Portfolio Charts - The Golden Butterfly

[yoda34]

https://portfoliocharts.com/portfolio/golden-butterfly/

Have any of you looked at the Golden Butterfly over at Portfolio Charts? It's intriguing me....

It goes against most of my investing philosophy, is heavy on treasuries and gold, and yet has pretty decent returns with a much lower std than the market as measured by many of the benchmarks.

I've run independent back test with CapIQ and other data sets using different start years and it is remarkably consistent over the past 15-20 years.

Because of the lower volatility it supports a little bit higher SWR - which has all kinds of benefits for FIRE.

Just wondering if other folks have looked at it, what your thoughts are, and if anyone is actually investing using this allocation.

[wienerdog]

I think Tyler created that portfolio as he is a fan of the Permanent Portfolio and was trying something a little different but spun off of the PP.  You can read his commentary below.  The gold portion doesn't sit well with people here so you probably won't find anyone using it.

https://portfoliocharts.com/2015/09/22/catching-a-golden-butterfly/

The portfolio charts website isn't is an excellent tool and we are lucky to have people like Tyler here.  I am sure he will chime in.

Edit: Changed isn't to is.  Sorry Tyler.  LOL

[workathomedad]

Portfolio Charts is awesome!

Tyler independently verified how awesome the Permanent Portfolio / Golden Butterfly is with his Portfolio Finder tool:

https://portfoliocharts.com/portfolio/portfolio-finder/

You punch in a desired historical CAGR and it gives you the portfolio with the lowest volatility / risk, which is *exactly* what you, as an early retiree, need and want! Lower volatility / drawdowns translates directly into a higher Safe Withdrawal Rate over longer terms.

Almost any number you punch in <10% independently comes up with a "Golden Butterfly" type portfolio.

I didn't know it was a forum member's website. Thank you for the amazing tools!

[AdrianC]

You punch in a desired historical CAGR and it gives you the portfolio with the lowest volatility / risk, which is *exactly* what you, as an early retiree, need and want! Lower volatility / drawdowns translates directly into a higher Safe Withdrawal Rate over longer terms.

Portfolio charts is very good and Tyler has done excellent work. However...just remember that we are looking at a back-test.

"Past Performance is Not Necessarily Indicative of Future Results".

[Tyler]

I'm glad you guys find the site helpful!

Wienerdog is correct -- I've been a Permanent Portfolio fan for a while, and the Golden Butterfly is an offshoot of the same theory.  Anyone interested in the Golden Butterfly might also be interested in reading the books supporting the Permanent Portfolio.  While past performance does not guarantee future returns, it does verify that the GB has worked as designed and done quite well in all types of economic conditions including ones where other portfolios have really struggled.

Also, Workathomedad correctly observes that the specific funds are up for personal interpretation.  There's a big cluster of similar portfolios around the Golden Butterfly in the Portfolio Finder that have two parts stocks, two parts bonds, and one part commodities/gold/REITs.  So don't let one asset you struggle with (most commonly gold or SCV) deter you from the concept altogether.  Buy the strategy, not the assets.   

I recently pulled the trigger and transitioned from the Permanent Portfolio to a slightly modified Golden Butterfly.  The only real difference is that I chose a combination of total stock market and small cap blend over large cap blend and small cap value (I already owned VTI, and I liked VB as its natural complement).  I believe that different people require different portfolios, so I'm not at all suggesting that everyone should do the same.  I'm happy to be the guinea pig and report back how it goes. 

[Interest Compound]

It goes against most of my investing philosophy, is heavy on treasuries and gold, and yet has pretty decent returns with a much lower std than the market as measured by many of the benchmarks.

...

Because of the lower volatility it supports a little bit higher SWR - which has all kinds of benefits for FIRE.

Tyler. This is why you got so much flak in your other thread. This kind of information lures financial amateurs in like a moth to the flame. Your Golden Butterfly page has quotes like, "match or beat the total stock market in long-term real CAGR with fewer stocks and lower volatility". What could possibly go wrong? While you're experienced enough to understand the inherent risks, the vast majority of people reading this information simply aren't.

Yoda34, do you understand the inherent risks of Retiring Early with a portfolio of 40% stocks, 40% treasuries, and 20% a single commodity?

Do you?

I'm genuinely not sure what you can do about this Tyler, as the disclaimers on your site are pretty clear. I just don't think your readers can responsibly handle the information you're giving them.

[Tyler]

Tyler. This is why you got so much flak in your other thread. This kind of information lures financial amateurs in like a moth to the flame. Your Golden Butterfly page has quotes like, "match or beat the total stock market in long-term real CAGR with fewer stocks and lower volatility". What could possibly go wrong? While you're experienced enough to understand the inherent risks, the vast majority of people reading this information simply aren't.

Yoda34, do you understand the inherent risks of Retiring Early with a portfolio of 40% stocks, 40% treasuries, and 20% a single commodity?

Do you?

I'm genuinely not sure what you can do about this Tyler, as the disclaimers on your site are pretty clear. I just don't think your readers can responsibly handle the information you're giving them.

I hear what you're saying, but I give people more credit than that.  If factual data that focuses on the big picture gets people excited about learning more about asset allocation, I consider that a success.   As JZinCO points out, a major goal of the site is to help visualize what uncertainty and risk really look like so that people can compare alternatives and make educated decisions. 

FWIW, I'm personally comfortable enough with the Golden Butterfly as an early retirement portfolio that I'm investing that way with my own money.  YMMV.

[Interest Compound]

I've run independent back test with CapIQ and other data sets using different start years and it is remarkably consistent over the past 15-20 years.

Of course it has been remarkably consistent. Why else do you think it's on your radar? Why else do you think this portfolio, out of the almost infinite possibilities of portfolios, caught your eye? A specific portfolio being remarkably consistent should not surprise you. If it wasn't this one, it'd be some other random portfolio. With a large enough dataset, it is expected to find outliers, but don't be confused...that doesn't mean anything can be learned from these outliers.

Here's a fun example. Imagine yourself in 1972. Doing this same analysis. You're about to choose a similar slice-n-dice portfolio that looked remarkably consistent over the previous ~60 years (1915-1972).

• What kind of portfolio do you think you would've ended up with? Do you think it would've looked anything like The Golden Butterfly?
• How would that have done over the next 40 years?
• The 1972 version of yourself had access to millions of different portfolios, what are the chances you would've been successful if they employed this same logic?

 

Now for some data (last 100 years):

The Golden Butterfly is 40% bonds. Let's see how bonds were doing in the years before 1972:



The Golden Butterfly is 20% gold. Let's see how gold was doing in the years before 1972:



Do you believe the 1972 version of you would've chosen these two assets for 60% of your Early Retirement portfolio?

