Things about the PP that do not make sense to me

General Discussion on the Permanent Portfolio Strategy

Moderator: Global Moderator

FF9000
Associate Member
Associate Member
Posts: 25
Joined: Mon Jul 18, 2016 9:26 am

Things about the PP that do not make sense to me

Post by FF9000 » Mon Jul 18, 2016 9:42 am

I've read a lot about the PP and am strongly considering using it for the bulk of my assets. But a few things haven't made sense to me and so I wanted to check in with the PP experts.

  • Why the 25/25/25/25 allocation? I understand the concept - be in a good place to handle any of the four types of economic scenarios. However, it appears that an evenly-distributed allocation assumes an equal probability for all four scenarios. Particularly in the case of deflation, which is a much less common occurrence than inflation, and would imply that cash holdings are overweighted in the portfolio. While Browne's idea that "you cannot predict the future by looking at history" is accurate, it is overly simplistic. You could use the same logic to say "just because the US Government has not defaulted doesn't mean it won't tomorrow". Again true, but not particularly helpful from an allocation perspective. Or "The PP has worked in the past but might not in the future." If we use historical returns, volatilities, correlations, etc. we can be better at predicting than just assuming equal weighting for all scenarios. And so 25/25/25/25, while elegant and probably "good enough", seems to be "leaving money on the table" even at the same level of risk. In a nutshell, 25/25/25/25 seems like an overly simplistic short-cut. The idea of the four asset classes makes rational sense, but surely there is a better way to determine how to maximize returns for a specific level of risk based on asset correlations, returns, and volatility. Has this been assessed by anybody?
  • In terms of projecting out PP growth, the Portfolio Visualizer isn't a bad tool, but it does not use asset correlations when forecasting based "statistical returns". Has anyone here done a stochastic analysis using not only historical returns for each asset class but also using correlations between asset classes (without using historical years for bootstrapping)? I am thinking something like a Principal Component Analysis could accomplish this but I have not gotten around to doing it myself yet.
There will be more questions, but these were the first two I wanted to pose. Thanks.
User avatar
Tyler
Executive Member
Executive Member
Posts: 1875
Joined: Sat Nov 12, 2011 3:23 pm
Contact:

Re: Things about the PP that do not make sense to me

Post by Tyler » Mon Jul 18, 2016 11:28 am

Hi FF9000. Welcome! Good questions.

Why the 4x25 allocation? The short story (from my perspective) is that the PP was designed for "money you can't afford to lose". So the 4x25 is the agnostic allocation that protects your money equally in any economic outcome. If you want to try to tweak the risk-adjusted returns, that's what the variable portfolio is for. Many people here own various VP holdings, but the PP is the solid foundation everything is built on.

Regarding projecting growth into the future, that's always a challenging proposition. I personally don't care for stochastic analyses because the real world is more complicated and interconnected than the mathematical methods can ever hope to account for, and the resulting precision is really deceptive. I prefer to look at history as a whole and appreciate the total range of outcomes. For example, play around with this calculator and change the different asset percentages. Some portfolios are more uncertain than others, and while you'll never be able to predict any future path with certainty you can get a good feeling for the range of possibilities by studying enough history. FWIW, the PP is one of the most steady asset allocations I've found to date.
Mechanical engineer, history buff, treasure manager... totally not Ben Gates
User avatar
dualstow
Executive Member
Executive Member
Posts: 10268
Joined: Wed Oct 27, 2010 10:18 am
Contact:

Re: Things about the PP that do not make sense to me

Post by dualstow » Mon Jul 18, 2016 12:58 pm

FF9000 wrote:I've read a lot about the PP and am strongly considering using it for the bulk of my assets. But a few things haven't made sense to me and so I wanted to check in with the PP experts.

  • Why the 25/25/25/25 allocation? ...
    it appears that an evenly-distributed allocation assumes an equal probability for all four scenarios. Particularly in the case of deflation, which is a much less common occurrence than inflation,.
It's kind of funny to read this in a time when yields are plummeting and long bonds (not cash) are doing their job.

I'm not sure how others feel about the probabilities, but the way I see it, there is a significant enough chance of any of the four scenarios happen. Once you acknowledge that, you want to be sufficiently prepared for each scenario, and you prepare by having enough of the corresponding asset.

Harry acknowledged that that 25% figure is arbitrary, keeping it simple, and it probably wouldn't kill you to knock it down to 20%. But, if you want to reduce the allocation more than that because you think that's in proportion to a probability, maybe don't bother to hold that asset (or the pp) at all.

