Page 1 of 1

Permanent Portfolio vs 60/40 1980 to 1998

Posted: Fri Nov 29, 2019 9:33 pm
by vnatale
From looking this table in Craig & Tex's book: "Table 3.6 Multi-year Real Returns for Permanent Portfolio Compared to 60/40 Stock/Bond Portfolio from 1972 to 2011."

It seemed that for every year ( including each corresponding rolling five / 10 year time period) that the Permanent Portfolio return was inferior to a 60 / 40 investment?

Is the simple reason is that gold's run up was over in 1979 and it was the reign of stock started around 1980 and continued through 1998? Thus, an investment underweighted in gold (actually none) and, in comparison, overweighted in equities was going to be superior? I assume, however, that on a risk-adjusted basis the difference in returns would not be as great?

Vinny

Re: Permanent Portfolio vs 60/40 1980 to 1998

Posted: Sat Nov 30, 2019 10:14 am
by Kbg
The simplest answer is more stock...historically more stock equals more return with more volatility. The PP is an excellent hold on to what you have with a some growth portfolio. It is not a good grow what you have portfolio comparatively.

PortfolioCharts makes it very easy to compare various portfolios and their pros/cons.

Re: Permanent Portfolio vs 60/40 1980 to 1998

Posted: Sat Nov 30, 2019 10:34 am
by sophie
Your post got me interested, so I pulled out the book and took a look at Table 3.6.

What you're missing is the preceding discussion in that chapter about the asymmetry of performance for volatile assets. That is, when an asset drops in value by X%, you need a subsequent gain much larger than X% to get back to your starting point. So you have to be careful about looking at year by year figures. Similarly, you have to be careful about just looking at CAGR over the long term, because that doesn't take into account the fact that you had to pull money out of your shrunken portfolio to live on, compounding the asymmetry problem.

It is true that during the stock bull market years that dominated the years 1980-2008 in terms of #years, the rolling 3 year returns of the PP lagged the 60/40 in most years. In other words, it must have been a very discouraging portfolio to hold all that time, when everyone else you know is clocking huge double digit annual gains with their stock-heavy portfolios. Sort of similar to what's going on now in fact!

But, check out the 10 year returns and you see where the magic lies: the PP is consistently returning 3.7% - 6.3% throughout, while the 60/40 goes through periods of near zero or even negative returns. It's those long periods with low yields that can kill your portfolio when you're depending on it for income. Tyler's charts on his portfolio visualizer website show this even more clearly.

Here's how I look at it: On average, the PP will give you a return comparable to the 60/40, but with less downside AND less upside. With the 60/40, you might end up seeing your portfolio take off into the stratosphere. Unfortunately that potential lottery win comes with the risk of running out of money due to a prolonged bad run. With the PP, you have given up your chance of getting those gigantic gains, but in exchange your portfolio is as close to bulletproof as it's possible to get. Which one of these scenarios you prefer is ultimately your choice of course, but the reason this forum exists is that several of us have chosen the safer (and less nail-biting) path.

About the 1972-1973 gold runup: People love to quote that like it's the only consideration in judging the PP. Well, fine if they're looking for an excuse but please keep it to yourself if you're just using it as a shield, is my take on that. They forget that there were reasons in 1972-1973 for gold to shoot up, and also note that the price dropped again afterwards so the PP would have lost most of its artificial gains. What Table 3.6 shows is that there's way more to the PP than just those two years. A 10 year rolling return starting in 2002 can hardly be explained by the gold price in 1973!!!

Re: Permanent Portfolio vs 60/40 1980 to 1998

Posted: Sat Nov 30, 2019 12:15 pm
by barrett
Just adding to what Sophie posted.

Regarding the 1980-1999 period, it emcompasses both the longest bull market in US history from 1982 to 1999 as well as the most challenging period to date for the PP.

Check out this site:

www.peaktotrough.com

Unfortunately the data on there stopped updating a few years back but I believe it is all valid through 2015 or so. peaktotrough.com allowed one to look at monthly or even daily data. Play around with the years 1980 to 1983 and you'll see that the PP was negative in nominal terms from the middle of 1980 to the middle of 1982. You can eyeball inflation data for those years and see that in real terms that period was even worse. Tyler's data over at Portfolio Charts is awesome but it is limited to yearly views as far as I know. The recovery for the PP at the very end of 1982 was so strong that it is necessary to look at a more granular view to see how severely the PP was tested in real terms.

You can see monthly inflation data here:

https://www.usinflationcalculator.com/i ... ion-rates/

Re: Permanent Portfolio vs 60/40 1980 to 1998

Posted: Sat Nov 30, 2019 1:59 pm
by mathjak107
    I show if we compare the pp to 60/40 like vanguard vbinx and use this website we see the following results , however if you notice he pulled the cash position out , so in reality model to model the pp is not going to be like the 60/40 unless you cannibalizes it by removing a needed component .

    Once you start pulling out required components you are no longer comparing the models for what they are ..

    a 60/40 user may decide to hold no cash ,, or in the case of the pp I hold all my spending cash outside the pp and the cash required by the pp is held in addition since the model calls for a 25% allocation .so the only real way to compare is to compare with the actual required components .

    Any cash not an actual component can not be compared since there may be not much held or in our case I have a lot of cash both for spending and eventually deploying somewhere ..but it certainly has no bearing on the model portfolios I use unless cash is a required position if I try to compare models .

    It is no different then Tyler’s work comparing portfolios for what they are ...he does not add cash to models that don’t use permanent cash positions nor does he omit The cash positions in the pp or butterfly..it would be a silly comparison so unless this guy on the website really is not tracking the pp nor are the comparisons he made valid ...wherever he wants to call his version it certainly is not the pp as designed .

    Looking at the cannibilized version

    No cash included according to the website
    Ytd Pp= 19.4 Vbinx 19.7

    1 year pp 20.1 Vbinx 14.4

    3 year pp 9% Vbinx 10.1

    5 year pp 6.5. Vbinx 7.6

    10 year pp 7.50. Vbinx 9.50

    15 year pp 8.4 Vbinx 7.3

    http://www.myplaniq.com/LTISystem/jsp/p ... n?ID=38351

    Re: Permanent Portfolio vs 60/40 1980 to 1998

    Posted: Sat Nov 30, 2019 5:31 pm
    by vnatale
    Thank you for each of your responses.

    I have quickly read them once but will coming back to each of them for rereading.

    Thanks again.

    Vinny