I'm not sure what you'll conclude from this example, but to me it's quite clear. There is essentially 0 chance I would've chosen this portfolio, so nothing can be learned from studying an outlier like The Golden Butterfly.

[MustacheAndaHalf]

Gold should not be back tested in 1973-1974 without a large disclaimer: it tripled in those years, earning over 70% per year.  Back testing assumes the past did not have one off events, but does anyone expect gold to triple again in a 2 year period?

How much did gold beat inflation 1972-2015?  About 2.9% ("all the available data", or 44 years)
How much did gold beat inflation 1976-2015?  About 1.0% ("the last 40 years")
https://www.portfoliovisualizer.com/backtest-asset-class-allocation

[yoda34]

I've run independent back test with CapIQ and other data sets using different start years and it is remarkably consistent over the past 15-20 years.

Of course it has been remarkably consistent. Why else do you think it's on your radar? Why else do you think this portfolio, out of the almost infinite possibilities of portfolios, caught your eye? A specific portfolio being remarkably consistent should not surprise you. If it wasn't this one, it'd be some other random portfolio. With a large enough dataset, it is expected to find outliers, but don't be confused...that doesn't mean anything can be learned from these outliers.

Here's a fun example. Imagine yourself in 1972. Doing this same analysis. You're about to choose a similar slice-n-dice portfolio that looked remarkably consistent over the previous ~60 years (1915-1972).

• What kind of portfolio do you think you would've ended up with? Do you think it would've looked anything like The Golden Butterfly?
• How would that have done over the next 40 years?
• The 1972 version of yourself had access to millions of different portfolios, what are the chances you would've been successful if they employed this same logic?

 

Now for some data (last 100 years):

The Golden Butterfly is 40% bonds. Let's see how bonds were doing in the years before 1972:



The Golden Butterfly is 20% gold. Let's see how gold was doing in the years before 1972:



Do you believe the 1972 version of you would've chosen these two assets for 60% of your Early Retirement portfolio?

I'm not sure what you'll conclude from this example, but to me it's quite clear. There is essentially 0 chance I would've chosen this portfolio, so nothing can be learned from studying an outlier like The Golden Butterfly.

Wow.  Not sure where to begin with that. First, I don't see anywhere that I said I was putting my entire FIRE Portfolio into that allocation (or even a single dollar). I simply asked what other members thought of that allocation and thoughts. The point of this forum is to discuss these things, correct? I didn't realize that certain topics were censored or would generate such strong response.

A couple of thoughts about back testing. We all know that past results don't make any kind of performance guarantee on the future, but people seem to forget that it works both ways. A common mistake in backtesting (beyond the normal survivor bias, mispricing etc) is the tendency to forget that the economic environment, political environment etc was vastly different in the past than today. Let's take your Gold chart above for a few moments (bear in mind that I hate gold as an asset and most likely will never put any significant portion of my portfolio in gold). Does how Gold performed in the 1920s have any bearing on gold today? Are there perhaps massive changes in the way we tie gold into our financial system that could change the way it performs in the market. Are there perhaps not several international agreements, wars, and perhaps a shock in1971 that totally changed the way gold performs in the market?

It's the basic problem with backtesting. Too little and you don't have enough information to be meaningful at all, too long (especially with a single asset commodity such as gold) and you include data that is nonsensical in the context in which you are trying to review.

Finally, I appreciate your concern about all the amateurs on the site but I think I'll be just fine. I passed the 7 and the 66, I've worked in the industry for several years, and yes I know how to calculate a CAGR appropriately

[Tyler]

I'm not sure what you'll conclude from this example, but to me it's quite clear. There is essentially 0 chance I would've chosen this portfolio, so nothing can be learned from studying an outlier like The Golden Butterfly.

I gather you don't care for backtesting and prefer your own investment method.  I admire your confidence, and I encourage you to keep doing what you're doing.  Others find value in looking at how portfolios have performed in both good times and bad to compare options and to learn from history.  We can only work with the data we have available, but what else are we supposed to go on?  To each his own.  I choose to agree to disagree rather than argue point by point. 

Gold should not be back tested in 1973-1974 without a large disclaimer: it tripled in those years, earning over 70% per year.  Back testing assumes the past did not have one off events, but does anyone expect gold to triple again in a 2 year period?

How much did gold beat inflation 1972-2015?  About 2.9% ("all the available data", or 44 years)
How much did gold beat inflation 1976-2015?  About 1.0% ("the last 40 years")
https://www.portfoliovisualizer.com/backtest-asset-class-allocation

Your gold disclaimer is now included on its asset page. 

The Golden Butterfly information linked in the OP also displays all of the timeframes beginning after 1974, including the worst case for gold beginning in 1980 when it lost over 80% if its value over the next 20 years.  One can easily see the impact of start dates and draw their own conclusions on how important certain years are to a portfolio, which may perform much differently than the individual underlying assets by design.  IMHO, the Golden Butterfly stands up quite well.  But I agree that it (as originally described) is a poor choice for individuals with a strong distaste for gold. 

[AZryan]

-I don't see anywhere that I said I was putting my entire FIRE Portfolio into that allocation (or even a single dollar).-

Right. But being highly intrigued does kinda imply you're seriously looking into going this direction, and that's pretty much how I saw the counterpoint presented to you. Seemed fair.

I didn't realize that certain topics were censored or would generate such strong response.

Nothing got censored. You're over-reaching in your defensiveness. And yeah, it got a strong response because I.C. sees some pretty major warning signs in this portfolio. Just the sort of thing that gets strong responses. Again... pretty fair.

Finally, I appreciate your concern about all the amateurs on the site but I think I'll be just fine. I passed the 7 and the 66, I've worked in the industry for several years, and yes I know how to calculate a CAGR appropriately

Again... pretty overly defensive IMO. Usually it's amateurs that are still sorting out what their basic overall asset allocations are -as you seemed to imply. But post information can be useful to others who are just reading and not directly participating, so it doesn't mean it's all about you.
Also... what the hell does 'I passed my 7 and the 66' mean? And 'I've worked in the industry for several years' is terribly vague if it's meant to show how you know stuff people don't know you know, ya' know?

[wienerdog]

Ahhhh the three Gs creating forum posts daily.

I think OP is referencing  Series 7 and Series 66 licenses.

[AZryan]

Tyler,
It seems beyond dispute that the Golden Butterfly is a tremendously intriguing portfolio. The debate, of course, is whether it's sensible for the future. And I see the issue coming down almost entirely to being a question about GOLD.
Would you disagree?
Would I.C. disagree?