One of the hardest things is thinking of the pp as a package and not worrying too much about the underlying assets, because at least one looks crazy at any given time, usually more than one.

Continue kicking the tires, though. It's your money, and it's important to understand a portfolio before putting one's hard-earned cash in it, even one that purports to be conservative.
dragoncar
Executive Member
Executive Member
Posts: 1111
Joined: Wed Aug 10, 2011 7:23 pm

Re: Things about the PP that do not make sense to me

Post by dragoncar » Mon Jul 18, 2016 2:12 pm

Others have proposed allocations based on duration or volatility, but I think they are still fairly close to the 4x25. I personally agree the PP is slightly skewed towards outperformance in bad times, but you have to remember that HB considered your salary to be the number one wealth generator. So in good times it's easy to get a job and be paid well - you don't need your portfolio to kick ass. In bad times you may lose your job and that is when you really want portfolio protection
User avatar
Xan
Administrator
Administrator
Posts: 2505
Joined: Tue Mar 13, 2012 1:51 pm

Re: Things about the PP that do not make sense to me

Post by Xan » Mon Jul 18, 2016 2:43 pm

There are also advantages to being a bit counter-cyclical. Yes, it stinks to be lagging a bit when stocks are hot. But then when there's a crash you can do things like buy a nicer house than you'd normally be able to.
User avatar
Kbg
Executive Member
Executive Member
Posts: 1542
Joined: Fri May 23, 2014 4:18 pm

Re: Things about the PP that do not make sense to me

Post by Kbg » Mon Jul 18, 2016 3:14 pm

Welcome FF!

I would take a look at MachineGhosts research resort. He's one of the board's more prolific data torturers. I'm a minor one. I think it's important to understand that Harry Browne philosophically did not believe in the ability to predict the markets which was a primary driver of asset selection and allocation. However, he did classify market types similarly to/the same as all-weather portfolios as you are probably aware. Over time he also migrated toward greater simplicity. Finally, he deliberately chose high volatility assets so that any one asset could carry the load when others were not doing well. I'm personally not a 4x25 purist nor was HB. He was pretty firm about not going below 20% for any one or over 35% however.

So, why the background explanation? If one is going to embark on additional analysis a necessary prerequisite is to determine if they are going to adhere to the same basic philosophies...I.e. market agnostic, minimalist, volatile, the four market types. To me the most important is the first. PP is a passive portfolio. If in their analysis one moves to active techniques then they are comparing the proverbial apples and oranges. The second most important is the four market types ( or another valid classification method ) for asset composition.

After that, I think the field is wide open for investigation.
clacy
Executive Member
Executive Member
Posts: 1128
Joined: Mon Mar 14, 2011 8:16 pm

Re: Things about the PP that do not make sense to me

Post by clacy » Mon Jul 18, 2016 5:28 pm

There is absolutely nothing wrong with (IMO), tweaking the PP if that's what it takes to sleep at night.....

-A portfolio made up of something like this certainly wouldn't be the end of the world:

20%- stocks
20%- gold
20%- treasuries
20%- cash
20%- (pick your asset class... high yield bonds, EM, REIT's, etc)

It's your portfolio and you're the one that has to live with whatever consequences come with it.
User avatar
MachineGhost
Executive Member
Executive Member
Posts: 10054
Joined: Sat Nov 12, 2011 9:31 am

Re: Things about the PP that do not make sense to me

Post by MachineGhost » Mon Jul 18, 2016 8:14 pm

FF9000 wrote: [*] Why the 25/25/25/25 allocation? I understand the concept - be in a good place to handle any of the four types of economic scenarios. However, it appears that an evenly-distributed allocation assumes an equal probability for all four scenarios. Particularly in the case of deflation, which is a much less common occurrence than inflation, and would imply that cash holdings are overweighted in the portfolio. While Browne's idea that "you cannot predict the future by looking at history" is accurate, it is overly simplistic. You could use the same logic to say "just because the US Government has not defaulted doesn't mean it won't tomorrow". Again true, but not particularly helpful from an allocation perspective. Or "The PP has worked in the past but might not in the future." If we use historical returns, volatilities, correlations, etc. we can be better at predicting than just assuming equal weighting for all scenarios. And so 25/25/25/25, while elegant and probably "good enough", seems to be "leaving money on the table" even at the same level of risk. In a nutshell, 25/25/25/25 seems like an overly simplistic short-cut. The idea of the four asset classes makes rational sense, but surely there is a better way to determine how to maximize returns for a specific level of risk based on asset correlations, returns, and volatility. Has this been assessed by anybody?
The PP is a simplified version of the portfolio that is in use in the PRPFX fund. None of the tweaks to account for volatilites and correlations improve it significantly enough to be worth the bother in having to use different rebalancing bands on different weights than 25%. Search the forum; there's been plenty posted about this (mostly by me) over the years.
FF9000 wrote: [*] In terms of projecting out PP growth, the Portfolio Visualizer isn't a bad tool, but it does not use asset correlations when forecasting based "statistical returns". Has anyone here done a stochastic analysis using not only historical returns for each asset class but also using correlations between asset classes (without using historical years for bootstrapping)? I am thinking something like a Principal Component Analysis could accomplish this but I have not gotten around to doing it myself yet.[/list]
It's been done and it winds up being exactly just like the PP. If you want to knock yourself out, be my guest, but you'll arrive right back at the PP. Statistics and correlations are not real world drivers of returns -- they're artifacts.

