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:

Re: Constant $ withdrawal

Post by stone » Tue Jan 09, 2018 4:38 pm

It is really helpful to have those much older time series as well as the 1970-2017 period. However I guess there is a very big difference between a 60%domestic-stock:40%bond portfolio and a 25%emerging market:25domestic:25%bond:25%gold portfolio. Looking at Tyler's calculators for the 1970-2017 period, a mix of emerging market and various developed world stockmarkets hasn't suffered the "lost decade" phenomenon to anything like the same extent as any single regional stock market has. I wonder whether that also holds for the older periods?

I'm with you guys though that now looks like a bad time to increase stock allocations. Perhaps something to think off once stocks crash again.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Tue Jan 09, 2018 4:54 pm

i have seen so few actual crashes in my 30 plus years as an investor . as peter lynch said , more money has been given up in preparation for or waiting for the next down turn then ends up actually being lost in the downturn .

it really does not matter what you use to model at this point . in the past we had no numbers to hang our hat on as far as how much is a safe draw rate so like building a house to withstand the strongest hurricane , the worst retiree outcomes became the benchmarks .

today we don't need to do that anymore . all knowing those past periods did is give us a magic number that we can hang our hat on ..

that number is the 2% real return or higher the first 15 years .

of course that may leave you with a buck at the end of 30 years but the good news is we have yet to hit anything near the worst times in the last 50 years .
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: Constant $ withdrawal

Post by Kriegsspiel » Tue Jan 09, 2018 9:14 pm

1. Save up a fuck load of money. Put at least like 12 years worth of expenses into a PP.
2. Stop worrying about the initial 15 year period?
3-infinity. Drink beer.
User avatar
Xan
Administrator
Administrator
Posts: 4392
Joined: Tue Mar 13, 2012 1:51 pm

Re: Constant $ withdrawal

Post by Xan » Tue Jan 09, 2018 9:45 pm

Kriegsspiel wrote:1. Save up a fuck load of money. Put at least like 12 years worth of expenses into a PP.
2. Stop worrying about the initial 15 year period?
3-infinity. Drink beer.
Wouldn't 12 years' expenses saved up mean you'd need an 8.3% rate of return in order for it to last forever? That's a lot.

The typical goal is 25 years' expenses, which means the traditional 4% would be "safe" (not accounting for Mathjak's order of returns).
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: Constant $ withdrawal

Post by Kriegsspiel » Tue Jan 09, 2018 9:56 pm

Xan wrote:
Kriegsspiel wrote:1. Save up a fuck load of money. Put at least like 12 years worth of expenses into a PP.
2. Stop worrying about the initial 15 year period?
3-infinity. Drink beer.
Wouldn't 12 years' expenses saved up mean you'd need an 8.3% rate of return in order for it to last forever? That's a lot.

The typical goal is 25 years' expenses, which means the traditional 4% would be "safe" (not accounting for Mathjak's order of returns).
If mathjak/Kitces was implying that 4% withdrawal wasn't safe with a certain asset allocation because that allocation has had very bad 15 year returns, I suggested having an allocation that generally has solid returns. I think 12 years worth of PP would be a good defense against a bad 15 year run. The rest of your fuck load could be more PP, stocks, whatever. Just a fuck load of it. At least 25 years.
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Wed Jan 10, 2018 2:24 am

I suppose the crux quandary I was wondering about was whether there is any historical/geographic example where a 25%domesticstocks:25%EM:25%LTT:25%gold portfolio would have run out of money with a constant dollar draw down that a HBPP would have survived? Basically it's whether cash provides more diversification/"growth" than EM stocks when viewed over the sort of time frame where a portfolio would be entirely used up (ie >15years-ish). Cash often is a better diversifier over a year but that is less relevant for this because a portfolio takes several years to run out.

I guess the Japanese bubble of 1990 was followed by an EM crisis in 1997-1998. Maybe Japan gives an example of the HBPP being the safest option?