Beyond gold, it's just the extreme ends (most and least volatile) of basically stocks and bonds (treasuries).

I don't know a lot about gold, but it seems like it has had some radically crucial changes to it compared to the ups and downs of stocks/bonds, so it's backtesting appears to be awful for projecting the future -unlike the usefulness that stock/bond backtesting 'seems' to have.

Maybe you need to present a good argument for holding gold in the future? The backtesting doesn't seem to make that argument at all IMO. If it did, I'd be all in on the golden butterfly.

But I agree that it (as originally described) is a poor choice for individuals with a strong distaste for gold.

Why? You word it like a tautology -'gold is bad for people who don't like gold'?

Would you say that 'vaccines are a poor choice for anti-vaxxer parents' whose beliefs are based on unscientific nonsense? No. Their judgment is demonstrably wrong. Their opinion doesn't matter.

I think gold's a bad choice for me, but that doesn't mean it actually is. I could simply be totally wrong, and I might be much better off with 20% gold in my portfolio as a uniquely excellent diversifier.

[AZryan]

Ahhhh the three Gs creating forum posts daily.

What's that mean?

I think OP is referencing  Series 7 and Series 66 licenses.

Ah, the old rhetorical trick of referencing the most obscure insider jargon-type words or phrases without any explanation of the point.

The trick is to have the least amount of people understand you, in the hope that this will make more people believe you have super special/secret/advanced knowledge on a subject that implies their actual questionable points are somehow better reasoned than they might actually be.

I saw this a lot on high end audio forums. Pure B.S. 9 times outta 10.

[MustacheAndaHalf]

Tyler - I appreciate the disclaimer of gold's unique 1973-1974 performance.

[wienerdog]

What's that mean?

God, Guns and Gold they always run the post count up.

Ah, the old rhetorical trick of referencing the most obscure insider jargon-type words or phrases without any explanation of the point.

The trick is to have the least amount of people understand you, in the hope that this will make more people believe you have super special/secret/advanced knowledge on a subject that implies their actual questionable points are somehow better reasoned than they might actually be.

I saw this a lot on high end audio forums. Pure B.S. 9 times outta 10.

Don't blame me I only have a drivers license.  Look at the software he used to back test.  Didn't look like something the average joe would have access to.

[workathomedad]

Wow, I can't believe how upset people get at showing backtest results from a portfolio different than their favorite allocation.

I've read and seen a lot about portfolio allocation, and it's pretty easy to beat a classic 60/40 portfolio in either drawdown, volatility, or returns by adding additional diversification into commodities, real estate, gold, international, etc.

http://mebfaber.com/2013/07/31/asset-allocation-strategies-2/

No portfolio allocation is sacrosanct, and no portfolio allocation will replicate it's backtesting results... including whatever you chose.

What's super cool about the Permanent Portfolio types is just how amazingly well they have down in reducing volatility and drawdown, which is uniquely beneficial for early retirees. They certainly are not designed to optimize return, which almost everyone agrees would require a higher stock allocation. When that stock allocation is down >50% though, if you're expecting a higher SWR, you may need to adjust your spending during those years or face earlier portfolio depreciation.

[Tyler]

Maybe you need to present a good argument for holding gold in the future? The backtesting doesn't seem to make that argument at all IMO. If it did, I'd be all in on the golden butterfly.

That's where we may have a fundamentally different perspective.  I do not recommend any asset for its projected performance in the future.  I prefer to evaluate asset allocations by first assuming that I'm really unlucky and to check what the worst times have been for such a portfolio, and the site tools intentionally show all of the best and worst times side by side (for as many years as I can get my hands on).  Now that doesn't mean the future won't be the new worst case, but IMHO it's a reasonable place to start and a healthy application of backtesting.  Truly diverse portfolios that don't bet so heavily on one asset (including stocks) naturally rise to the top, and gold can be an important ingredient in some of these options.  My best succinct explanation to date is in this link. 

I understand that my neutral-outlook mentality runs counter to the typical value investing approach.  I'm heavily influenced by the writings of Harry Browne (Ray Dalio is also on the same page), so reading his books (I recommend this) will help you understand my perspective.  It's deserving of its own full-length post to do the topic justice -- thanks for the idea! 

But I agree that it (as originally described) is a poor choice for individuals with a strong distaste for gold.

Why? You word it like a tautology -'gold is bad for people who don't like gold'?

Would you say that 'vaccines are a poor choice for anti-vaxxer parents' whose beliefs are based on unscientific nonsense? No. Their judgment is demonstrably wrong. Their opinion doesn't matter.

If you don't get vaccines, you could die.  If you don't invest in gold, there are other good portfolio options that don't require it. 

A better metaphor might be cooking.  Imagine gold is like cinnamon.  In the right proportions, I find cinnamon to be a delicious addition to certain baked sweets.  But it's completely inedible on its own, and anyone suggesting that cinnamon is a universally bad ingredient that nobody should use because they once tried it on their favorite sandwich and didn't like the results would be rightly mocked.  Still, many people genuinely dislike cinnamon even when it's used properly.  Who am I to tell them that cinnamon cookies are unquestionably the best choice when other options make them much happier and are clearly tasty in their own right?  Some recipes work better with different ingredients, and some people prefer different recipes. 

While I do believe the facts support that gold can play an important role in certain portfolios (including the Golden Butterfly), I am practical about it and concede that far more important than owning gold is owning a portfolio that you believe in and are comfortable with.  A portfolio you can't hold for the long run often does more harm than good. 

[yoda34]

Ahhhh the three Gs creating forum posts daily.

What's that mean?

I think OP is referencing  Series 7 and Series 66 licenses.

Ah, the old rhetorical trick of referencing the most obscure insider jargon-type words or phrases without any explanation of the point.

The trick is to have the least amount of people understand you, in the hope that this will make more people believe you have super special/secret/advanced knowledge on a subject that implies their actual questionable points are somehow better reasoned than they might actually be.

I saw this a lot on high end audio forums. Pure B.S. 9 times outta 10.

You're right, perhaps I was a bit defensive. I shouldn't have posted in between meetings when I just had read the thread and got irritated, so my apologies for that.

A few points - saying I'm intrigued about something, at least to me, doesn't mean that I'm rushing to invest or even need you to explain it to me. I asked simply for a discussion of the portfolio. This forum has some pretty smart folks, and in general, I value what people have to say - that's why I read it. Some appropriate responses could have been a discussion of the bond barbell with long and short term treasuries. I think the discussion about not liking gold could is also appropriate if you just turned down the attitude a bit. That's what I'm looking for - reasoned discussion about the pros and cons of an idea. I admit I over reacted, but I still do not believe that's what IC did.

