Canadian Money Forum banner

Permanent portfolio and asset allocation

312K views 644 replies 56 participants last post by  Covariance 
#1 ·
I thought I'd start a new thread because it looks like we had hijacked another thread but had some material that would be good to discuss in more detail.

I've always been interested in diversification, and had wandered on my own towards a strategy that turned out to be a lot like Browne's Permanent Portfolio which is equal weights of stocks, government bonds, gold, cash. I'm also interested in reading about other diversification approaches.

Here's some data I have for a permanent portfolio, Jan 2006 to today (10+ years) using 25% XIC, 25% XGB, 25% MNT, 25% XSB. Some of the numbers are imperfect because data wasn't available for the full range.

Performance for the full 10+ years: 5.8%, worst year -2.9%
Benchmark: 50/50 XIU and XBB: 4.3%, worst year -12.6%
Annual returns:
2006: 11.5%
2007: 6.6%
2008: 2.9%
2009: 11.7%
2010: 12.3%
2011: 4.4%
2012: 3.8%
2013: -2.9%
2014: 7.5%
2015: 0.8%
2016: 3.0%

This looks pretty appealing to me. Not only is the long term performance greater, but the volatility and risk is less.

Others had the following comments to this:

I would highly recommend you read Global Asset Allocation by Meb Faber. It's pretty cheap as an e-book and he regularly has promotions on his website where you can get it for free or even cheaper.

He compares various 'famous' portfolios with different asset allocations including Browne's permanent portfolio. From what I recall, the permanent portfolio is actually the worst performing of the bunch but the ultimate conclusion is that they all beat the index over long periods and many end up actually having similar exposure once you simplify the asset classes.

Why compare what is essentially a balanced portfolio containing mix of fixed income and equity with XIU (an all equity etf)?? The comparison should be with a balanced fund or etf.

10 yrs is not really a long enough period to use to make a decision. But sometimes longer term data may be harder to find.

I don't think there is any magic allocation that will work for all times. Those are just for those who do not have time or inclination to adjust portfolios for changing times.

"I don't think there is any magic allocation that will work for all times."

Yup - depending upon the timeframe you pick, to measure that allocation performance, you could look like a hero or a goat.

You need to compare apples to apples as much as possible. I'm not a huge fan of Browne's Permanent Portfolio but that's because my approach is different. Doesn't make it right or wrong, just is :)
 
See less See more
#2 ·
I should add that in the permanent portfolio, it's perfectly feasible to have US exposure within that 25% stock allocation too and realistically you'd probably do 12.5% TSX, 12.5% S&P 500. To simplify the comparison I just used the Canadian stock index in this comparison.
 
#3 ·
If holding entirely in registered accounts, I'd go 100% global for the stock allocation. Why limit yourself to North America?

In non-registered there's a tax advantage to Canadian dividends so I would probably go a little heavier on Canadian in that situation.
 
#6 ·
Yes, I am worried james :)

Kidding aside, in terms of diversification I believe you can own the top sector holdings across our Canadian market and avoid using the Canadian index as part of your portfolio. I just wrote about this actually since I've been thinking about this stuff of late....
http://www.myownadvisor.ca/canadian-stocks-to-buy-and-hold/

Then, after you own your "top-stocks" you supplement the Canadian portfolio using VTI and VXUS in your RRSP.

No bonds here but that's because I have a small pension and I consider that a big bond. So, my asset allocation is 100% stocks.
 
#7 ·
My Own Advisor, I understand how one can optimize or geographically diversify the stock part of the holdings, but I'm curious why you don't like the idea of the broader diversification outside of stocks?

You said earlier,
depending upon the timeframe you pick, to measure that allocation performance, you could look like a hero or a goat.
But analyses on diversified portfolios like the Permanent Portfolio have gone back many decades, on US data, it definitely shows strong performance with undeniably less volatility and milder down years. Isn't that very compelling? It covers a number of periods including high inflation/low inflation, high interest rates, low interest rates, and the permanent portfolio had steady performance and low volatility throughout all of this.

Pure stock allocation just can't do that. Bonds & stocks together do it to some degree -- and I often endorse balanced funds -- but why not add more diversification since it reduces volatility further?
 
#13 ·
My Own Advisor, I understand how one can optimize or geographically diversify the stock part of the holdings, but I'm curious why you don't like the idea of the broader diversification outside of stocks?

