Author Topic: Portfolio Charts - The Golden Butterfly  (Read 275095 times)

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #350 on: October 15, 2016, 02:13:24 AM »
I think if that I were worried about it, and if it were at all possible, I would do two sets of analysis - one with those years included and one without. I would then base any conclusions on the worst case out of the two.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #351 on: October 15, 2016, 02:28:01 AM »
It's not possible to know the future.  Just make our best guess.

Our best guess doesn't have to come solely from back testing.

Business ownership (via equities, in this case) is fundamentally different than gold ownership.  While I can come up with circumstances where 100% gold beats 100% stocks, they're much less likely than vice-versa.

Where you choose your mix for your AA relies on a number of things (including the ever important psychology); backtesting is merely one of these things.  It shouldn't be the only thing, or you get an overfitted portfolio that's almost guaranteed to UNDERperform.

Thanks, I agree with all of this.

I'd say that my own current tendency towards a GB-like allocation isn't solely based on back-testing. It is more based on my belief that diversification is a good thing and will remain a good thing.

the right diversification is a good thing . the wrong diversification can hurt you permanently over the long term . being a long term investor that is what hedging does .  it gives you temporary relief from short term events that may mean nothing in the long term , but cuts your compounding over the long term .

diversification in to all sectors like financials ,healthcare ,consumer staples ,etc is a good thing .

just buying bonds or gold can be a bad thing performance wise over the long term  .it always has . we are starting to see bond rates rise after 35 years . that can be a heavy weight with long term bonds . bonds got hit hard again yesterday . it has been a blood bath for the permanent portfolio the last 3 months which is similar to the gb .

i mean a 45% drop is a lot since it peaked at brexit. that is just the thing it is not supposed to do .
« Last Edit: October 15, 2016, 02:32:07 AM by mathjak107 »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #352 on: October 15, 2016, 03:44:39 AM »
it has been a blood bath for the permanent portfolio the last 3 months which is similar to the gb .

i mean a 45% drop is a lot since it peaked at brexit. that is just the thing it is not supposed to do .

A quick check of some charts and I can't see how a 45% drop was possible for PP in the last 3 months. Looks like more like 6% or 7% might have happened.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #353 on: October 15, 2016, 03:59:27 AM »
well the portfolio i still track was up 135k  90 days ago . it is now up only 74k

that value has fallen by 45%. going from being up 15 or 16% to only up 7% or so , that  drop represents a 45% fall  in dollars  . is that low volatility to you ? 
« Last Edit: October 15, 2016, 04:06:39 AM by mathjak107 »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #354 on: October 15, 2016, 04:05:42 AM »
Those values look like profit, rather then total portfolio value.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #355 on: October 15, 2016, 04:07:14 AM »
yes , but a 45% drop in your gains is a lot for that model . we are talking almost  a 70k fall . it got hit pretty hard .on the other hand my conventional model is still ahead of the pp by 25k  using same starting amount .

you would think with all the volatility this year the pp would have been the better choice  but the assets playing against each other are doing more harm than good and that is my point  . trying to hedge your bets is working against the portfolio and not for it .

we seem to be duplicating what we see in times of tight money even though that is not the case. that is not good for models like the  pp or gb .

unlike equity's , interest rate cycles and gold cycles can be out of favor for very long periods of time . it took interest rates 35 years to get to these levels, who knows how long it will take going up . . gold has never had an extended long term time frame where something else did not do better .

if you bought your gold the first year gold traded here in 1975 for 175 an ounce today a 1 month t-bill rolled over did better .

my opinion is i think going forward these long term treasury /gold heavy portfolio's are not going to be the low volatility models they were .

i still think a diversified mix of equity's and a dynamic bond portfolio that is ready to shift to different kinds of bond funds as the big picture unfolds is the way to go .

buy and die was a lot easier when rates were falling for decades . taking a 17% hit in TLT  with just a 1 point rise in rates is going to be quite painful to those models that use long term bonds if we have the meager stock returns that are anticipated  .


« Last Edit: October 15, 2016, 04:27:59 AM by mathjak107 »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #356 on: October 15, 2016, 04:26:05 AM »
trying to hedge your bets is working against the portfolio and not for it .

I think this is to be expected at times for any portfolio with more than one asset class in it.

Quote
if you bought your gold the first year gold traded here in 1975 for 175 an ounce today a 1 month t-bill rolled over did better

In terms of total return, but that's not the most important thing for a lot of people. It is for some though.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #357 on: October 15, 2016, 04:33:07 AM »
Quote
taking a 17% hit in TLT  with just a 1 point rise in rates is going to be quite painful to those models that use long term bonds

I think re-balancing will help to soften the blow.