I also get irritated when people post misleading backtests to make a point (either for or against an allocation). The fact that our approach to currency and gold fundamentally changed in 1971 is well known - using charts in 1920 is a form of twisting data to make your point stronger and yes, it irritates me.

While I really appreciate you calling me a liar or full of BS, the Series 7, 65, 66, CFA I II III etc are not obscure or some secret thing. They are literally the tables stakes to have any job in financial services for a broker, RIA firm, or analyst. If you don't know what they are, or didn't recognize it, then you truly do not have any clue about the industry. It's not my fault that you're ignorant of the subject matter. If you want to call BS because you don't understand that's fine *shrug*

This thread was not intended to be about Gold. It was about a specific allocation including long term bonds, short term bonds, total stock market, value funds, and yes some gold. The fact that we can't have a conversation about it is kinda sad. Regardless this thread will clearly not accomplish what I had hoped and my post here is wildly off topic so I'm done with this discussion and this thread. Hope everyone has a good night.

[AZryan]

Tyler,

I don't consider big index funds, esp. the Total Stock Market, to be reasonably called 'one asset'. One asset class, sure. But it's super diversified on it's own, and it pretty much beats most other allocations divided among various other types of asset classes. You know this. We're both working with the exact same facts.

I don't think your cooking with cinnamon analogy works in reference to gold because neither of us are talking about using gold/cinnamon as their only investment/ingredient. We both agree it would be horrible.
And that some people don't like how cinnamon tastes doesn't correlate to the actual value of a gold investment. It only works if you're talking about how gold might make people scared (and you are). But that fear aspect is not what I'm talking about. I'm only talking about the actual value gold brings to a portfolio.

So, like vaccines, you could be an irrational anti-vaxxer, but happens you and your kids don't get sick and die. Most of these dummies don't because of herd immunity. But it doesn't change the fact that the vaccines do provably work if you do use them properly and contact the thing they guard against.
Meaning...
Getting 'scared and changing your portfolio' should NOT be a factor in evaluating an asset's real or potential performance within that portfolio.

And again... it really does seem that ~20% gold is the main factor that makes the Golden Butterfly so unique and kick ass. Yes, there are other portfolios that can get similar results, but there really isn't anything that you can swap gold for to get similar results in the Golden Butterfly, right? And it seems that shoving ~20% gold into damn near any other mix of stuff has turned out great because it was so uncorrelated to anything else.

[AZryan]

Yoda34,

It seems like you really haven't decided if you were being too defensive or not.

I didn't call you a liar or full of BS, but yeah... posting 'I passed the 7 and 66' is a pretty obscure reference. Has it EVER been referenced before in the literally thousands upon thousands of posts in this entire forum? It smacked of the 'argument from authority' logical fallacy.

It's like me saying you used a comma wrong because I'm a published author. It's a daffy way to argue.

Who cares if I know jack about 'the industry'!? I never claimed to, and it doesn't matter to any of this. Think you need to 'turn down the attitude'.

Just go back to the actual investment topics at hand.

"-This thread was not intended to be about Gold. It was about a specific allocation including long term bonds, short term bonds, total stock market, value funds, and yes some gold.-"

Gold is the one very particular, unique, equally large, and highly questionable aspect of this portfolio.
Literally everyone except you appears basically fine with the discussion and the meaningful tangents flowing from it.

You're free to post about other aspects, and that's fine, too. No need to take your toys and storm off.

[wienerdog]

Regardless this thread will clearly not accomplish what I had hoped and my post here is wildly off topic so I'm done with this discussion and this thread. Hope everyone has a good night.

I think if you stick around you'll realize the ones that preach tolerance are usually the most intolerant.  It's never fun coming to a thread and getting called amateur and then the BS card called on your credentials.  A quick google of your CapIQ program would have easily turned the sure 90% bet into realizing they were in the 10%.  Anyway enjoy your evening.  Sorry your post turned into this.   

[Tyler]

And again... it really does seem that ~20% gold is the main factor that makes the Golden Butterfly so unique and kick ass. Yes, there are other portfolios that can get similar results, but there really isn't anything that you can swap gold for to get similar results in the Golden Butterfly, right? And it seems that shoving ~20% gold into damn near any other mix of stuff has turned out great because it was so uncorrelated to anything else.

Dropping the analogies, gold does have a demonstrably positive effect on many portfolios.  That includes the Golden Butterfly, as it complements the other assets quite well.  A lot of it does come down to correlations, as it's rare to find an asset negatively correlated to both stocks and bonds.  But gold isn't inherently magical and other assets can also help. 

It's never fun coming to a thread and getting called amateur and then the BS card called on your credentials. ... Anyway enjoy your evening.  Sorry your post turned into this.   

+1

[arebelspy]

Don't let a single poster or two, on a forum of thousands, drive you off.  The internet can be a rough place sometimes, but it's worth it; if you want discussion, stick around.  :)

[arebelspy]

I'm not sure what you'll conclude from this example, but to me it's quite clear. There is essentially 0 chance I would've chosen this portfolio, so nothing can be learned from studying an outlier like The Golden Butterfly.

I gather you don't care for backtesting and prefer your own investment method.  I admire your confidence, and I encourage you to keep doing what you're doing.  Others find value in looking at how portfolios have performed in both good times and bad to compare options and to learn from history.  We can only work with the data we have available, but what else are we supposed to go on?

I do think Interest Compound makes a good point, in that if you were picking a portfolio based on backtesting, at the point in time of, say, 1970, you probably wouldn't pick anything close to the Golden Butterfly.

[Tyler]

I do think Interest Compound makes a good point, in that if you were picking a portfolio based on backtesting, at the point in time of, say, 1970, you probably wouldn't pick anything close to the Golden Butterfly.

That may be true, especially since investing in gold made no logical sense prior to 1971 because of Bretton Woods.  But the average investor in the 70's wouldn't exactly be beating down the door for stocks, either.  Nobody can see the future.

http://www.crestmontresearch.com/docs/Stock-Matrix-Tax-Exempt-Real3-11x17.pdf

In any case, I think that's a red herring.

I'm not at all suggesting that anyone should make their investing decisions (Golden Butterfly or otherwise) based simply on backtesting.  As I've said elsewhere, relying solely on good backtested numbers with no understanding of why a portfolio worked (and may continue to work) is a bad idea.  But making decisions based solely on dogma while ignoring history (or assuming the data that contradicts your beliefs must somehow be misleading) is similarly shortsighted.  Wise investors use both theory and history together as complementary pieces to the puzzle.

