Dynamic Allocation based on the Intelligent Investor

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
CFK
Posts: 13
Joined: Fri Oct 06, 2017 11:06 am

Dynamic Allocation based on the Intelligent Investor

Post by CFK » Sun Feb 18, 2018 8:39 pm

After reading the Intelligent Investor, I am thinking about switching from a set allocation of 80% stocks, 20% bonds, to a more dynamic allocation that would change slightly based on interest rates and the Schiller PE. I am thinking of something like the following:

1. If the Schiller PE is less than 17, I'll adjust the ratio to 85/15.

2. If the Schiller PE is between 17 and 23, then the allocation would be 80/20.

3. If the Schiller PE is between 23 and 30, then the allocation would be 75/25.

4. If the Schiller PE is between 30 and 35, then the allocation would be 70/30.

5. If the Schiller PE is above 35, then the allocation would be 65/35.

When the ten year treasury yield is greater than the inverted Schiller PE, then I'd add an additional 5% in bonds.

I don't want to be more aggressive than 85/15, or more conservative than 60/40.

This has the benefit of being mechanical, and easy to follow, much like a set allocation. It also presses in the right direction, at least, of not chasing returns. And it circumscribes the risk of a dot.com or Japan 1989 style meltdown. Further, if a bear market doesn't occur then I'll be fine - I'm still substantially in stocks. And if a bear market does occur, then I'll have the benefit of some additional money to buy stocks as their prices collapse, instead of just being along for the ride.

Primary expected criticism: It's market timing. I disagree, at least based on the definition of speculation that Graham uses in the Intelligent Investor. Nothing in this is based on any anticipation about market moves in the short or intermediate term.

Thoughts?

AlohaJoe
Posts: 3199
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Dynamic Allocation based on the Intelligent Investor

Post by AlohaJoe » Sun Feb 18, 2018 8:52 pm

CFK wrote:
Sun Feb 18, 2018 8:39 pm
And it circumscribes the risk of a dot.com or Japan 1989 style meltdown.
This is the Shiller PE in Japan over the past ~30 years
Image
Notice how it went above 40 in 1984 -- many years before the crash.
Also notice how even after the biggest stock market crash in world history, it never went below 30 other than a few brief moments.
Now, three decades later, it has slightly dropped below 30. (It is around 28 right now, I think.)
According to the numbers you picked, you'd be at 70/30 even after the massive crash and you'd have switched to 72/25 in Japanese equities a few years ago.
Similarly, if you had come up with your strategy in 1980, when US historical CAPE looked like this
Image
You would have picked very different numbers. Instead of "anything above 35 is very dangerous" you'd probably say "anything above 20 is very dangerous".
I don't know about you but when I see CAPE10 numbers like in your original post, it is hard not to feel that they are overly fitted to recent US history and that makes me think they are very fragile with a high likelihood of being wrong about the future.

CFK
Posts: 13
Joined: Fri Oct 06, 2017 11:06 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by CFK » Sun Feb 18, 2018 9:49 pm

I don't know about you but when I see CAPE10 numbers like in your original post, it is hard not to feel that they are overly fitted to recent US history and that makes me think they are very fragile with a high likelihood of being wrong about the future.
I hear you. This is colored by recency, and so might be wrong. But that's always a risk with investing. My thinking was that the only time the Schiller PE has gone under 17 since the early 90s, was during recessions. So I'm worried that if I lower the PE ratio thresholds, I run a risk of being way too conservative.

The alternative is just to hold a constant allocation, which doesn't seem like an improvement - it just ignores valuations and bond yields altogether. Not claiming that this is perfect, only that it's a reasonable alternative to holding a constant allocation and just going along for the ride.

taguscove
Posts: 135
Joined: Sun Dec 11, 2016 6:49 pm

Re: Dynamic Allocation based on the Intelligent Investor

Post by taguscove » Sun Feb 18, 2018 11:14 pm

[edited to remove content]

pkcrafter
Posts: 12646
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Dynamic Allocation based on the Intelligent Investor

Post by pkcrafter » Mon Feb 19, 2018 11:59 am

CFK:
Nothing in this is based on any anticipation about market moves in the short or intermediate term.