I wonder if more frequent re-balancing in more uncertain times is a way to help lower the perceived risks without upsetting the longer term success or asset allocations of the portfolio? It could probably work both ways. It's also been discussed in another thread no doubt.
« Last Edit: October 15, 2016, 04:35:55 AM by MrNotRobot »

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #358 on: October 15, 2016, 04:59:09 AM »
rebalancing under these circumstances just ends up sending good money after bad . more often than not rebalancing cuts the gains and adds to the losers . 70% of the time equity's go up . that has you betting against the house .

you are also taking money out of equity's which have always been the long term winner and putting them in assets that over the long term have never out performed equity's .

for rebalancing to really work you need to be a good market timer .

kitces looked at the effects of rebalancing and found long term it actually hurt more than helped .

but in this case i think it  is the fact that we have never had a situation with rates this low and stocks this high . that means we can have slow growth in equity's and poor conditions for bonds and gold .

i still say taking the most likely horse in the race , your equity's and strapping the weight of long term bonds and gold to them will not allow them to gain any traction . the wild card would be betting on long term bonds going forward to do the heavy lifting .

you would have to bet on a long term disaster to the markets . i think if not then all these portfolio's that attempt to hedge will have more volatility than those using them would want . the last 3 months may be an indication of the new normal for the pp and gb .
« Last Edit: October 15, 2016, 05:06:30 AM by mathjak107 »

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #359 on: October 15, 2016, 05:19:48 AM »
I'd say that my own current tendency towards a GB-like allocation isn't solely based on back-testing. It is more based on my belief that diversification is a good thing and will remain a good thing.

This is to me a principle that you can follow. A well diversified asset allocation gives you a good chance of having lower but more stable returns. I think the principle is important rather than any back-testing.

Still you have to look at the down side of this type of allocation. Your greatest chance of success over the long term is probably with a 100% stock allocation.
« Last Edit: October 15, 2016, 04:59:36 PM by steveo »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #360 on: October 15, 2016, 06:36:40 AM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #361 on: October 15, 2016, 07:02:37 AM »
the problem is that conditions today have turned non correlated assets into correlated ones and not only are they correlated but the volatility in the opposite asset classes you typically need to add ballast  which have plenty of ooomph , now work against you making the supposedly  non volatile portfolio even more volatile .

you take something like the pp with long term bonds and gold  and those volatile investments can add far more swings and volatility when  teaming up  than a conventional stock fund  with  short to intermediate term bond funds will . that is why the pp is now worth less than my conventional model  after it was much ahead .
« Last Edit: October 15, 2016, 07:07:29 AM by mathjak107 »

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 27688
  • Age: -999
  • Location: Traveling the World
Re: Portfolio Charts - The Golden Butterfly
« Reply #362 on: October 15, 2016, 07:17:09 AM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

In the short run, volatility is a better predictor of success.

In the long run, returns are a better predictor of success.

A portfolio with higher returns is likelier to last longer, even with sequence of returns risks.  This is doubly true if you can be flexible about spending (because being flexible about spending early, temporarily, during a crash is much easier than permanently reducing spending forever due to eroded purchasing power due to lower returns).
We are two former teachers who accumulated a bunch of real estate, retired at 29, and now travel the world full time with two kids.
If you want to know more about me, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #363 on: October 15, 2016, 07:30:40 AM »
i have mixed feelings about volatility being a measure of success unless you have a low pucker factor  but if that is the case i wouldn't call being gun shy a success .. if it is long term money then volatility should not be an issue in the short term  . if it is shorter term money than perhaps you are mis-matching time frames  and investments and hoping the portfolio does not fall to much if you need the money in the shorter term .  in that case hope is the strategy .

i think we have to forget about what was for the most part as driving and looking in the rear view mirror on a road we have never been before does little good .

we have never had all asset classes over valued at the same time before .even when rates were low in the 1930's in real return they were very high since the cpi fell 18%
« Last Edit: October 15, 2016, 07:35:20 AM by mathjak107 »

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 27688
  • Age: -999
  • Location: Traveling the World
Re: Portfolio Charts - The Golden Butterfly
« Reply #364 on: October 15, 2016, 07:35:37 AM »
I said predictor, not measure.
We are two former teachers who accumulated a bunch of real estate, retired at 29, and now travel the world full time with two kids.
If you want to know more about me, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #365 on: October 15, 2016, 07:36:37 AM »
i am not even sure i would say that about the short term .

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 27688
  • Age: -999
  • Location: Traveling the World
Re: Portfolio Charts - The Golden Butterfly
« Reply #366 on: October 15, 2016, 07:56:31 AM »
i am not even sure i would say that about the short term .

I certainly would.

In the short term, return won't do much for you.  It's decades of compounding that have the big impact.

Volatility can do a lot against you though.

That's why different investments are recommended for shorter term time horizons, more conservative investments.
We are two former teachers who accumulated a bunch of real estate, retired at 29, and now travel the world full time with two kids.
If you want to know more about me, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #367 on: October 15, 2016, 07:59:20 AM »
but as i said the models like the pp and gb which are heavy in to long term bonds and gold have been proving to be more volatile in the short term today. just look at the last 3 months since there was just a scent of inflation and rate increases in the air .

the strong pulls from those assets that are supposed to be in  the opposite direction are teaming up and pulling things down instead of up .

that is my point ,under different circumstances what was less volatile can end up more volatile and we are seeing that now  in any portfolio's that use heavy positions in long term bonds and gold . . 6 or 7% in gains evaporated in 90 days time  from the pp. if you just started in the pp , how would you feel about its short term volatility being down 7%  in 3 months time  ? my conventional model is down less than 2% the last 3 months from where it was .
« Last Edit: October 15, 2016, 08:30:16 AM by mathjak107 »

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1114
  • Age: 43
Re: Portfolio Charts - The Golden Butterfly
« Reply #368 on: October 15, 2016, 11:10:06 AM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

In the short run, volatility is a better predictor of success.