For the theory beyond the attractive backtests, I've already recommended a few supporting books. 

[wienerdog]

And again... it really does seem that ~20% gold is the main factor that makes the Golden Butterfly so unique and kick ass. Yes, there are other portfolios that can get similar results, but there really isn't anything that you can swap gold for to get similar results in the Golden Butterfly, right?

Nope

Read the other thread Monkey Uncle summarizes it very well:

Rather, there are a few principles that seem to be emerging:

1. Focus the core of the portfolio on high-return, high-risk asset classes.
2. Those high-risk, high-return assets should be un-correlated as much as possible to balance out the ups and downs (i.e., don't just divide everything among various categories of u.s. small caps, for example). 
4. Some exposure to real estate and commodities seems to be a recurring theme.
3. Keep about a third of the portfolio in low-risk bonds to further cushion the downs during major melt downs when panic-selling impacts almost all equity-type asset classes (e.g., 2008).

Substitute REIT in for gold.  A little higher return than the GB but the std dev is up 4% so the of course the Sustainable WR is down from 5.2% to 4.6%

[YoungInvestor]

I have no interest for Treasuries with current interest rates. Their superb performance over the last 40 years is simply due to ever-decreasing rates. Gold has similar circumstances.

40% global stocks, 20% short-term bonds, 20% reits and 20% self-directed equity investments makes sense.(edit: to me, it makes sense, for my own situation).

[MustacheAndaHalf]

Meb Faber, in his book "Global Asset Allocation" claims all portfolios do the same because he makes them into the same thing.  He literally can't tell the difference between the S&P 500 and Berkshire Hathaway: his book calls the S&P 500 the Buffet portfolio.  I happen to like Larry Swedroe - who Faber represents by removing small/value stocks!  So get another source if you've been convinced by Meb Faber that all portfolios are similar.

If someone invests in gold after back testing for 40 years (1976-2015), so be it.  To me the disclaimer Tyler agrees with is the whole story: know what happened with gold, and decide with knowledge either way.

If you're in retirement and need "downturn insurance", you could consider gold.  During a sharp stock price drop, gold tends to move opposite.  If you need to pull money out when that happens, you can sell gold to cover expenses or to rebalance.  Personally I decided not to do this since gold has no expected rate of return - it requires someone else to buy at a higher price.  But gold does work differently than bonds or stocks, and could improve draw downs if that's important to you.

[bryan]

If I were to be serious about gold, I would probably modify the data to smooth out the jump due to the unlocking of the official exchange price from the market price. Similar story to other mispriced currencies or assets today.

wouldn't a value investor back in ~1970 be piling into gold?Seeing how expensive it was in India, Hong Kong, etc?

[Tyler]

Wise investors use both theory and history together as complementary pieces to the puzzle.

I wrote this in response to a PM and figured it would also be good to share.  Here's a high-level overview of the theory behind the GB.

The Golden Butterfly is certainly unintuitive and goes against conventional investing wisdom.  The natural instinct of many people is to immediately seek out any possible outlying data that they assume must make it look so good (bond tailwind since '81, the gold switch in '71, etc.).  But they miss the fact that it also did very well when treasuries were crushed as interest rates skyrocketed in the 70's, when gold lost 80% of its value in the 80's and 90's, and when stocks lost money for more than a decade starting in 2000.  So clearly there's something more fundamental backing the consistent returns than a few great years.



For comparison:



As I mentioned earlier, the Golden Butterfly is built upon the same foundation of the Permanent Portfolio.  So to understand how it works we first need to talk about the PP. 

The basic theory of the Permanent Portfolio is to select assets that do particularly well in the four possible economic environments. Generally speaking:

Prosperity: stocks
Recession: cash
Inflation: gold
Deflation: long term treasuries

As an aside, some people like to argue about how well gold truly tracks inflation.  That's fine, but gold has other benefits as well -- especially how it is negatively correlated to stocks and bonds.  So even if its inflation-fighting properties aren't as direct as some people claim, it's still a valuable component to protect the portfolio when stocks and bonds are both struggling.  Think of the 2000's when stocks went sideways for a decade with low inflation, but gold still responded and did great. 

Harry Browne liked to explain how the PP protects your money from market turmoil by referring to the four very different assets as having "firewalls" between them.  So even if gold does terribly for a while, the damage is isolated only to a small portion of the portfolio.  And because the assets are selected based on economic conditions, when gold is struggling that usually means that at least one other asset is doing great and more than making up the difference.  Importantly, it works the same way when bonds do terribly and when stocks inevitably crash.  This outlook-neutral approach that doesn't try to predict the future but instead protects you and grows your money no matter what the future holds is the core idea that makes it tick.

The GB follows the same philosophy -- invest in volatile uncorrelated assets that cover every economic condition, and you'll do pretty well with limited downside no matter what happens in the markets.  The difference is that it tilts the portfolio slightly towards prosperity, the most common of the four conditions.  And it does so by intentionally selecting an additional stock asset that complements the normal Permanent Portfolio stock index fund for rebalancing purposes and higher overall returns.  The large/small barbell works well for this but other options are also fine.  The permanence of the value premium is also debatable, and as I mentioned I personally punted and deferred to my neutral Harry Browne mindset by using small cap blend. 

Of course any short summary like this will over-simplify certain issues.  If you want to learn more, read the two books listed here.  You also may disagree with some of the theory and methods, and that's fine.  Different portfolios approach investing in different ways. 

[pka222]

Tyler- Great work- I've really enjoyed using and learning from your tools. 
I've got a portfolio currently mostly in TSM (VTSAX) now but will be deploying enough to totally revamp my allocation.
Due to your tools I'm finally seeing the value in gold- I still don't like it much but damn it does have an amazing effect, even at relatively low  proportions of a assets allocation.
I'm now looking at a 5 fund portfolio weighted toward prosperity and value and using gold and LLT for a bit of downside protection.  It back tests well having a max 4 year draw down (-19%) with an average CAGR of 8.2%.
Based on my time line (still accumulating) I can let this portfolio grow for another 10 years before needing more stability.
Cheers

TSM 30%
SCV 20%
EM  15%
LLT 25%
GOld 10% 

 

[Monkey Uncle]

The GB follows the same philosophy -- invest in volatile uncorrelated assets that cover every economic condition, and you'll do pretty well with limited downside no matter what happens in the markets.

That's the whole topic in a nutshell.

Now, off to find some REIT and bond exposure to balance my stock-heavy portfolio...

[Interest Compound]

But they miss the fact that it also did very well when treasuries were crushed as interest rates skyrocketed in the 70's, when gold lost 80% of its value in the 80's and 90's, and when stocks lost money for more than a decade starting in 2000.  So clearly there's something more fundamental backing the consistent returns than a few great years.