No?

I don't really see much wrong with your strategy since your AA range is not extreme. Maybe just one small suggestion. Make your move on what the Shiller P/E will be, then you will be in vogue, part of the avant-garde. :wink:

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

randomguy
Posts: 5552
Joined: Wed Sep 17, 2014 9:00 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by randomguy » Mon Feb 19, 2018 1:36 pm

AlohaJoe wrote:
Sun Feb 18, 2018 8:52 pm
CFK wrote:
Sun Feb 18, 2018 8:39 pm
And it circumscribes the risk of a dot.com or Japan 1989 style meltdown.
This is the Shiller PE in Japan over the past ~30 years
You can't really compare PE10s accross countries. 20 in the US isn't the same as 20 in Japan given the different accounting rules. There is also some bias where some markets will have higher PE10s and those tend to be sticky.

But yeah the historical numbers are different. But they way earnings is calculated is also different. Discount the current numbers by 20-25% to account for that and much of the difference goes away. The current number for the past 18 years are still high (The average PE10 would be something like 15 and we basically would have been below average for 1 year out of the past 20+) but not quite as crazy high.

At a implementation level, you would want some bands so that you don't yo-yo around based on flucations. If the market alternated between 22 and 24, it makes no sense to be buying and selling. And of course you have to decide if you are doing the reallocations on a monthly, quarterly or yearly basis. I also would hesitate to pay taxes for minor tweaks (i.e. going from 80/20 to 65/35 is noticeable. Going to 75/25 really isn't) if you can't do rebalancing in tax advantaged space.

There are some momentum issues where you can be buying on the way down (especially with monthly rebalancing) and emotionally that could be hard. Think about what it would have felt like in March of 2009. You just went 85/15 and over 3 months you watched your investments plummet 20%. But if you execute, you will be fine. If PE10s stay at 30+ for the next 50 years, you will be holding a 65/35 AA. That is pretty sane for anyone. If they drop, odds are you will not be worse off holding 85/15 after PE10s hit 17 than just holding 80/20.

This type of market timing is unlikely to help or hurt much. The scary timing is selling on feelings (i.e. market is going to crash so I am going 100% cash. market is roaring time to go 100% stocks) and going to extremes (holding 80/20 is fine for a lot of people. So is 65/35. Holding say 20/80 ot 100/0 are allocations that apply to fewer people).

magneto
Posts: 934
Joined: Sun Dec 19, 2010 10:57 am
Location: On Chesil Beach

Re: Dynamic Allocation based on the Intelligent Investor

Post by magneto » Mon Feb 19, 2018 2:13 pm

CFK wrote:
Sun Feb 18, 2018 8:39 pm
After reading the Intelligent Investor, I am thinking about switching from a set allocation of 80% stocks, 20% bonds, to a more dynamic allocation that would change slightly based on interest rates and the Schiller PE. I am thinking of something like the following:
1. If the Schiller PE is less than 17, I'll adjust the ratio to 85/15.
2. If the Schiller PE is between 17 and 23, then the allocation would be 80/20.
3. If the Schiller PE is between 23 and 30, then the allocation would be 75/25.
4. If the Schiller PE is between 30 and 35, then the allocation would be 70/30.
5. If the Schiller PE is above 35, then the allocation would be 65/35.
When the ten year treasury yield is greater than the inverted Schiller PE, then I'd add an additional 5% in bonds.
I don't want to be more aggressive than 85/15, or more conservative than 60/40.
This has the benefit of being mechanical, and easy to follow,
Primary expected criticism: It's market timing. I disagree, at least based on the definition of speculation that Graham uses in the Intelligent Investor. Nothing in this is based on any anticipation about market moves in the short or intermediate term.
Thoughts?
The bands maybe a little chunky in operation, but seem generally reasonable to this investor.
Chosen CAPEs make sense, noting the median anchor gets dragged from decade to decade.
Used a similar Variable Ratio over last quarter century plus, using real yields of competing Asset Classes, and quite content with the results.
Some thought might be given as to how targets are approached. In what increments and over what period of time?
Agree that since neither immediate anticipation nor time are included in the calcs, only the most dogmatic of Constant Ratio Investment Formula followers might insist on classifying as 'Market Timing'.

