Constant $ withdrawal

A place to talk about speculative investing ideas for the optional Variable Portfolio

Moderator: Global Moderator

User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Constant $ withdrawal

Post by stone » Sun Dec 31, 2017 12:57 pm

I've discovered Tyler's (from here) fabulous website and have been playing around with his interactive charts. One thing that struck me was that the style of withdrawal seems to have major implications for what's safest as a portfolio composition. If a constant % withdrawal is used (ie 4% of whatever the portfolio happens to be at any given time) then a very low volatility portfolio is better because a volatile portfolio will get eaten up whenever the value spikes up. But if a constant $ withdrawal is used, then a much more volatile portfolio with a higher CAGR gives a much safer result. I used the chart at https://portfoliocharts.com/portfolio/r ... -spending/ and compared two UK portfolios. The low volatility one had 15%TDM,10%NorthAmerica,15%emergingMkt, 20%LTT,20%Tbills, 20%gold; the high volatility portfolio had 25%TDM,25%emergingMkt,25%LTT, 25%gold. The low volatility portfolio looked like it would keep going for perpetuity with a 4.7% withdrawal rate at constant$ or at 6% at constant%. The high volatility portfolio looked like it would keep going for perpetuity with a 6% withdrawal at constant$ or at 6.3% at constant%. The "ulcer index" was better for the low volatility portfolio so it might be thought of as "safer" but seems to me far from being the safest option since constant$ is probably closer to a real life situation. Am I in a muddle?
PS I'm not saying 6% constant$ withdrawal is recommendable, simply that having a portfolio like that would give a bigger margin of safety with a conservative constant$ withdrawal.
farjean2
Executive Member
Executive Member
Posts: 284
Joined: Thu Feb 23, 2017 12:51 am

Re: Constant $ withdrawal

Post by farjean2 » Sun Dec 31, 2017 1:59 pm

Yes, you are in a muddle and I suggest that since it's New Years Eve you should just chill out, have a beer or two or three or more and let tomorrow worry about itself.

I retired last year and it's really not as complicated as you are making it. We have accumulated X dollars that has to last for Y number of years with a realistic Z percent ROI and it looks good on paper but we don't know what tomorrow brings so we don't spend a lot of time and energy worrying about it. We live conservatively staying within the parameters of our plan but not denying ourselves anything within those parameters due to worries about things in the future we can't control. We will deal with those things if/when they come.
User avatar
sophie
Executive Member
Executive Member
Posts: 3488
Joined: Mon Apr 23, 2012 7:15 pm

Re: Constant $ withdrawal

Post by sophie » Sun Dec 31, 2017 3:30 pm

Stone, nice to hear from you!! Do you remember this thread? Maybe it should be stickied.

viewtopic.php?f=9&t=4887&hilit=safe+withdrawal&start=12

I didn't realize Tyler had turned his charts from that thread to a page on his portfoliocharts site. Tyler, what assumptions went into your retirement calculator? I saw something about withdrawals pegged to 4% of portfolio value if the portfolio increases.

In any case, I don't think I'd be trying to push withdrawal rates that hard. The future may or may not produce the boom years of the 1980s and early/mid 90s again, which have a lot to do with the lustrous return from stocks.
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Sun Dec 31, 2017 4:34 pm

Hi Sophie, Tyler's site is a wonderland of geeky fun :) . My aim wasn't to push withdrawal rates hard but to see what sort of portfolio would give the greatest margin of safety such that a modest withdrawal rate would still be OK even if things ended up generally giving lower returns. I just thought it was interesting that in principle a portfolio that is designed to give a very low "ulcer index" and so a very good maximum safe constant% withdrawal rate falls short on the safe constant$ withdrawal rate.
PS, That "high volatility" portfolio I described actually has the high returns scattered all over the time series. The 2000s were a boom for it.
User avatar
Tyler
Executive Member
Executive Member
Posts: 1876
Joined: Sat Nov 12, 2011 3:23 pm
Contact:

Re: Constant $ withdrawal

Post by Tyler » Sun Dec 31, 2017 5:40 pm

stone wrote:One thing that struck me was that the style of withdrawal seems to have major implications for what's safest as a portfolio composition.
I'm glad you find the site useful!

It doesn't surprise me that certain withdrawal methods may pair well with certain portfolios. I honestly haven't explored it enough to have an opinion on why that is, but I do think it deserves further study.

Oh, and I appreciate that you're taking a balanced approach to this that doesn't merely maximize withdrawal rates. I prefer to use these tools to validate the retirement performance of a portfolio you already like for other reasons, as withdrawal rates are meaningless if they're calculated for a portfolio that you're unable to stick with for the long haul.
sophie wrote: I didn't realize Tyler had turned his charts from that thread to a page on his portfoliocharts site. Tyler, what assumptions went into your retirement calculator? I saw something about withdrawals pegged to 4% of portfolio value if the portfolio increases.
Yep -- this forum has provided all sorts of inspiration for my own calculations over the years, and I'm happy to have found a way to share. :)

For reference, the Withdrawal Rate assumptions are here and the Retirement Spending assumptions are here.
Mechanical engineer, history buff, treasure manager... totally not Ben Gates
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Mon Jan 01, 2018 2:30 am

Tyler, your website is a wonder to behold. It is set out so intuitively that even a dunce like me can just start messing around and everything becomes clear :).

My impression about safe withdrawal methods and portfolio composition was that the safest withdrawal method would be a composite between the constant% and constant$. If the inflation adjusted portfolio value fell below the starting point, then constant% would be withdrawn until the portfolio regained its value, at which point constant$ would be withdrawn. That probably would be what most cautious people would just do instinctively anyway. Such a withdrawal method would be even more tilted in favour of a portfolio that had consistently high 10year CAGR rolling returns but worse max-draw-down statistics.