Therein lies the critical flaw. There does not necessarily need to be a fundamental backing. This seems to be the thought process, please correct me if I'm wrong:

• Look at all possible portfolio combinations, for all available data.
• Find the ones which outperformed.
• Out of these portfolios, find the ones which seem to have a fundamental backing (a story for why they make sense).

Or maybe it's the opposite:

• Theorize about possible portfolios which you believe have a fundamental backing (a story for why they make sense).
• Look at all possible portfolio combinations, for all available data.
• Did your theorized portfolio outperform? If not, keep theorizing.

It's my suspicion that this is how all slice-n-dice portfolios are created. So, I set out to do the opposite. Let's find a portfolio which makes perfect sense, but drastically UNDERperformed! Then let's find a portfolio which makes perfect sense, backtested well, THEN after that point in time, drastically UNDERperformed! Your Portfolio Finder seems perfect for this, but it only shows me the BEST portfolios given a set of parameters. How can I find the WORST?

While this wouldn't fully get me there, as it won't include:

• All the other assets/asset-classes over the years (which made perfect sense in the moment), but didn't do well, so you don't have data for them.
• The fact that Gold was illegal to own until 1975, and until the advent of gold ETFs in 2005 gold had to be purchased as bullion or coins, which carried very high spreads and storage costs that are never factors into gold returns in backtesting.
• Transaction costs (including taxes) with rebalancing equally weighted assets. Transaction costs may have been prohibitively high for many of these portfolios.

...etc

I think it'll get me close enough.

[Tyler]

It's my suspicion that this is how all slice-n-dice portfolios are created. So, I set out to do the opposite. Let's find a portfolio which makes perfect sense, but drastically UNDERperformed! Then let's find a portfolio which makes perfect sense, backtested well, THEN after that point in time, drastically UNDERperformed!

Interesting idea. 

The Heat Map is a pretty good tool for what you're looking for, although it requires a bit of individual legwork.  Let's say that an investor did a lot of research and decided that putting all of their money in VTSAX makes perfect sense.  They could enter 100% total stock market into the Heat Map and get this:



Using your method, let's put ourselves in the shoes of an investor at the end of 1999 evaluating that portfolio.  Backtesting the stock market to 1972, the inflation adjusted CAGR was a whopping 8%! (Look at the 1972 row, and follow it to the right to year 28).  I'd definitely classify that as backtesting extremely well!  After buying in, the stock market proceeded to immediately dive underwater and not emerge for 13 years.  From 2000 to today, the real CAGR is only 2.3%.  (Look at the 2000 row and follow it all the way to the right).  I'd also classify that as drastically underperforming relative to expectations.

(Please understand I'm not picking on anyone who invests that way nor am I suggesting it's a bad idea.  A 100% TSM portfolio is included on the site because I believe that's a perfectly good portfolio for many investors.  I'm just using it as an example).

Using the same methodology, one thing I like about the Golden Butterfly his how it really doesn't matter what year you pick for the same analysis.  (Although if you're curious, the numbers for the same reference date are 6.5% before and 5.3% after). 



No matter what start year you choose, it's more or less just as blue after that as it was before.  Now that doesn't mean it will always be that way and it can't be taken as a guarantee, but it's still a positive quality and it makes me want to look deeper. 

I do sorta like your idea for automating that kind of start date dependency analysis.  I'll have to think about that.

Transaction costs (including taxes) with rebalancing equally weighted assets. Transaction costs may have been prohibitively high for many of these portfolios.

For reference, the returns do include expense ratios for the underlying low-cost Vanguard funds.  Before those funds existed, previous returns are discounted by the same ER to make the analysis fair.  Using the base assumption that you rebalance once a year, I consider the brokerage trading fees ($8 per trade at Fidelity, for example) to be negligible.  It's absolutely true that taxes are a big consideration, but individual circumstances (income, state, filing status, tax deferred accounts, tax-loss harvesting, etc) are so different that calculating one "after tax" number is mostly worthless.  Depending on their circumstances, one person using just VTSAX with no rebalancing required may still owe a sizable tax bill on dividends, while another with the complex Merriman portfolio may owe nothing.  Also, tracking 44 years of ever-changing tax law in a calculator like this is something that just doesn't sound very fun.  ;)

That said, I completely agree that some portfolios may be better than others tax-wise depending on your situation.  Always make that part of your decision making process.    [/list]

[Mrs. Healthywealth]

Tyler,
Thank you for all the hard work you put into building your website and wanting to help others.  I've been reading your responses to other people, and you're really great about being respectful and patient.

Been staying up late for the past few nights absorbing all this information and reading Portfolio Charts and learning about the other options.  The calculators are great, and are making me consider a little less volatility to my current allocation which is a 3-fund portfolio of 70/10/20 (stock/international/bond). I have a pension which is why i'm so aggressive in my allocation, but going thru your website makes me realize there are other potential options to get me to the same place with less volatility. 

Looking at the heat table for TSM and gold is amazing, given their negative correlation regardless of the issues surrounding gold.  I read so many negative things about gold, but this makes me reconsider it.  I'm looking to FIRE in 8yrs, and I had already planned to tone down the stocks in the portfolio. No intentions to make any quick decisions, and I will be reading some of the books you recommended.

The Golden Butterfly does get me excited, but so does the latest Tesla...will take deep breaths and research before jumping into anything--which I hear you saying.

[Interest Compound]

Unfortunately, the Heat Map won't get me there. Let me explain a different way. Here's what I noted in the other thread, then titled, "The Case Against 100% Stocks (and what to do about it)":

Out of 83,000 total portfolios, about half of them did better than TSM in terms of risk/reward. So that means about half of them worse. Isn't that the point of just buying the whole stock market? I understand you're looking at many asset classes which aren't composed of stocks, but that's the most interesting part. Despite looking at so many different asset classes, that dynamic was still in play! Buy the whole stock market, and you'll get average returns.

Buy the whole stock market, and you'll get average returns. Looking at your charts, there seems to be an invisible Invisible horizontal barrier at the Total Stock Market level. Only 1 of the super-portfolios surpassed it. As is a common theme around here, beating the average seems to be a tough nut to crack:



When you deviate from the average, you could end up as any one of the little dots on this chart. As you didn't comment on your thought process, I'll assume my suspicions are correct, and it's the first one:



Don’t you see? You're making the same mistake everyone else did when creating their slice-n-dice portfolio, they were just using different data. I'm sure at some point all those other portfolios looked just as amazing, and now look at them. Seemingly randomly-distributed dots, almost all comfortably under the invisible line of Average. Perhaps the Golden Butterfly will end up being closer to the Permanent Portfolio in a few years, after-all, they have a similar story.