P.S. Schiller note is spelt Shiller !!!
To avoid confusion.
"Schiller was a German poet, philosopher, physician, historian, and playwright" wikipedia
'There is a tide in the affairs of men ...', Brutus (Market Timer)

CFK
Posts: 13
Joined: Fri Oct 06, 2017 11:06 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by CFK » Mon Feb 19, 2018 3:48 pm

At a implementation level, you would want some bands so that you don't yo-yo around based on flucations. If the market alternated between 22 and 24, it makes no sense to be buying and selling. And of course you have to decide if you are doing the reallocations on a monthly, quarterly or yearly basis. I also would hesitate to pay taxes for minor tweaks (i.e. going from 80/20 to 65/35 is noticeable. Going to 75/25 really isn't) if you can't do rebalancing in tax advantaged space.
My thinking is I'd implement this mostly through new money - which is feasible in my situation since I've only been investing about 2 years, and save a decent chunk each month. So, for example, I'm currently at somewhere around 75/25, but based on my plan I should be between 70/30 and 65/35. So when I make monthly contributions I'd contribute maybe 50/50 or 40/60 - still buying stocks every month, but moving the needle in the right direction. Every six months I might move money around in tax advantaged accounts to get closer to the allocation target.

P.S. Schiller note is spelt Shiller !!!
To avoid confusion.
"Schiller was a German poet, philosopher, physician, historian, and playwright" wikipedia
:oops: thanks for the heads up!

staythecourse
Posts: 5640
Joined: Mon Jan 03, 2011 9:40 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by staythecourse » Mon Feb 19, 2018 3:57 pm

I ran the original concept of Mr. Graham of % stocks vs. bonds based on earning yield and treasury note (if I remember that correct). I crunched all the data and it was no better then a 50/50 portfolio from 1972- current (done around 2012-13 or so).

Do as you want as it is your money, but whatever you are doing IS market timing as you are making guesses on valuations. Simple as that. You may be correct or you may be completely wrong. You may make a ton of money, but you may end up broke. If you have that fallback inheritance go for it, but if you have one shot at retirement I wouldn't take a chance of getting something so wrong.

The thing is I am surprised at is HOW MANY folks will come up with some random simple valuation metric and think it will do better then a buy/ hold technique when there is a TON of data on buy/ hold both in efficiency and effectiveness and NO DATA on valuations on an ex ante basis. I just don't see how any rational individual wouldn't come up with what you are thinking of doing and then have a light bulb turn on above their head where they go, "Hey if this is so simple and there are MILLIONS of ivy league grads and MILLIONS of computers running all sorts of programs and computations for the last 20-40+ years how could this not have been found already?" This just another case of the professor and the student walking down the street seeing the $20 bill on the street.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

User avatar
randomizer
Posts: 1249
Joined: Sun Jul 06, 2014 3:46 pm

Re: Dynamic Allocation based on the Intelligent Investor

Post by randomizer » Mon Feb 19, 2018 3:58 pm

CFK wrote:
Sun Feb 18, 2018 9:49 pm
My thinking was that the only time the Schiller PE has gone under 17 since the early 90s, was during recessions. So I'm worried that if I lower the PE ratio thresholds, I run a risk of being way too conservative.
"since the early 90s" isn't really anything at all in terms of market history. I wouldn't want to make any big decisions based on what has happened to happen over only 2 or 3 decades. (But yes, I think you run the risk of being too conservative. Unless of course, you need to be that conservative in order to sleep at night and not panic and make behavioral mistakes.)
75:25 AA / Expected retirement: 2097