If you want some fish to fry, then focus on diversifying the Prosperity assets and look at tactical allocation rather than tweaking the strategic allocation.
"All generous minds have a horror of what are commonly called 'Facts'. They are the brute beasts of the intellectual domain." -- Thomas Hobbes

Disclaimer: I am not a broker, dealer, investment advisor, physician, theologian or prophet.  I should not be considered as legally permitted to render such advice!
User avatar
Kbg
Executive Member
Executive Member
Posts: 1542
Joined: Fri May 23, 2014 4:18 pm

Re: Things about the PP that do not make sense to me

Post by Kbg » Mon Jul 18, 2016 9:24 pm

With regard to tactical allocation and the four assets...I've hoed that row quite a bit and I don't find much to commend it either that isn't an artifact of data mining. At the end of the day, I think HB came to the right conclusion...if you have strong feelings/opinions toward an asset class or future market type then just tilt accordingly and live with your decision and associated performance.

Depending on age (younger) I'd definitely tilt some to equities.

In any event, sometimes the board can almost sound psychopantic about HB but if you get dirty with the data you eventually come to: wow, smart guy.

One other place to read: Go to GestaultU and I believe in 2011 or 12 they wrote an excellent series of articles on the PP and various/more advanced asset weighting methods.
User avatar
dualstow
Executive Member
Executive Member
Posts: 10268
Joined: Wed Oct 27, 2010 10:18 am
Contact:

Re: Things about the PP that do not make sense to me

Post by dualstow » Mon Jul 18, 2016 9:34 pm

Hey, sycophants are not always psychos. ;)
Dieter
Executive Member
Executive Member
Posts: 364
Joined: Sat Sep 01, 2012 10:51 am

Re: Things about the PP that do not make sense to me

Post by Dieter » Mon Jul 18, 2016 10:51 pm

The PP is one of the few portfolios that can include your emergency cash as part of the portfolio. (not everyone does, of course.)

If include, retirement accounts may be 30/30/30/10 with emergency funds outside the 401(k) or IRA rounding out the Cash allocation.

Also, as Clacy says, the 5 20's, with Small Cap Value being another choice give more growth opportunity with less downside protection --
40% geared to prosperity, 40% in volatile 'other' (Gold/LTT), and 20% cash.

Conversely, part of the idea is that your human capital is worth the most during prosperity and is the real wealth builder.
FF9000
Associate Member
Associate Member
Posts: 25
Joined: Mon Jul 18, 2016 9:26 am

Re: Things about the PP that do not make sense to me

Post by FF9000 » Tue Jul 19, 2016 10:29 am

Thanks for all the great replies, guys. I really appreciate the helpful and intelligent discussion. I'm excited about joining the community, although honestly I hope I am very inactive since that would mean that my money is safely growing slowly and off my mind. :D

Since yesterday, I have gone on a spree and have built a stochastic VBA model that uses historical block bootstrapping (to preserve asset class correlations in a simple manner). Using this model, I have run various allocation scenarios to see what the probability is that I do not reach $X in Y years. Still tweaking it, but it gives me a lot of confidence in the PP as a way of decreasing the risk of not reaching my goals.

By way of background, I am turning 32 shortly and am aiming to definitely be financially independent by 40. The PP can get me there in 99.3% of (50,000) stochastic scenarios, assuming it performs as it has throughout history (big if, I know). And it can give me a 16% chance of getting there by 38.

Compare to a 40/60 stock/bond portfolio that gives me a 96.6% chance by 40 and 21% chance by 38.

Not a huge difference, but the low standard deviation makes me feel better about it.

I will share the spreadsheet when it is complete.
Post Reply