PS; Tyler's interactive charts seem to have hit a glitch (at least on my browser).
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Wed Jan 10, 2018 2:28 am

there are 2 issues with that premise . there is no guarantee the pp will hold even 2% real return when rates rise on bonds . gold ,stocks and long term treasuries could suck .

the other is the balance left is a consideration too . i don't use the pp because as a long term investor , there really was no logic to mitigating temporary short term dips (which are irrelevant as a long term investor ) and permanently reducing my long term gains .

in a tough retirement outcome the cushion in the balance with the pp may be weak .
User avatar
stone
Executive Member
Executive Member
Posts: 2627
Joined: Wed Apr 20, 2011 7:43 am
Contact:

Re: Constant $ withdrawal

Post by stone » Wed Jan 10, 2018 3:38 am

mathjak107 wrote: i don't use the pp because as a long term investor , there really was no logic to mitigating temporary short term dips (which are irrelevant as a long term investor ) and permanently reducing my long term gains .
I suppose that was sort of the realization I was seeing too. But I was wondering whether almost everyone is in reality a "long term investor". Perhaps some saving schedules do require all of the savings to be drawn down at once. Perhaps saving up for buying something expensive outright at an auction or whatever. But retirement drawn downs don't look like that and a >15year draw down period in itself pushes things over to being in the "long term investor" realm.
mathjak107 wrote: there is no guarantee the pp will hold even 2% real return when rates rise on bonds . gold ,stocks and long term treasuries could suck .......in a tough retirement outcome the cushion in the balance with the pp may be weak .
I hadn't got my head around the possibility of sustained rising interest rates without severe inflation. In the 1970s there were rising interest rates but outpaced by inflation (at least in the UK) so gold did very well indeed. The hike in real interest rates at the start of the 1980s was so short it was survivable. I must admit I find it hard to see why a central bank would hike up interest rates unless battling out of control inflation.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Wed Jan 10, 2018 4:26 am

even spending down directly from 100% stocks as a hypothetical retiree (which i would not do ) , you actually did just fine . without the weight of cash and bonds the difference in gains over time allows spending even in down years with almost the same success rate as 50/50 .

if i remember 50/50 has a 96% success rate at 4% while 100% equity has a 94% success rate .

it really is a mental issue which way you do it not a financial one . the only risk , and it never happened is an extended down turn day 1 .

even 2008 was a non event to a retiree and if you retired in 2008 on 100% equities , today you are no different than any other average retiree group .

a typical portfolio with bonds too will likely never be spending down stocks at a loss in a down turn as rebalancing will always have bonds being sold .in fact if the drop is steep enough conventional rebalancing will have you buying stocks as well as spinning off spending cash .

yeah we can all create visions in our heads of all kinds of nasty exceptions to things and scenario's where things don't work out as they typically do , but it just never plays out that way .

peter lynch was 100% correct when he said more money is lost or given up in anticipation and preparation for a downturn than is actually lost in any downturn . unless of course you exhibit poor investor behavior and hurt yourself ..


so the reality is that since 1871 , at least 35% equities has always maintained a very high success rate of at least lasting 30 years under the worst outcomes so far but the greater the equity allocation the greater the cushion and the balance left .

for the record i am retired and we do live 80% off our portfolio . which i let range from 40 to 50% equities . as of last week i rebalanced my portfolios and are back to the 40% range . i think there is more risk than reward at this point in markets . they had grown from 40 to over 50% the last few years so i scaled them back to have money to put back in when there are more values .

but i was always 100% equities for much of my 30 years as an investor.

that difference in balance from what i had vs would have had if i went much more conservative , can cushion a whole lot of awe craps today compared to where i would have been .

as an example , i have been using the fidelity insight newsletter for 30 years . a 100k in 1987 in the growth model is 2.75 million today . that is what i used for almost all the years since i started in 1987 up until about 5 or 6 years from retiring .

if you were more aggressive and followed their sector model , that actually started a year later and is 3.70 million .

but a growth and income model is about a million less than the growth model . so that mental comfort cost you big time .

in fact there is no evidence that people stick around in more conservative portfolio's . these people tend to be more gun shy and their trigger points are generally lower too so they are just as inclined to bail and run when they see losses mount . morningstar data shows this to be true .
Kbg
Executive Member
Executive Member
Posts: 2815
Joined: Fri May 23, 2014 4:18 pm

Re: Constant $ withdrawal

Post by Kbg » Fri Jan 12, 2018 8:29 am

I’ve said many times here...the vanilla PP is not a good port for a younger person.