In the long run, returns are a better predictor of success.

A portfolio with higher returns is likelier to last longer, even with sequence of returns risks.  This is doubly true if you can be flexible about spending (because being flexible about spending early, temporarily, during a crash is much easier than permanently reducing spending forever due to eroded purchasing power due to lower returns).

The idea "well, if returns are bad for a few years, I'll get some part time work or cut back spending", is a repeated sentiment on these forums.  I would agree that it is generally predictable when a portfolio is in trouble and these are workable solutions.  However, it's a solution outside of the portfolio and not necessarily the only one available. 

Sacrificing returns in the first 10-15 years of drawdown to minimize volatility is also a viable solution.  One that does not require working or spending reduction (or softens the need for these things).  Why not choose the best AA for whatever your goals are in a specific period of life?  If someone's "goal" in the beginning of drawdown is to minimize sequence risk, why choose a portfolio with historical and theoretical high volatility?  It seems counter productive.  I can see why someone would argue GB's historical performance may not repeat, but the theory is sound (in my mind).  Then, once sequence risk is overcome, one could move back towards a equity heavier portfolio to combat the next issues, long term compounding inflation and maybe legacy.

I also think Mathjack107 is correct in one account, that it is wise to watch macroeconomic conditions and make adjustments.  Not that anyone can precisely predict the future, rather get a general idea of long term trends.  Put another way, when sitting at the beach I can't predict when the next big wave will hit, but I can watch the tide coming in and slightly adjust my lawn chair to avoid the oncoming water.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #369 on: October 15, 2016, 12:08:57 PM »
i am retired and have toned my portfolio down in volatility  considerably . but i do not use long term treasury's at this stage nor gold as a fixed position . i do own gld  as a trading vehicle for quick profits .

at this stage i would keep neither of those as a permanent part of the portfolio  if i wanted to tame volatility for the reasons i mentioned . going forward they may make the portfolio more volatile and not less volatile if the last 3 months are an indication .

not only will you hinder long term gains permanently but you will do more damage volatility wise in the short term then you hoped for . that could be a double whammy going forward .


« Last Edit: October 15, 2016, 12:26:39 PM by mathjak107 »

Classical_Liberal

  • Handlebar Stache
  • *****
  • Posts: 1114
  • Age: 43
Re: Portfolio Charts - The Golden Butterfly
« Reply #370 on: October 15, 2016, 12:30:56 PM »
@ mathjack107  I understand your position on these investments, particularly LTT's as they have little upside and large downside at this moment in time. IMO gold has more upside than LTT. Be cautious on short term predictions. As Europe & Japan has shown, zero or negative yields on treasuries are not out of the question, even if the Fed makes a small rate hike.  Three months is a wave, not the tide.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #371 on: October 15, 2016, 12:50:58 PM »
i agree , there is a remote chance rates can go lower . but if someone is that concerned about their portfolio volatility in the short term it would not make a lot of sense making a heavy bet on the flyer .

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #372 on: October 15, 2016, 05:04:08 PM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

I hear what you are stating but I think those statements are pretty close to the same thing if we are talking a long period of retirement. If you have a long period of retirement which maybe greater than 30 years your greatest chance of success is probably going to align with the greatest chance of maximum returns. Low returns over a long period will get you through shorter time periods but not longer time periods.

Personally I'm a little volatility averse because I want to ensure my portfolio holds up under the short term as well. I have some bonds in my portfolio and I intend to keep them there. I also intend to draw down on bonds at the start of my retirement assuming stocks aren't going gang busters. That is my safety net. At the same time I think that in most situations high equity allocations will give you the greatest chance of success over the longer term.

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #373 on: October 15, 2016, 05:05:23 PM »
the problem is that conditions today have turned non correlated assets into correlated ones and not only are they correlated but the volatility in the opposite asset classes you typically need to add ballast  which have plenty of ooomph , now work against you making the supposedly  non volatile portfolio even more volatile .

you take something like the pp with long term bonds and gold  and those volatile investments can add far more swings and volatility when  teaming up  than a conventional stock fund  with  short to intermediate term bond funds will . that is why the pp is now worth less than my conventional model  after it was much ahead .

The problem with this statement is that who knows what the future will bring. It's all guess work. You might be right for the next 2 years and then wrong for the next 30 years.

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #374 on: October 15, 2016, 05:07:04 PM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

In the short run, volatility is a better predictor of success.

In the long run, returns are a better predictor of success.

A portfolio with higher returns is likelier to last longer, even with sequence of returns risks.  This is doubly true if you can be flexible about spending (because being flexible about spending early, temporarily, during a crash is much easier than permanently reducing spending forever due to eroded purchasing power due to lower returns).

Exactly. You make another critical point. If you can be flexible that also gives you added security.

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #375 on: October 15, 2016, 05:10:54 PM »
Sacrificing returns in the first 10-15 years of drawdown to minimize volatility is also a viable solution.  One that does not require working or spending reduction (or softens the need for these things).  Why not choose the best AA for whatever your goals are in a specific period of life?  If someone's "goal" in the beginning of drawdown is to minimize sequence risk, why choose a portfolio with historical and theoretical high volatility?  It seems counter productive.