asif408
Posts: 1251
Joined: Sun Mar 02, 2014 8:34 am
Location: Florida

Re: Dynamic Allocation based on the Intelligent Investor

Post by asif408 » Mon Feb 19, 2018 4:21 pm

staythecourse wrote:
Mon Feb 19, 2018 3:57 pm
The thing is I am surprised at is HOW MANY folks will come up with some random simple valuation metric and think it will do better then a buy/ hold technique when there is a TON of data on buy/ hold both in efficiency and effectiveness and NO DATA on valuations on an ex ante basis. I just don't see how any rational individual wouldn't come up with what you are thinking of doing and then have a light bulb turn on above their head where they go, "Hey if this is so simple and there are MILLIONS of ivy league grads and MILLIONS of computers running all sorts of programs and computations for the last 20-40+ years how could this not have been found already?" This just another case of the professor and the student walking down the street seeing the $20 bill on the street.
I won't necessarily knock a methodology that takes places over many years. As many of the professional investors will tell you, look out beyond 2 years and the competition thins out considerably. At least from the old timers in the industry I've followed every indication is that there is a much greater focus on short term performance nowadays than in the past, and the definition of short-term has dropped considerably (to the points of months, days, and minutes). The competition is admittedly high in the short term, and any amateur who thinks they can compete with the folks in this cage is probably delusional. I surely wouldn't jump in that cage.

I have no doubt that many, if not most of the pros are aware of things like CAPE, it's just that most of them don't have a long enough time frame to take advantage of it. Do you really think the big institutional and hedge fund managers wouldn't get fired or replaced if they used a strategy like CAPE and it didn't pan out in a year or two or they underperformed considerably? And how many individual investors would stick with a strategy that has years of under or lagging performance with the promise of longer-term performance?

The OP may or may not do better, but his method appears slow, methodical, and will take place over many years, so it's unlikely he will dramatically underperform or outperform the market.

randomguy
Posts: 5552
Joined: Wed Sep 17, 2014 9:00 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by randomguy » Mon Feb 19, 2018 4:42 pm

randomizer wrote:
Mon Feb 19, 2018 3:58 pm
CFK wrote:
Sun Feb 18, 2018 9:49 pm
My thinking was that the only time the Schiller PE has gone under 17 since the early 90s, was during recessions. So I'm worried that if I lower the PE ratio thresholds, I run a risk of being way too conservative.
"since the early 90s" isn't really anything at all in terms of market history. I wouldn't want to make any big decisions based on what has happened to happen over only 2 or 3 decades. (But yes, I think you run the risk of being too conservative. Unless of course, you need to be that conservative in order to sleep at night and not panic and make behavioral mistakes.)
25 years is a good percentage of the 90-130 years of market history. Who knows if you will look like an idiot buying at these prices (i.e. PEs regress to the mean of the years up to 1990), genius (PE go to 40 and stay there) or something in between. In the OP case being 65/35 instead of 80/20 is unlikely make a huge difference. And if you figure on average the AA will be more like 70-75% stocks we are talking even smaller differences. Of course you can say the same thing about the upsides.

CFK
Posts: 13
Joined: Fri Oct 06, 2017 11:06 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by CFK » Mon Feb 19, 2018 5:15 pm

staythecourse wrote:
Mon Feb 19, 2018 3:57 pm

Do as you want as it is your money, but whatever you are doing IS market timing as you are making guesses on valuations. Simple as that. You may be correct or you may be completely wrong. You may make a ton of money, but you may end up broke. If you have that fallback inheritance go for it, but if you have one shot at retirement I wouldn't take a chance of getting something so wrong.
I doubt I'll make a ton of money, given that as equity markets become more expensive, I should become more conservative. Also pretty sure I won't end up broke, for the same reason. And given that I'm never more than 85% or less than 60% in stocks, I think extreme outcomes are unlikely. If I do end up broke, it will only be because of a massive collapse in the economy, in which case having a set allocation wouldn't have helped.

staythecourse
Posts: 5640
Joined: Mon Jan 03, 2011 9:40 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by staythecourse » Tue Feb 20, 2018 12:55 pm

CFK wrote:
Mon Feb 19, 2018 5:15 pm
staythecourse wrote:
Mon Feb 19, 2018 3:57 pm