I’m not sure so sure I agree with mj’s post retirement spin on things. Historically his market history math may be just fine; however, would a retiree just roll easily with a 50% drawdown of their source of income? My guess is a good portion of them would panic and do something emotionally and financially stupid. Therefore, I think a portfolio that is easier to deal with psychologically is a better way to go.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Fri Jan 12, 2018 8:48 am

mental is a different issue from what makes financial sense .

my wife would not let me ever go higher than 50% equities in retirement . she already lost half her savings in the dot com crash when her first husband died and she went to the adviser at her bank for help . so he loaded her up on tech and dot coms , the moron .

so what we do mentally is a whole other story . i was always 100% equities up until just about retirement . but 40-50% suits us just fine .

but even 40-50% can be a white knuckle ride as we get more and more dollars in our account .

a mere 7% drop today in portfolio value would represent 9 years of maxing out my 401k at catchup .

so when the crap hits the fan if your dollars are high enough there is no such thing as a "good nights sleep " in a down draft regardless of allocation unless relatively few dollars are effected the body reacts .. "everyone has a plan-until they get punched in the face " is so true . thank you tyson .

we humans are prewired to hate loosing money more than making money and it makes us exhibit poor investor behavior all the time regardless of amounts . there is no evidence balanced funds exhibit any better behavior in rough times than growth funds do ...

you would think they would be easier to stay the course , but nope , down is down and investors exhibit poor behavior regardless .
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: Constant $ withdrawal

Post by Kriegsspiel » Fri Jan 12, 2018 9:29 am

mathjak107 wrote:there are 2 issues with that premise . there is no guarantee the pp will hold even 2% real return when rates rise on bonds . gold ,stocks and long term treasuries could suck .
Granted. Although I'd wager that every investment would suck in that situation. The best you could expect would be that the PP would not suck as much.
the other is the balance left is a consideration too . i don't use the pp because as a long term investor , there really was no logic to mitigating temporary short term dips (which are irrelevant as a long term investor ) and permanently reducing my long term gains .

in a tough retirement outcome the cushion in the balance with the pp may be weak .
I see your point if you're working, spending some of your income and saving some (and not touching it for 20 years). If you're retired, you aren't a long term investor. You are spending your savings. According to your own posts, you do want to mitigate first-15-year dips because they ruin retirements.
mathjak107 wrote:even spending down directly from 100% stocks as a hypothetical retiree (which i would not do ) , you actually did just fine . without the weight of cash and bonds the difference in gains over time allows spending even in down years with almost the same success rate as 50/50 .
In 1907, 1929, 1937, 1966?

I legit can't tell what you're saying... Just so we're clear, you're saying that the retirements that failed due to 15 year bad returns for the balanced portfolios would have succeeded if they were in 100% stocks? No bonds, no cash at all?

The stocks had worse 15 year returns than bonds in a few of those cases, from what you posted, so you can see why I'm asking.
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Fri Jan 12, 2018 9:36 am

no ,those dates are the worst historical dates for retirees in any allocation and they are the dates the 4% safe withdrawal rate was based on . all of those dates failed and you needed to reduce to about 3.70% with any allocation over 35% equities to survive the stress test 100% .

but those dates failing still represent over a 90% success rate at 4% so it is still acceptable .

but the common denominator to all failures at 4% are conditions were so bad the first 15 years that even the best of years coming along later could not salvage them .

so mathematically it takes about a 2% real return the first 15 years to have 4% hold . of course what balance is left at 30 years depends on your actual allocation , but the income stream should hold up ok

the funny thing is researcher michael kitces found that if you made it to the 15 year mark okay , that even if your retirement went out longer than 30 years you still were okay .