I agree with this premise but I also look at the counter point that arebelspy and I made. I think looking at your downside to a certain degree is smart but I personally wouldn't take it as far as implementing a GB type portfolio. That seems way to extreme if all you are worried about is the first 5 or so years of sequence of return risk.

lordmetroid

  • Stubble
  • **
  • Posts: 112
Re: Portfolio Charts - The Golden Butterfly
« Reply #376 on: October 15, 2016, 05:28:55 PM »
Would anybody care to speculate how GB would have done for a Japanese investor i.e Replace TSM for Nikkei?, SCV and treasuries for Japanese equivalents?

I live in Sweden and was interested in the Permanent Portfolio so I back-traded that portfolio to see how it worked in Sweden and also in Japan as a case study. Turns out the results are greatly different. USA is a unique market because your currency, the Dollar.

For instance, gold behaves totally different. First of all gold is traded in USD so you have to add your own currency-volatility to the asset and second when their is stock market panic outside of US. It has no effect on gold whatsoever.

Getting your hands on government bonds and bills can also be rather tricky. In Sweden for instance only wholesalers can buy them from the state and they do not sell retail. The only bonds you can get is through mutual funds that have a mix of different kinds of bonds/bills. State, municipal, corporate and mortgages.

So the permanent portfolio isn't so permanent outside of USA.

lordmetroid

  • Stubble
  • **
  • Posts: 112
Re: Portfolio Charts - The Golden Butterfly
« Reply #377 on: October 15, 2016, 05:29:33 PM »
why buy Japanese stocks just because you may live there ?    i would never buy in a country i wasn't happy with market or economic wise .

To avoid the currency-risk.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #378 on: October 16, 2016, 04:07:18 AM »
the problem is that conditions today have turned non correlated assets into correlated ones and not only are they correlated but the volatility in the opposite asset classes you typically need to add ballast  which have plenty of ooomph , now work against you making the supposedly  non volatile portfolio even more volatile .

you take something like the pp with long term bonds and gold  and those volatile investments can add far more swings and volatility when  teaming up  than a conventional stock fund  with  short to intermediate term bond funds will . that is why the pp is now worth less than my conventional model  after it was much ahead .

The problem with this statement is that who knows what the future will bring. It's all guess work. You might be right for the next 2 years and then wrong for the next 30 years.

actually you have to be pretty right in the beginning. the first 5 years have a big effect and the first 15 determine the entire outcome . after a run up you can pretty much get by with doing anything . but it is the beginning years that need a whole lot of being right more

you just have to much heavy exposure with pp and gb if assets move together as they have been as rates rise  in this new normal . but that is my opinion . i won't do either one today . to much short term  volatility  risk now and to much gain cutting later . i think you would  be   buying the worst outcome's most likely in both situations .
« Last Edit: October 16, 2016, 04:12:40 AM by mathjak107 »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #379 on: October 16, 2016, 06:01:20 AM »
Everything does seem expensive right now.

I think what I finally end up going with will only be GB-like rather than the pure original. It will be customized to have a larger share component of at least 50% and smaller in both bonds and gold. I posted what it might look like a few days ago.


mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #380 on: October 16, 2016, 06:45:53 AM »
i can see doing that .

the usual long term driver is prosperity to put it in harry's terms . so with that potential somewhat dim you don't want to much grabbing your equity's by the collar and yanking them backwards . cutting some of those  volatile other positions that may not move opposite much going forward may be a good thing .

at this point my bond position is very flexible . it has changed much over the last 2 years . it has reduced it's position in total bond funds and corporate bond funds to go in to other more productive segments of the bond market .

high yield bonds was a great proxy for stocks too . my high yield fund has had double digit returns with only half the volatility of the s&p 500. as the party ends there that bond money will find other types of area's in the bond universe .

what i wouldn't do anymore is buy and die with long bonds or just a total bond fund which really has nothing total about it . total bond funds are about 1/2 treasury and gov't bonds as well as are missing so many segments of the  bond market .
« Last Edit: October 16, 2016, 07:02:45 AM by mathjak107 »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #381 on: October 16, 2016, 10:21:23 AM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

I hear what you are stating but I think those statements are pretty close to the same thing if we are talking a long period of retirement. If you have a long period of retirement which maybe greater than 30 years your greatest chance of success is probably going to align with the greatest chance of maximum returns. Low returns over a long period will get you through shorter time periods but not longer time periods.

Personally I'm a little volatility averse because I want to ensure my portfolio holds up under the short term as well. I have some bonds in my portfolio and I intend to keep them there. I also intend to draw down on bonds at the start of my retirement assuming stocks aren't going gang busters. That is my safety net. At the same time I think that in most situations high equity allocations will give you the greatest chance of success over the longer term.

I think success being subjective is the issue here. If the portfolio doesn't meet my objectives then I wouldn't be inclined to call it a success. One of my objectives is to not only minimize the size of draw-downs but also minimize the length of those draw-downs. I'm even more afraid of longer draw-downs than larger ones because it would likely make me lose faith and not stick to the plan. One arguable advantage of GB is that you may find out it isn't working to plan sooner.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #382 on: October 16, 2016, 10:35:35 AM »
Would anybody care to speculate how GB would have done for a Japanese investor i.e Replace TSM for Nikkei?, SCV and treasuries for Japanese equivalents?