This is a dangerous game to play, and I believe this is what's being lost on your readers. You can either take your average returns, or you can play the "choose a dot" game. The problem is, there's no way of knowing where the dot will slide, for your specific timeline! What if it moves down here, next to the Permanent Portfolio?



What if it moves further down, close to the Desert Portfolio?



What if it moves further down still?



When you play the "choose a dot" game, you can end up anywhere. While you talk a lot about volatility risk, I believe this risk is underplayed. This doesn't leave much wiggle-room for unexpected expenses in retirement, or sub-par investment returns. While I'm sure you understand this, I don't believe it's coming across clearly. You have enough data to play with this. "If you played this game on a 10 year time frame, what are the chances you’d end up with a bad dot after 10 more years? What about a 15 year time frame? 20? 30?

Based on this, if you decide to play this game on a ~40 year time frame (data from 1972), what are the statistical chances of picking a bad dot? If so, what are the consequences?

It wouldn't be perfect, and there's only so much you can do with limited data, so maybe you can come up with a better idea? ¯\_(ツ)_/¯

In any case, maybe you can give me a list of some of the worst portfolios (not assets) in the chart above? This might go a long way in uncovering why it's believed there *needs* to be a fundamental backing to what seems to be just another ordinary dot.

[Tyler]

If you played this game on a 10 year time frame, what are the chances you’d end up with a bad dot after 10 more years? What about a 15 year time frame? 20? 30?

Well, you're using an old chart and I've already done this.  The Portfolio Finder has since been updated to look at all 15-year rolling timeframes and only report the worst one for each portfolio.  So now all of the dots are shown at their lowest CAGR points from a start-date independent perspective, which helps you evaluate whether the portfolio would have met your needs even in its least flattering light.  Notice how it really flattens the chart.



With that said, I'd like to respect the OP and keep this thread about the Golden Butterfly.  There's another one dedicated to the Portfolio Finder if you still want to discuss that further.

On that note, you seem to misunderstand how the Golden Butterfly was created.  Note that I posted it nearly six months before I created the Portfolio Finder.  I've explained how the concept is based on the Permanent Portfolio, which I believe was first proposed by Harry Browne in the early 80's and has held up very nicely.  Yes I used back testing to help me find the one extra asset, but I don't believe that's unreasonable.  The Portfolio Finder independently verified after the fact that it is a very consistent portfolio compared to almost any other option.  It was not at all the result of pure data mining. 

BTW, your conclusions regarding withdrawal rates are also over-simplified.  A portfolio can have a lower return than the stock market and still have a higher SWR depending on the volatility.  But there are other threads and posts about that, too.

[Interest Compound]

If you played this game on a 10 year time frame, what are the chances you’d end up with a bad dot after 10 more years? What about a 15 year time frame? 20? 30?

Well, you're using an old chart and I've already done this.  The Portfolio Finder has since been updated to look at all 15-year rolling timeframes and only report the worst one for each portfolio.  So now all of the dots are shown at their lowest CAGR points from a start-date independent perspective, which helps you evaluate whether the portfolio would have met your needs in its least flattering light.  Notice how it really flattens the chart.



With that said, I'd like to respect the OP and keep this thread about the Golden Butterfly.  There's another one dedicated to the Portfolio Finder if you still want to discuss that further.

On that note, you seem to misunderstand how the Golden Butterfly was created.  Note that I posted it nearly six months before I created the Portfolio Finder.  I've explained how the concept is based on the Permanent Portfolio, which I believe was first proposed by Harry Browne in 1983 and has held up very nicely.  Yes I used back testing to help me find the one extra asset, but I don't believe that's unreasonable.  The Portfolio Finder independently verified after the fact that it is a very consistent portfolio compared to almost any other option.  It was not at all the result of pure data mining. 

BTW, your conclusions regarding withdrawal rates are also over-simplified.  A portfolio can have a lower return than the stock market and still have a higher SWR depending on the volatility.  But there are other threads and posts about that, too.

The dollar amounts I gave (half the amount saved, and one third the amount saved), from the Permanent Portfolio and the Desert Portfolio were taken directly from modeling those portfolios with a 4% withdrawal rate. It was not a simplification. Since I don't know what the other portfolios consist of (which is what I've been asking for), I picked a few that were identified.

It seems my posts still aren't clear, so I'll be more brief and direct.



This is very much related to The Golden Butterfly.

[Tyler]

The dollar amounts I gave (half the amount saved, and one third the amount saved), from the Permanent Portfolio and the Desert Portfolio were taken directly from modeling those portfolios with a 4% withdrawal rate. It was not a simplification. Since I don't know what the other portfolios consist of (which is what I've been asking for), I picked a few that were identified.

I was only referencing your chart that showed a horizontal line at the TSM and seemed to imply that anything below that is "almost guaranteed portfolio wipe at the 4% rule".  That's not necessarily true, but after re-reading your post I believe I misunderstood your intent to show "what if" the GB went lower than a 4% CAGR.  In any case, I've already provided data for just how low all of the portfolios have gone so there's no need to guess.

It seems my posts still aren't clear, so I'll be more brief and direct.

This is very much related to The Golden Butterfly.

One of the worst examples is a portfolio with 100% gold. Others around that are things like 50% gold and 50% cash.  Most of the ones at the bottom of your bounded box are portfolios with nothing but various bonds.  The closest one marked on your chart is the TSM.  I'll leave it to you whether that sounds as plausible as the Golden Butterfly.

I disagree that this is related to the Golden Butterfly.  If you want to start another thread discussing your opinion on the best way to select a portfolio, I'm sure that would be an interesting topic in its own right. 

[Interest Compound]

The dollar amounts I gave (half the amount saved, and one third the amount saved), from the Permanent Portfolio and the Desert Portfolio were taken directly from modeling those portfolios with a 4% withdrawal rate. It was not a simplification. Since I don't know what the other portfolios consist of (which is what I've been asking for), I picked a few that were identified.

I was only referencing your chart that showed a horizontal line at the TSM and seemed to imply that anything below that is "almost guaranteed portfolio wipe at the 4% rule".  That's not necessarily true, but after re-reading your post I believe I misunderstood your intent to show "what if" the GB went lower than a 4% CAGR.  In any case, I've already provided data for just how low all of the portfolios have gone so there's no need to guess.

It seems my posts still aren't clear, so I'll be more brief and direct.

This is very much related to The Golden Butterfly.

