobby' reach -- just getting that much clay moved around would require a truck and several large people. Still light compared to my fellow students chiseling away on hunks of solid stone, though. Our school takes people to Italy every year to carve Carrara marble with hand-made tools just the way Michelangelo used to do... maybe one of these years...

 

[TromboneAl]

Great sculptures -- neat.

The stuffed beaver had something to say about one of the sculptures, but luckily I caught it before he posted it.