I live in Sweden and was interested in the Permanent Portfolio so I back-traded that portfolio to see how it worked in Sweden and also in Japan as a case study. Turns out the results are greatly different. USA is a unique market because your currency, the Dollar.

when their is stock market panic outside of US. It has no effect on gold whatsoever.

Getting your hands on government bonds and bills can also be rather tricky. In Sweden for instance only wholesalers can buy them from the state and they do not sell retail. The only bonds you can get is through mutual funds that have a mix of different kinds of bonds/bills. State, municipal, corporate and mortgages.

So the permanent portfolio isn't so permanent outside of USA.

I hadn't considered the lesser effects of gold for non-US stock markets. I'm in Australia and there is a possibility of Australia-only issues ahead due to extremely high private debt and expensive housing. The super-heavy market cap weightings of our banks doesn't help the case for our stock market at all.

There are very limited exchange traded bond options here too.

Do you remember any of the specifics of your back-testing results?

Thanks for the insights.
« Last Edit: October 16, 2016, 10:43:31 AM by MrNotRobot »

tommie

  • 5 O'Clock Shadow
  • *
  • Posts: 10
Re: Portfolio Charts - The Golden Butterfly
« Reply #383 on: October 16, 2016, 10:56:03 AM »
i still think a diversified mix of equity's and a dynamic bond portfolio that is ready to shift to different kinds of bond funds as the big picture unfolds is the way to go .

If you think you need some sort of 'dynamic allocation' for bonds i wonder how you imagine this 'dynamic' part to work and i also wonder why this dynamic part isn't included for your other asset types.

I previously outlined and talked about something a bit more dynamic; about that a GB-style portfolio isn't the 100% way to go due to the fact that we simply do not know the future and first och foremost do not know the correlations of different assets in the future. It could however be a base for something else and what it highlights can, in my opinion, be of some use (it highlights that diversification and keeping track of negative correlations isn't the most stupid idea in the world).

In my mind it's also a tad insane to justify a strategy based on data that includes the 70's like todays globalized economy resembles the one in the 70's. More current data should of course carry a lot more weight, else one can simply create portfolios that is optimized for historical data. If you check my post history you'll find that i outlined such a strategy as an example of overoptimizing for historical data.

Personally i do this:
1. Broad diversification (<asset class A, B, C, D...>)
2. $UNRATE + MA12 (12 month average)
3. MA10 for all assets if $UNRATE > MA12 @ $UNRATE

As of october 2016 $UNRATE is above its MA12 average (unemployment is higher then usual for the current 12m period) so my MA10 is set to on for each asset class. This does not mean that i sell everyting, it's basically just something that raises a flag. It's a basic parent -> child relationship.

Radagast

  • Handlebar Stache
  • *****
  • Posts: 1450
  • Location: West of the Mountains, East of the Sea
  • One does not simply work into Mordor
Re: Portfolio Charts - The Golden Butterfly
« Reply #384 on: October 16, 2016, 02:15:00 PM »
Would anybody care to speculate how GB would have done for a Japanese investor i.e Replace TSM for Nikkei?, SCV and treasuries for Japanese equivalents?

I live in Sweden and was interested in the Permanent Portfolio so I back-traded that portfolio to see how it worked in Sweden and also in Japan as a case study. Turns out the results are greatly different. USA is a unique market because your currency, the Dollar.

when their is stock market panic outside of US. It has no effect on gold whatsoever.

Getting your hands on government bonds and bills can also be rather tricky. In Sweden for instance only wholesalers can buy them from the state and they do not sell retail. The only bonds you can get is through mutual funds that have a mix of different kinds of bonds/bills. State, municipal, corporate and mortgages.

So the permanent portfolio isn't so permanent outside of USA.

I hadn't considered the lesser effects of gold for non-US stock markets. I'm in Australia and there is a possibility of Australia-only issues ahead due to extremely high private debt and expensive housing. The super-heavy market cap weightings of our banks doesn't help the case for our stock market at all.

There are very limited exchange traded bond options here too.

Do you remember any of the specifics of your back-testing results?

Thanks for the insights.
If you are in Australia I'd say that definitely changes things. For one, I would guess you'd want more like 30% Australian / 70% international stocks. For another, gold is relatively big in Australia along with other minerals which behave somewhat similarly to gold. The nation's economic fortune is already tied more to the price of gold than it is in many places, especially the US. This might change the relationships between the asset classes quite a bit, but I couldn't tell you what they should be be as I haven't studied investing in Oz.

mathjak107

  • Pencil Stache
  • ****
  • Posts: 558
Re: Portfolio Charts - The Golden Butterfly
« Reply #385 on: October 17, 2016, 01:58:34 AM »
i still think a diversified mix of equity's and a dynamic bond portfolio that is ready to shift to different kinds of bond funds as the big picture unfolds is the way to go .

If you think you need some sort of 'dynamic allocation' for bonds i wonder how you imagine this 'dynamic' part to work and i also wonder why this dynamic part isn't included for your other asset types.