You said earlier,


But analyses on diversified portfolios like the Permanent Portfolio have gone back many decades, on US data, it definitely shows strong performance with undeniably less volatility and milder down years. Isn't that very compelling? It covers a number of periods including high inflation/low inflation, high interest rates, low interest rates, and the permanent portfolio had steady performance and low volatility throughout all of this.

Pure stock allocation just can't do that. Bonds & stocks together do it to some degree -- and I often endorse balanced funds -- but why not add more diversification since it reduces volatility further?
I don't hold bonds because while bonds cushion the blow from bad equity markets, my investing time horizon is 10+ years. I feel over that time, equities will far outperform bonds. I could be wrong.

I also feel bond yields have nowhere to go but up over time, which means of course, prices will fall over time. There is little capital appreciation to be had in the decades to come from bonds. I could be wrong.

Lastly, I don't hold any bonds because I have a workplace pension plan (not gold-plated) and I consider that a big bond. This bond will provide me with, hopefully, a decent amount of fixed income in retirement.

For those key reasons, I diversify my equities for sure, but I don't hold any fixed income or bonds. I do hold some cash for as my emergency fund and we some cash for savings (i.e., house improvements, trips, etc.)

Volatility is a short-term headache.
 
#8 · (Edited)
... a strategy that turned out to be a lot like Browne's Permanent Portfolio which is equal weights of stocks, government bonds, gold, cash.

Not only is the long term performance greater, but the volatility and risk is less.
Dan and Justin did an analysis of the PP vs the GCP here. Here's a quick summary:

December 1979 to August 2011:

Global Couch Potato Portfolio:
Annualized Return: 10.13%
Annualized Standard Deviation: 8.83%
Sharpe Ratio: 0.40

Permanent Portfolio:
Annualized Return: 8.81%
Annualized Standard Deviation: 6.73%
Sharpe Ratio: 0.32

GCP has a better return but with a greater ASD. I can't think of a single reason why someone in the accumulation phase, years out from retirement, would put 25% of their portfolio in cash. It's sure to get killed vs inflation. In order to get market returns you need to accept market risk, not cower in cash. My two cents, anyway.

Edit to add: Why are you only looking at Canadian funds when Canada is ~8% of the world's markets. Too much home bias IMO.
 
#187 · (Edited)
And even more direct proof of equivalence for Canada. I'm using just XIC for the stock component but you'd obviously put a mix of stock indices into a real one.

https://www.portfoliovisualizer.com...ocation4_2=25&symbol5=XBB.TO&allocation5_2=50

This visualization shows two portfolios: first, the "ideal" PP broken into 4 pieces including ZFL (long term bonds) and XSB (short term bonds). This is compared to my simplified PP which groups the bonds into plain old XBB.

The overall performance, and year to year behaviour, is virtually identical. Combined with the long term US backtest above I think this proves that the PP can be easily implemented using a single 50% XBB component.

This also addresses one of the biggest criticisms raised about the PP, which was mentioned back on page 1 of this thread

I can't think of a single reason why someone in the accumulation phase, years out from retirement, would put 25% of their portfolio in cash. It's sure to get killed vs inflation. In order to get market returns you need to accept market risk, not cower in cash.
Solved :)

Now you can invest in the PP without any cash component, and we've shown it still matches the original/literal PP. The big trick here... which I'm really proud of ... is simultaneously solving the reluctance to hold cash and the reluctance to hold the long term treasuries.

It took me 3 years to "solve" the above problem and I don't think many others have figured this out.

PP =
25% in XIC / XAW / etc
50% in XBB or VAB, or maybe XGB
25% in MNT or CGL.C​
 
#9 ·
Thanks, that's a neat comparison with the Global Couch Potato Portfolio.

The comparison gives volatility measures but stated that way, it doesn't make intuitive sense to people. To get a better sense look at rough periods and max declines. For example in 2008, the GCP fell -20% vs +3% for the PP. That's a huge difference and I think it shows the value of the PP.

I agree that GCP gave higher returns than PP (an advantage of 1.32% per year in the very long term). We all talk about how you should just accept market risk and sit through volatile periods to achieve the best returns in the long term -- that's the theory. In practice, volatile periods cause a lot of difficulty.

When you see your life savings drop by 20% in one year, it's scary. You start to ask legitimate questions like, am I really willing to experience another 20% drop like that? What if this is the start of a prolonged bear market and it's ten years before the value recovers? What if you lose your job during a depression/bear market... e.g. 2008... and suddenly need to tap into the cash. There are certainly people who post on these forums who had to abandon the stock investments for various reasons. Job loss, illness, emergency, etc.