Portfolios such as the HBPP minimize the "ulcer index" but perhaps that mainly has purely emotional benefit for people who check on their portfolio values more than the once a year that Harry Browne recommended. In terms of genuine financial security, it is much better to have a portfolio that has say grown twice as large as you need but has occasional, brief, 20% draw-downs. What really matters is the extent of draw-downs to below the starting value of the portfolio during the withdrawal period.
Last edited by stone on Mon Jan 01, 2018 11:04 am, edited 1 time in total.
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Mon Jan 01, 2018 3:15 am

Image
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Mon Jan 01, 2018 4:25 am

6% constant$ withdrawal from "constant$-safe-portfolio" (ie CDSP :) )
Image

6%constant$ withdrawal from HBPP
Image

6%constant$ withdrawal from GoldenButterfly
Image

CDSP benchmarked against GoldenButterfly
Image

CDSP Ulcer index:
Image

HBPP Ulcer Index
Image
barrett
Executive Member
Executive Member
Posts: 1524
Joined: Sat Jan 04, 2014 2:54 pm

Re: Constant $ withdrawal

Post by barrett » Sun Jan 07, 2018 7:34 am

stone wrote:Portfolios such as the HBPP minimize the "ulcer index" but perhaps that mainly has purely emotional benefit for people who check on their portfolio values more than the once a year that Harry Browne recommended. In terms of genuine financial security, it is much better to have a portfolio that has say grown twice as large as you need but has occasional, brief, 20% draw-downs. What really matters is the extent of draw-downs to below the starting value of the portfolio during the withdrawal period.
stone, correct me if I am wrong but aren't you finding that variations on Tyler's Golden Butterfly portfolio have generally done better than the HBPP during the testable timeframe? It looks like the AAs that you are plugging in are all 40% - 50% stocks, right?

As Sophie alluded to, it was really the stock bull market of 1982 to 1999 that separated a high stock allocation from the HPBB. Or are the numbers telling you something different?

The GB is appealing because by holding a higher percentage of stocks, it diminishes the feeling of being "left behind" when stocks are roaring. And it fits nicely into Machine Ghost's synopsis (probably slightly misquoted here) that "stocks drive PP returns and everything else is just hedges."

That 40% stock allocation is so tempting... if only the Shiller CAPE weren't up over 33. Hmm.
User avatar
sophie
Executive Member
Executive Member
Posts: 3488
Joined: Mon Apr 23, 2012 7:15 pm

Re: Constant $ withdrawal

Post by sophie » Sun Jan 07, 2018 8:11 am

Barrett - that's why I was thinking that instead of using a stock-heavy PP variant, you could instead think of the PP as a safe, conservative core, and put any additional savings into 100% stocks with the expectation that it might take off like it has the past few years, or go nowhere for 10 years like in the 1999-2009 period. Because you have the PP for core expenses, this wouldn't be as much of a concern.

The idea that it's ok to have a portfolio that drops 20% (or 40%) but outpaces the conservative approach is fine, as long as you have a plan for dealing with the possibility of prolonged negative returns or a deep drawdown. Like, hold 5 years expenses in cash or increase target portfolio size by 20-30%. If you hold the cash though, you'd have to factor that into return calculations, in order to get a fair comparison to the PP. I suspect the lustrous-looking CAGR of, say, a 60/40 portfolio would drop quite a bit if you did.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 2722
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Tue Jan 09, 2018 5:50 am

to sustain a 4% safe withdrawal rate it has taken a 2% average real return over the first 15 years of a retirement time frame to last .
if you don't have at least a 2% real return average the first 15 years no matter how good the next years are the risk of failing is very very high .

all 30 year time frames for all the worst 30 year periods going as far back as 1871 failed in the first 15 years because the real return average was to low to support 4%.

1965 /1966 the poster children for the worst group on record also included the greatest bull market in history in their time frame . but it was to little to late .

so if 5 years in you see you are not averaging at least a 2% real return , a red flag should go up to the fact you may be taking a pay cut if this keeps up
User avatar
mathjak107
Executive Member
Executive Member
Posts: 2722
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Tue Jan 09, 2018 8:34 am

all the stress testing of all these portfolio allocations do is identify what failed . it is the high speed numbers crunching we have today that was able to analyze the common denominator to all the worst case outcomes for a retiree .

1907,1929,1937.1965/1966 were all the poster children for the worst outcomes and when the time frames were analyzed (thanks to michael kitces )

the same common denominators all came up .

the 30 year results were all pretty average . you would really not know anything bad happened . but all failed in the first 15 years when the real return average sucked .


30 year results looked like :

1907 stocks returned 7.77% -- bonds 4.250-- rebalanced portfolio 7.02- - inflation 1.64--

1929 stocks 8.19% - - bonds 1.74%-- rebalanced portfolio 6.28-- inflation 1.69--

1937 stocks 10.12 - - bonds 2.13 - rebalanced portfolio -- 7.24 inflation-- 2.82

1966 stocks 10.23 - -bonds 7.85 -- rebalanced portfolio 9.56- - inflation 5.38

for comparison the 140 year average's were:

stocks 8.39--bonds 2.85%--rebalanced portfolio 6.17% inflation 2.23%

now look at the 15 year periods

1907--- stocks minus 1.47%---- bonds minus .39%-- rebalanced minus .70% ---inflation 1.64%

1929---stocks 1.07%---bonds 1.79%---rebalanced 2.29%--inflation 1.69%

1937---stocks -- 3.45%---bonds minus 3.07%-- rebalanced 1.23%--inflation 2.82%

1966-stocks minus .13%--bonds 1.08%--rebalanced .64%-- inflation 5.38%
Post Reply