I previously outlined and talked about something a bit more dynamic; about that a GB-style portfolio isn't the 100% way to go due to the fact that we simply do not know the future and first och foremost do not know the correlations of different assets in the future. It could however be a base for something else and what it highlights can, in my opinion, be of some use (it highlights that diversification and keeping track of negative correlations isn't the most stupid idea in the world).

In my mind it's also a tad insane to justify a strategy based on data that includes the 70's like todays globalized economy resembles the one in the 70's. More current data should of course carry a lot more weight, else one can simply create portfolios that is optimized for historical data. If you check my post history you'll find that i outlined such a strategy as an example of overoptimizing for historical data.

Personally i do this:
1. Broad diversification (<asset class A, B, C, D...>)
2. $UNRATE + MA12 (12 month average)
3. MA10 for all assets if $UNRATE > MA12 @ $UNRATE

As of october 2016 $UNRATE is above its MA12 average (unemployment is higher then usual for the current 12m period) so my MA10 is set to on for each asset class. This does not mean that i sell everyting, it's basically just something that raises a flag. It's a basic parent -> child relationship.

like nudging a  big ship to keep it on course there is no bond fund type that is appropriate in advance anymore like it used to be .  as the bigger picture unfolds different types of bond funds will be swapped out

steveo

  • Handlebar Stache
  • *****
  • Posts: 1944
Re: Portfolio Charts - The Golden Butterfly
« Reply #386 on: October 17, 2016, 03:48:03 AM »
If you had of said this, then I would be in 100% agreement:

"Still you have to look at the down side of this type of allocation. Your greatest chance of success better returns over the long term is probably with a 100% stock allocation."

I hear what you are stating but I think those statements are pretty close to the same thing if we are talking a long period of retirement. If you have a long period of retirement which maybe greater than 30 years your greatest chance of success is probably going to align with the greatest chance of maximum returns. Low returns over a long period will get you through shorter time periods but not longer time periods.

Personally I'm a little volatility averse because I want to ensure my portfolio holds up under the short term as well. I have some bonds in my portfolio and I intend to keep them there. I also intend to draw down on bonds at the start of my retirement assuming stocks aren't going gang busters. That is my safety net. At the same time I think that in most situations high equity allocations will give you the greatest chance of success over the longer term.

I think success being subjective is the issue here. If the portfolio doesn't meet my objectives then I wouldn't be inclined to call it a success. One of my objectives is to not only minimize the size of draw-downs but also minimize the length of those draw-downs. I'm even more afraid of longer draw-downs than larger ones because it would likely make me lose faith and not stick to the plan. One arguable advantage of GB is that you may find out it isn't working to plan sooner.

I basically completely agree with your comments here however there is an additional point in that if you are going to be drawing down for a long period of time you will need to have good returns over the lifetime of the portfolio. If you get subpar returns but low drawdowns over 50 years your portfolio might go to nothing and then your screwed.

On your other point I'm also Australian and I just follow the 4% rule and invest in Australia. I use Vanguard ETF's outside of super. At the moment I'm 50% Australian stocks, 25% Bonds and 25% International stocks. Super is 90/10 stocks/bonds/cash with 50% of the stock being international. The outside Super stock allocation is what I intend to draw down at the start of FIRE so I support I want some buffer there. I also own my house which to me is a massive buffer.

For international readers most of our stock market is made up of Banks. I think we are more at risk if our financial sector goes bust rather than the resource sector although the resource sector tanking could impact the financial sector.

Personally I'm not that worried. I can't see Australia going bust unless the rest of the world is also screwed. We didn't really get hit by the GFC for instance. We are too freaken small.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #387 on: October 18, 2016, 06:11:05 AM »
I just want to get a sanity check about these withdrawal rates:



The PWR shows the worst case maximum withdrawal to end up with at least the same amount as you started with, after adjustment for inflation.

So looking at the chart, with the start test year being 1972 and ending year being 2015, that would mean that in ANY of the years from 1972 and 2005 you could have withdrawn 3.6% and end up 10 years later having at least your same balance after inflation (ignoring tax and fees), but quite possibly you could have withdrawn a lot more if you were lucky to have had a good starting year. That's 33 starting years tested! Quite incredible isn't it? Unless I'm interpreting it incorrectly.

Another observation about this chart is that as you move further to the right and look at longer time periods, these results must be based on less tests. For example for the years ranging from 1972 to 2015 there are only 4 tests possible which have a length of 40 years. They are starting 1972, 1973, 1974 and 1975. So to interpret this chart correctly you cannot assume that the better PWRs are solely because of a longer holding period, because they could just look better because of the more limited number of tests, which didn't necessarily start during a bad year. Is that a fair statement?

I've checked all of the portfolios and GB has the highest 10 year PWR of all of them. Quite a few, including TSM, have a negative 10 year PWR. But it needs to be remembered that this is worst case. No doubt there were specific years that would have given much better than GB's 3.6%. How much would people read into this? Would you consider the 10 year PWR a pretty good comparative measure between the portfolios? I suppose it is may only be a comparison of volatility.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 962
  • Location: Cincinnati
Re: Portfolio Charts - The Golden Butterfly
« Reply #388 on: October 18, 2016, 06:13:31 AM »
I spent a little time today looking at alternative portfolios. I wondered if a portfolio much heavier in stocks could beat the GB over long time periods while making 4% withdrawals. Portfolio 1 below is GB. Portfolio 2 is 80% stock/20% total bond. Portfolio 3 is the same stock but with 10% total bond and 10% gold (for the gold bugs).