Do as you want as it is your money, but whatever you are doing IS market timing as you are making guesses on valuations. Simple as that. You may be correct or you may be completely wrong. You may make a ton of money, but you may end up broke. If you have that fallback inheritance go for it, but if you have one shot at retirement I wouldn't take a chance of getting something so wrong.
I doubt I'll make a ton of money, given that as equity markets become more expensive, I should become more conservative. Also pretty sure I won't end up broke, for the same reason. And given that I'm never more than 85% or less than 60% in stocks, I think extreme outcomes are unlikely. If I do end up broke, it will only be because of a massive collapse in the economy, in which case having a set allocation wouldn't have helped.
Okay let me rephrase then moving from 60-80% stocks based on some RANDOM valuation that you have thought of which I am sure has already been computer tested probably 20 years will likely make you LESS rich then you could have been. If you buck the probabilities then you end up a little richer. Doesn't seem like a good bet, but your money.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

staythecourse
Posts: 5640
Joined: Mon Jan 03, 2011 9:40 am

Re: Dynamic Allocation based on the Intelligent Investor

Post by staythecourse » Tue Feb 20, 2018 1:02 pm

asif408 wrote:
Mon Feb 19, 2018 4:21 pm
I have no doubt that many, if not most of the pros are aware of things like CAPE, it's just that most of them don't have a long enough time frame to take advantage of it. Do you really think the big institutional and hedge fund managers wouldn't get fired or replaced if they used a strategy like CAPE and it didn't pan out in a year or two or they underperformed considerably? And how many individual investors would stick with a strategy that has years of under or lagging performance with the promise of longer-term performance?
That is the usual mantra, but wonder how often institutions really do change money managers. Maybe a few, but I don't think very often. I remember MANY Of the institutions were lagging a simple sp500 the last 5 years several years ago and nearly NONE of them got fired. Who did get fired was the alternative guys who were farmed out by the money managers. Hey, somebody had to fall on the sword.

There is A LOT of money riding on those accounts and sadly for the beneficiaries they (money managers) are NOT chosen due to their investing acumen, but the relationships between the top admin/ pension committees and the money managers. Trust me, it is not so easy to pass up on John after EVERY year he takes the whole board out to Vegas rents out a race track and everyone drives a lambo of their choice and one gets to take on one home as a "gift". That is the reason they don't change. Has NOTHING to do with performance or lack of it.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

magneto
Posts: 934
Joined: Sun Dec 19, 2010 10:57 am
Location: On Chesil Beach

Re: Dynamic Allocation based on the Intelligent Investor

Post by magneto » Tue Feb 20, 2018 1:31 pm

CFK wrote:
Mon Feb 19, 2018 5:15 pm
staythecourse wrote:
Mon Feb 19, 2018 3:57 pm

Do as you want as it is your money, but whatever you are doing IS market timing as you are making guesses on valuations. Simple as that. You may be correct or you may be completely wrong. You may make a ton of money, but you may end up broke. If you have that fallback inheritance go for it, but if you have one shot at retirement I wouldn't take a chance of getting something so wrong.
I doubt I'll make a ton of money, given that as equity markets become more expensive, I should become more conservative. Also pretty sure I won't end up broke, for the same reason. And given that I'm never more than 85% or less than 60% in stocks, I think extreme outcomes are unlikely. If I do end up broke, it will only be because of a massive collapse in the economy, in which case having a set allocation wouldn't have helped.
+1
It is all about risk management.
I.E. 'Downside Risk'.

Posted yet again :-
Re so-called 'Market Timing'.
Benjamin Graham had this to say in The Intelligent Investor. "Since common stocks, even of investment grade, are subject to recurrent and wide fluctuations in their prices, the intelligent investor should be interested in the possibilities of profiting from these pendulum swings. There are two possible ways by which he may try to do this : by way of TIMING and the way of PRICING. By timing we mean the endeavour to anticipate the action of the stock …… . By pricing we mean the endeavour to buy stocks when they are below their fair value and to sell them when they rise above such value. … We are convinced that the intelligent investor can derive satisfactory results from pricing …. We are equally sure that if he places his emphasis on timing, in the sense of forecasting, he will end up a speculator and with a speculator’s financial results This distinction may seem rather tenuous to the layman, and it is not commonly accepted in Wall Street."

But this 'Market Timing' straw-man will never be slain. :!:
It will run and run as it always has :(
Of the several Investment Formula Plans out there, only the Constant Ratio inspires unusual devotion.
'There is a tide in the affairs of men ...', Brutus (Market Timer)

Post Reply