PP reduces the volatility substantially and eliminates many of these concerns. I guess it becomes a question of how much performance you're willing to sacrifice for that.
 
#11 · (Edited)
In their comparison, Dan and Justin used the 30 day t-bills for the cash component which is as literal as you can get with cash.

Personally I'd hold something like XSB as a short term bond fund (Americans use SHY which is similar to XSB). This is a common adjustment to the permanent portfolio, and you'd get some increase in performance. Quite a bit actually. XSB since inception returned 4.5%

Assuming that this XSB performance since inception is indicative, you gain another 1.1% performance in the permanent portfolio. Which, surprise surprise, now makes the Permanent Portfolio have equal performance to the Global Couch Potato portfolio. (let's call that the optimistic result)

I really don't think the PP performance is bad at all, and the XSB substitution is a pretty reasonable implementation of it. I'm not convinced that you sacrifice much performance with PP.

Even if I'm wrong on the XSB performance projected back to 1979, let's say that 1-3 year bonds definitely yield around 100 basis points more than 30 day bills (pessimistically) which closes the gap in the comparison by 0.25%. In any case, the message is that the performance gap is narrower than the 1.32% figure in their analysis and the long term annual PP performance is probably within 1% of the 60/40 couch potato.
 
#16 ·
CPA, you might call it sad, but short term bonds (XSB) have almost outperformed the TSX in the last decade. So if you think bond holdings are sad, then I'm sure you'll agree that stock holdings are even sadder.

The point really is that you don't know which asset will perform. If we enter a period of negative yields, and increasingly negative yields, then the bonds will continue to perform well. Nobody knows.
 
#17 ·
I think folks who are into their 60s, 70s and 80s probably want some form of fixed income, but that depends, it depends on their risk tolerance and estate planning. Bonds are not for everyone.

Nobody knows the future but if I was a betting man, and I am with my portfolio apparently, the long-term returns of stocks will outpace bonds, and bonds in turn should outpace idle cash. Again, I could be wrong!

If an investor feels better owning cash, owning more bonds than stocks - that's fine. Every investor is different. The perfect portfolio only exists in hindsight.
 
#18 ·
Nobody knows the future but if I was a betting man, and I am with my portfolio apparently, the long-term returns of stocks will outpace bonds, and bonds in turn should outpace idle cash. Again, I could be wrong.
That has been my thinking. I have not owned anything fixed income in over a quarter century. With my wife due to retire on a teacher's pension soonish, my concession to FI will be to wind down the leverage a bit.

hboy43
 
#19 ·
It will be interesting to see the permanent portfolio reaction to what looks like insane market volatility brewing right now. I've made a Canadian permanent portfolio index computed from equal parts (MNT,HBB,HXT,cash) so I can track it over time. I wouldn't actually invest using those, but they're useful to compute total returns.

My P.P index was 100 on 2016-03-22
June 23 at the close it was 101.240

The theory is that this construction with uncorrelated assets will dampen volatility.
 
#20 · (Edited)
Yup this was great on a volatile day! June 24 (on the Brexit outcome and global market turmoil), my P.P closed up 102.42 or +1.2% today.

But really the question is how it does over much longer time frames. And from what I can see, the answer is: it does quite well over the decades.

I think GICs can be used as the "cash" allocation of the permanent portfolio. The reasoning is that for diversification, the cash allocation must be something that can't decline. GICs can't decline. Also, with a ladder where amounts mature ever 6 or 12 months, you have liquidity. My current plan is to follow PP using 5 year GICs as the cash component.

Looking at Dan and Justin's 32 year comparison, substituting GICs for t-bills should completely bridge the gap. The resulting long-term performance would be the same for Permanent Portfolio and Global Couch Potato Portfolio.
 
#22 ·
Permanent portfolio was first thing I learned in investing, definitely a good allocation. Though I would modify it to 50% stocks, 25% gold, and 25% cash/bonds. Not in a rush to setup that allocation just yet with poor interest rates and metals not going anywhere for a few years, but something to work towards in the future.
 
#31 ·
Not in a rush to setup that allocation just yet with poor interest rates and metals not going anywhere for a few years
I challenge this assertion :) In the last 15 years, gold's performance is approx 275% cumulative (using CEF.A as approximator). Compare that to XIU's total return of 165%.

Wear the "permanent portfolio hat" for a second. The point of it was to gain exposure to assets that thrive under different economic conditions. Looking at that 15 year performance, there is a good chance we currently are in the environment where gold really thrives.