Data from https://www.portfoliovisualizer.com/backtest-asset-class-allocation
   
Portfolio 1 – Golden Butterfly
Large Cap Blend   20.00%
Small Cap Value   20.00%
Long Term Treasuries   20.00%
Short Term Treasuries 20.00%
Gold   20.00%

Portfolio 2 – Global 80/20
Large Cap Value   10.00%
Large Cap Blend   10.00%
Mid Cap Blend   10.00%
Small Cap Value   10.00%
REIT   10.00%
Intl Stock Market 10.00%
Intl Small Cap Stocks   10.00%
Emerging Markets 10.00%
Total Bond   20.00%

Portfolio 3 – Global 80/10 Total Bond/10 Gold

$1M, withdraw an initial $40K adjusted for inflation, annual rebalancing.

Real CAGR GB vs Global 80/20 vs 80/10/10

1972-2015   3.35% vs 5.46% vs 6.22%
1986-2015   2.65% vs 6.11% vs 5.80%
1996-2015   1.82% vs 2.97% vs 2.92%
2001-2015   1.61% vs 1.81% vs 2.49%
2006-2015   1.51% vs 0.36% vs 0.91%

The global 80/20 beats the GB in all cases except the last 10 years. Surprisingly, the 80/20 was better even when starting in 1972, though it didn’t begin to power ahead until 1984.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 962
  • Location: Cincinnati
Re: Portfolio Charts - The Golden Butterfly
« Reply #389 on: October 18, 2016, 06:26:25 AM »
I've checked all of the portfolios and GB has the highest 10 year PWR of all of them. Quite a few, including TSM, have a negative 10 year PWR. But it needs to be remembered that this is worst case. No doubt there were specific years that would have given much better than GB's 3.6%. How much would people read into this? Would you consider the 10 year PWR a pretty good comparative measure between the portfolios? I suppose it is may only be a comparison of volatility.

Why is the 10 year PWR so important? Why not 5 year or 15 year? On the 15 year the GB isn't so special.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #390 on: October 18, 2016, 06:39:31 AM »

Real CAGR GB vs Global 80/20 vs 80/10/10

1972-2015   3.35% vs 5.46% vs 6.22%
1986-2015   2.65% vs 6.11% vs 5.80%
1996-2015   1.82% vs 2.97% vs 2.92%
2001-2015   1.61% vs 1.81% vs 2.49%
2006-2015   1.51% vs 0.36% vs 0.91%

The global 80/20 beats the GB in all cases except the last 10 years. Surprisingly, the 80/20 was better even when starting in 1972, though it didn’t begin to power ahead until 1984.

Yes, the GB is beaten. But looking at the results between both 80/20 portfolios, I would say that the small gold allocation has a good effect compared to the 80/20 with no gold.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #391 on: October 18, 2016, 06:41:14 AM »
I've checked all of the portfolios and GB has the highest 10 year PWR of all of them. Quite a few, including TSM, have a negative 10 year PWR. But it needs to be remembered that this is worst case. No doubt there were specific years that would have given much better than GB's 3.6%. How much would people read into this? Would you consider the 10 year PWR a pretty good comparative measure between the portfolios? I suppose it is may only be a comparison of volatility.

Why is the 10 year PWR so important? Why not 5 year or 15 year? On the 15 year the GB isn't so special.

Because the 10 year is based on a greater number tests than the 15 year. The 5 year would be great to see, but it is not provided.

arebelspy

  • Administrator
  • Senior Mustachian
  • *****
  • Posts: 27688
  • Age: -999
  • Location: Traveling the World
Re: Portfolio Charts - The Golden Butterfly
« Reply #392 on: October 18, 2016, 06:42:37 AM »
I spent a little time today looking at alternative portfolios. I wondered if a portfolio much heavier in stocks could beat the GB over long time periods while making 4% withdrawals. Portfolio 1 below is GB. Portfolio 2 is 80% stock/20% total bond. Portfolio 3 is the same stock but with 10% total bond and 10% gold (for the gold bugs).

Data from https://www.portfoliovisualizer.com/backtest-asset-class-allocation
   
Portfolio 1 – Golden Butterfly
Large Cap Blend   20.00%
Small Cap Value   20.00%
Long Term Treasuries   20.00%
Short Term Treasuries 20.00%
Gold   20.00%

Portfolio 2 – Global 80/20
Large Cap Value   10.00%
Large Cap Blend   10.00%
Mid Cap Blend   10.00%
Small Cap Value   10.00%
REIT   10.00%
Intl Stock Market 10.00%
Intl Small Cap Stocks   10.00%
Emerging Markets 10.00%
Total Bond   20.00%

Portfolio 3 – Global 80/10 Total Bond/10 Gold

$1M, withdraw an initial $40K adjusted for inflation, annual rebalancing.