so you can see in the 30 year results for all those failure years 30 years was not bad . but the 15 year periods sucked .

we don't know how the pp would have held up over those benchmark years since gold's prices were really skewed but it is not important . thanks to kitces we know no matter what you use if you are not seeing a 2% real return the first 15 years you are in danger of not holding 4% so a pay cut may be in order. in fact if you are not holding a 2% real return 5 years in , i certainly would proactively cut spending as a precaution
User avatar
Kriegsspiel
Executive Member
Executive Member
Posts: 4052
Joined: Sun Sep 16, 2012 5:28 pm

Re: Constant $ withdrawal

Post by Kriegsspiel » Fri Jan 12, 2018 10:34 am

Just read an article by Kitces, he's saying the same thing I am.
One popular way to manage the concern of sequence risk is through so-called “bucket strategies” that break parts of the portfolio into pools of money to handle specific goals or time horizons. For instance, a pool of cash might cover spending for the next 3 years, an account full of bonds could handle the next 5-7 years, and equities would only be needed for spending more than a decade away, “ensuring” that no withdrawals will need to occur from the portfolio if there is an early market decline.

Yet the reality is that strict implementation of a bucket strategy is more than just an exercise in mental accounting; it can actually distort the portfolio’s asset allocation, leading to an increasing amount of equity exposure over time as fixed income assets are spent down while equities continue to grow. Yet recent research shows that despite the contrary nature of the strategy – allowing equity exposure to increase during retirement when conventional wisdom suggests it should decline as clients age – it turns out that a “rising equity glidepath” actually does improve retirement outcomes!

... ironically, it turns out that for those who do want to implement a rising equity glidepath, the best approach might actually be to explain it to clients as a bucket strategy in the first place!

https://www.kitces.com/blog/should-equi ... ly-better/
User avatar
mathjak107
Executive Member
Executive Member
Posts: 4456
Joined: Fri Jun 19, 2015 2:54 am
Location: bayside queens ny
Contact:

Re: Constant $ withdrawal

Post by mathjak107 » Fri Jan 12, 2018 1:54 pm

buckets are nothing more than mental masturbation . in fact not only is there no help from them but they can actually hurt you .

our brains like to compartmentalize . they are comfortable like that . but using a fixed allocation and simply rebalansing seems to get better results .

as kitces said:


EXECUTIVE SUMMARY

Financial planners have always sought to adjust their strategies and communication techniques to the realities of client needs, although the increasing volume of behavioral finance research is now beginning to document exactly how we as human beings sometimes think in very irrational ways, which in turn provides insight about how to best adapt to deliver effective advice. One common challenge area regarding investments in particular is our tendency for mental accounting – where we break up and categorize assets based on various needs and purposes, even if the underlying investments are flexible or entirely fungible – which in turn has spurred the growth of so-called “bucket strategies” that seek to allocate portfolios based on various goals, needs, or time horizons.

Unfortunately, though, recent research has shown that stringent applications of bucket strategies can potentially result in less optimal retirement outcomes, not better ones, particularly due to the “cash drag” and portfolios that can dial down too conservatively too fast; in addition, the reality is that mathematically, most of the benefits of bucket strategies are captured simply from traditional rebalancing strategies, which already ensure that stocks are bought (not sold) when they’re down and that cash and bonds are used for spending needs when appropriate.

Nonetheless, from the behavioral perspective, using bucket strategies remains appealing, if only to help clients stay the course during stressful times. But ultimately, perhaps the best solution is not just to weigh the trade-off between managing with buckets (even if the results are worse) versus helping clients psychologically (which is still better than having them bail out at the worst of times), but to accomplish both by improving performance reporting to overlay buckets and goals on top of the portfolio. In other words, maybe the key is not that we need to change how we invest for clients, but simply to more effectively frame how we report the results?

https://www.kitces.com/blog/Should-Fina ... -That-Way/
Post Reply