As low as interest rates are, there is also the possibility that we're heaving for a negative interest rate environment. Many European countries already have negative yields, and the US treasuries just hit all time historic low yields yesterday.

Looking at today's market conditions, I like the PP a lot. All of these are plausible -- stock bull market, gold bull market, treasuries bull market.
 
#23 ·
I think modifications are fine ... fundamentally it's the exposure to all 4 classes that I like. Basically I think PP is one of the only schemes that actually handles the matrix of all possible bull/bear scenarios -- e.g. our traditional balanced fund / buy and hold is really great when either stocks or bonds are in a bull, and is terrific when both stocks & bonds are a bull (as they've been since 1981), but falls on its face when stocks & bonds are both in bear markets.

By the way the P.P has done just amazingly in June despite the market turmoil. You can't even see the Brexit in the daily data, look how it smooths volatility in June:

2016-06-01 100.64
2016-06-02 100.91
2016-06-03 101.61
2016-06-08 101.47
2016-06-09 101.51
2016-06-10 101.36
2016-06-13 101.72
2016-06-14 101.61
2016-06-15 101.92
2016-06-16 101.69
2016-06-17 101.90
2016-06-20 101.72
2016-06-21 101.18
2016-06-22 101.18
2016-06-23 101.24
2016-06-24 102.42
2016-06-27 102.63
2016-06-28 102.44
2016-06-29 102.74
2016-06-30 102.81
 
#24 ·
Harry Browne was a Libertarian and the original intent of the Permanent Portfolio was to protect your money, not to grow it. This should be the number one criteria for choosing this portfolio. The growth that does occur is a by-product of the volatility of the four assets.

Modifications are fine but keep in mind, they will change the purpose of the PP. Weighting towards equities means it behaves more like a stock portfolio (bigger gains during bulls, bigger drawdowns during bears and crashes, longer recovery times.)

- Index funds were chosen to capture the returns of the market.
- Long term Federal government bonds were chosen for the magnified volatility and protection from defaults.
- Gold was chosen because it's the 2nd most popular currency in the world. When the USD has trouble, people pile into gold the most.
- One-year duration Treasury Bills were chosen for zero volatility, to keep up with inflation, and offer complete default protection. If the federal government ever defaults, Treasury bills are the first thing that's paid off. They also offer liquidity for emergency funds or buying opportunities.

These are the ETFs I chose for my own portfolio. I believe they are the best choices available in Canada for now.
- VCN / XAW
- ZFL
- MNT
- cash (no T-Bill ETF exists in Canada. I will hold pure cash until interest rates rise, or someone creates this fund.)
 
#28 ·
True, government default is very unlikely. I have difficulty justifying that scenario with my portfolio. Harry Browne was very distrustful of government and those with lots of power-- and rightly so. History is filled with events that no one thought could happen, but did. The point of the PP is to protect your worth under any scenario, even the collapse of a country. Keeping 25% of your wealth offshore in the form of physical gold is one of the ways he advocated. It may seem extreme today, but such things have happened before.
 
#30 · (Edited)
You're right, Harry Browne was a libertarian, but I see this portfolio construction as being appealing due to the uncorrelated assets that have different characteristics in different economic environments. I like JohnnyFactor's ETF picks.

For the cash component, I agree that t-bills are ideal and the safest, but I think some alternatives are OK within the spirit of PP. (For end-of-the-world resilience though, I agree you'd want to hold physical gold and either t-bills or physical cash).

Johnny, consider that pure "cash", if you're holding it in a bank account, is only as safe as CDIC insurance. So if you're holding it at a bank you might as well use GICs with CDIC insurance. Of course t-bills are safer and more liquid. I feel that GICs are an acceptable tradeoff as I don't seek liquidity in the PP; what I want is to ensure that the cash bucket can't possibly decline.

I suggested XSB because it is government-heavy, has high credit quality and demonstrated even at the worst of 2008 that it was extremely durable. Unlike corporate-heavy short term bond funds, XSB never even hiccuped at the worst of the crisis. In fact XSB's behaviour was identical to an actual US t-bill fund, SHY, which is often used in American PP. Take a look at the charts from 2008,

BSV, a US corp short-term bond fund that semi-crashed (don't want this in PP)
SHY, a US t-bill fund (used in American PP)
XSB, which I think is good enough for PP

Other alternatives for cash

BMO's ZFS appears to be the safest of any option, with 100% federal paper, short term maturity. JohnnyFactor, this is probably the best option for a Canadian t-bill fund.

iShares CLF is all federal & provincial, but lower credit grade than ZFS due to provincial exposure.
 
Top