Real CAGR GB vs Global 80/20 vs 80/10/10

1972-2015   3.35% vs 5.46% vs 6.22%
1986-2015   2.65% vs 6.11% vs 5.80%
1996-2015   1.82% vs 2.97% vs 2.92%
2001-2015   1.61% vs 1.81% vs 2.49%
2006-2015   1.51% vs 0.36% vs 0.91%

The global 80/20 beats the GB in all cases except the last 10 years. Surprisingly, the 80/20 was better even when starting in 1972, though it didn’t begin to power ahead until 1984.

You are too focused on CAGR.

The trinity study demolished CAGR as the measurement for ER success, due to sequence of returns risk. 

The CAGR may be 7% over a time frame, for example, but if bad returns hit early when you're withdrawing, the portfolio may never recover.  This is why looking at returns in conjunction with withdrawals is super important (e.g. cFIREsim scenarios).

Just CAGR, as you've done, is great for accumulation, when there's no withdrawals (and, in fact, extra volatility is likely good in accumulation).  So heavy stocks in accumulation is almost definitely the way to go, if you can stick with it during down markets and not sell.

But that isn't necessarily true in ER.

I have other issues with the GB, but it having a bit lower CAGR isn't one of them, as the point of it isn't the highest CAGR, but the highest safe withdrawal rate, which takes into account both CAGR and sequence of returns.
We are two former teachers who accumulated a bunch of real estate, retired at 29, and now travel the world full time with two kids.
If you want to know more about me, or how we did that, or see lots of pictures, this Business Insider profile tells our story pretty well.
We (rarely) blog at AdventuringAlong.com. Check out our Now page to see what we're up to currently.

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #393 on: October 18, 2016, 06:51:44 AM »
Imagine a scenario your retirement is 50 years, but the last 10 years of your life just happened to be a very poor 10 years for your portfolio. You are no longer caring that over the long term things will work out great.

Personally, if a portfolio were to provide a worst-case PWR of 3.6% over my entire retirement period I would be be over the moon. (Not that anything is guaranteed, as we know)
« Last Edit: October 18, 2016, 07:07:06 AM by MrNotRobot »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #394 on: October 18, 2016, 06:57:12 AM »

Why is the 10 year PWR so important? Why not 5 year or 15 year? On the 15 year the GB isn't so special.

Because the 10 year is based on a greater number tests than the 15 year. The 5 year would be great to see, but it is not provided.

Also, the most recent 15 year test would have been the year 2000, whereas the most recent 10 year test would have been 2005. Personally I think there is a case for giving slightly more credence to more recent tests.

AdrianC

  • Pencil Stache
  • ****
  • Posts: 962
  • Location: Cincinnati
Re: Portfolio Charts - The Golden Butterfly
« Reply #395 on: October 18, 2016, 08:49:54 AM »
You are too focused on CAGR.
...

The data I gave included an initial 4% WR indexed for inflation.

The 80/20 *usually* beats the GB while using a 4% SWR.
« Last Edit: October 18, 2016, 09:12:09 AM by AdrianC »

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #396 on: October 18, 2016, 09:52:53 AM »
Maybe we could start looking at this from another angle. If the GB gave 3.6% PWR for over 3 decades, what type of scenarios could cause it not to in the future. A bond market crash is one possibility.

P.S. I appreciate all posts being made here :) I can clearly see in my posts I am biased towards the point of view of the retirement years and I am selfishly looking at it from that perspective (for me this means I want lower volatility... and I'm willing to pay for it)

Daniel S

  • 5 O'Clock Shadow
  • *
  • Posts: 60
Re: Portfolio Charts - The Golden Butterfly
« Reply #397 on: October 18, 2016, 11:55:35 AM »
Would you consider the 10 year PWR a pretty good comparative measure between the portfolios? I suppose it is may only be a comparison of volatility.

I have other issues with the GB, but it having a bit lower CAGR isn't one of them, as the point of it isn't the highest CAGR, but the highest safe withdrawal rate, which takes into account both CAGR and sequence of returns.

So a low-length worst-case PWR/SWR could actually reflect sequence of returns risk, rather than broader volatility?

k9

  • Stubble
  • **
  • Posts: 241
  • Age: 39
Re: Portfolio Charts - The Golden Butterfly
« Reply #398 on: October 18, 2016, 03:54:32 PM »

1986-2015 (30 years)
GB final balance $4,059,788 CAGR real 2.12%
60/40 final balance $7,110,707 CAGR real 4.05%

You did just fine. But your neighbor who went with a simple 60/40 and rode out the extra volatility did quite a bit better. The GB gave a smoother ride, but quite a lot less money.
But then the guy who went all in on emerging markets made about 11% CAGR ! Who wants to make only 4% when you can earn an amazing 11% per year ?
I'm sarcastic yet a little serious. If you answer this question seriously (and ignore the fact I'm semi-trolling) you'll probably end up with arguments that answer the question "why would one choose GB instead of 60/40 ?" (or any mix of assets, anyway)

k9

  • Stubble
  • **
  • Posts: 241
  • Age: 39
Re: Portfolio Charts - The Golden Butterfly
« Reply #399 on: October 18, 2016, 03:57:48 PM »
why buy Japanese stocks just because you may live there ?    i would never buy in a country i wasn't happy with market or economic wise .
And then, the question is "what event would make you turn away from US stocks, and how would you know you're not just selling at the worst possible moment ?" Were you happy with US market/economy in 2008 ?