One of the worst examples is a portfolio with 100% gold. Others around that are things like 50% gold and 50% cash.  Most of the ones at the bottom of your bounded box are portfolios with nothing but various bonds.  The closest one marked on your chart is the TSM.  I'll leave it to you whether that sounds as plausible as the Golden Butterfly.

I disagree that this is related to the Golden Butterfly.  If you want to start another thread discussing your opinion on the best way to select a portfolio, I'm sure that would be an interesting topic in its own right.

Agreed, those portfolios don't see to make much sense. While I'm sure there are some that do, I'll just go with what I see:



I'd expect The Golden Butterfly to gravitate back towards the average, just like all the other similar-minded portfolios which were created at a time when they were a similar out-performing dot on the chart. The Permanent Portfolio, on which The Golden Butterfly is based, was created in 1982. From 1972 to 1982 The Permanent Portfolio looked even better on the charts than The Golden Butterfly does today! But had you gone all-in back then, you would've underperformed even 100% bonds.

TL:DR, unless you're comfortable with the very real possibility of getting about 4.5% inflation-adjusted returns for the long term, don't invest in The Golden Butterfly. That's the risk you take when trying to remove volatility from your portfolio. The, "match or beat the total stock market in long-term real CAGR with fewer stocks and lower volatility" description of The Golden Butterfly is not a realistic expectation. Don't bet your life-savings on it, there is no free lunch.

[Tyler]

TL:DR, unless you're comfortable with the very real possibility of getting about 4.5% inflation-adjusted returns for the long term, don't invest in The Golden Butterfly.

The start-date-independent chart makes your point very well.  It also shows that if you expect more than 4.5% real from any portfolio any time soon, history shows there's a good chance you'll be disappointed.  As you said, there's no free lunch.  But if you wait long enough and have both the tolerance and capacity to weather prolonged droughts, then sure -- there are higher earning long-term options than the Golden Butterfly.  Maximizing returns is not its primary purpose. 



I respect your opinion.  Clearly we have different perspectives, and I'm happy to agree to disagree.  Different people like different portfolios, and by reading both sides I'm confident others can come to their own conclusions.  We've said enough -- Let's give them space to do that.  :)

[Interest Compound]

TL:DR, unless you're comfortable with the very real possibility of getting about 4.5% inflation-adjusted returns for the long term, don't invest in The Golden Butterfly.

The start-date-independent chart makes your point very well.  It also shows that if you expect more than 4.5% real from any portfolio any time soon, history shows there's a good chance you'll be disappointed.  As you said, there's no free lunch.  But if you wait long enough and have both the tolerance and capacity to weather prolonged droughts, then sure -- there are probably higher earning long-term options than the Golden Butterfly.  Maximizing returns is not its primary purpose. 



I respect your opinion.  Clearly we have different perspectives, and I'm happy to agree to disagree.  Different people like different portfolios, and by reading both sides I'm confident others can come to their own conclusions.  We've said enough -- Let's give them space to do that.  :)

That's a good chart for someone planning for 15 years of retirement. A 75-80 year old perhaps. Early Retirement is typically much longer, MrMoneyMustache is planning for a 70 year retirement, most of us here are probably planning for 50 years. I don't believe a 15 year chart, spanning 43 actual years of data, is really that useful for people on this forum.

The Minimum Real 50-year CAGR for stocks is about 5%.
The Minimum Real 70-year CAGR for stocks is about 6%.

This moves TSM up to the 5-6% range for an Early Retiree:



The Minimum Real 50 year CAGR for 40/60 stocks/bonds is about 2%
The Minimum Real 70 year CAGR for 40/60 stocks/bonds is about 3%

This is about what I'd expect from The Golden Butterfly. The conservative 40/60 stocks/bonds portfolio has seen more than a few 50 year time periods where it hit a 5% yearly CAGR. But it would be silly to expect anything close to that on average. That's exactly what we're seeing with The Golden Butterfly. To the newbies exploring this portfolio, don't make the mistake of expecting it to continue coming anywhere near the returns of 100% stocks. As Tyler said, maximizing returns is not The Golden Butterfly's primary purpose. It's stability. Hence why I don't think it's wise to believe such a portfolio would stay anywhere near stocks for the long-term.

Especially when all the portfolios like it, indeed even the portfolio it was based on, is exhibiting this behavior. Consider this, the Minimum Real 50 year CAGR for The Permanent Portfolio (substituting Long-Term Bonds with Intermediate Bonds for more data), has a Minimum Real 50 year CAGR of 1%. That's a big shift:



Maybe it looks better with Long-Term Bonds. In any-case, one thing is clear. These portfolios are not an appropriate allocation for someone looking to Retire Early with the 4% rule.

We've said enough indeed, thanks again for the analysis and all the tools Tyler!

[arebelspy]

Awesome discussion.

I got a lot out of it.  Thanks to both of you.

[WerKater]

Awesome discussion.

I got a lot out of it.  Thanks to both of you.

Seconded. I love this forum for discussing this kind of stuff. I just read through this without a break.

[dandypandys]

me too, it sure is interesting.

[josstache]

I think the larger risk you assume with gold is that it is a single asset, and so the entire 20% of your portfolio is subject to certain risks.  I have a similar problem with US treasuries, which make up another 40% of the golden butterfly.

For instance, if you have 60% of your portfolio in US treasuries and gold, what happens if the US government has a fiscal crisis, in response to which it defaults on its debt, goes back onto the gold standard and forces a sale on unfavorable terms of all privately held gold.

60% of your portfolio would have its value significantly impaired (or even reduced to 0) by the actions of a single government.  Obviously stocks and other types of bonds would also have extreme volatility in response to such an event, but it's not clear that they would actually be similarly impaired.

[Radagast]

I think the larger risk you assume with gold is that it is a single asset, and so the entire 20% of your portfolio is subject to certain risks.  I have a similar problem with US treasuries, which make up another 40% of the golden butterfly.

For instance, if you have 60% of your portfolio in US treasuries and gold, what happens if the US government has a fiscal crisis, in response to which it defaults on its debt, goes back onto the gold standard and forces a sale on unfavorable terms of all privately held gold.

60% of your portfolio would have its value significantly impaired (or even reduced to 0) by the actions of a single government.  Obviously stocks and other types of bonds would also have extreme volatility in response to such an event, but it's not clear that they would actually be similarly impaired.

If you strictly followed the PP gold would be kept in New Zealand or somewhere to avoid that. In my opinion Tyler should change that 20% small company to instead be either total international or even emerging markets to help offset issues that affect only the US.

[Seppia]

Terrific discussion, thanks to the great Tyler and IC for the debate.