Valuation-based market timing with PE10 can improve returns?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Bongleur
Posts: 2037
Joined: Fri Dec 03, 2010 10:36 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Bongleur » Mon Apr 09, 2012 12:47 pm

All I can access is the draft.

I note that 50:50 yielded about 1/7th the gains of 100% timed. So a more conservative timing (being in the market 100% but less often) could increase "safety" and still provide something like 3 to 4 times the yield of a constant 50:50. So an investor who has calculated how much wealth is "enough" could find something useful in that.

Also exactly what Treasuries were used? Did it assume a constant 30 year maturity (ie trading yearly for a new issue), or buying a 30 year and holding until the timing said to swap? You could examine the effect of buying ladders as well. And a mix of maturities.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

jbaron
Posts: 115
Joined: Fri Mar 18, 2011 3:32 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by jbaron » Mon Apr 09, 2012 1:03 pm

I could not access the paper either, so not of much use to me.

User avatar
CaliJim
Posts: 2725
Joined: Sun Feb 28, 2010 8:47 pm
Location: San Fran Bay Area

Re: Valuation-based market timing with PE10 can improve retu

Post by CaliJim » Mon Apr 09, 2012 3:22 pm

-calijim- | | For more info, click this

chaz
Posts: 13601
Joined: Tue Feb 27, 2007 2:44 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by chaz » Mon Apr 09, 2012 3:33 pm

Bogleheads abhor timing.
Chaz | | “Money is better than poverty, if only for financial reasons." Woody Allen | | http://www.bogleheads.org/wiki/index.php/Main_Page

User avatar
BlueEars
Posts: 3531
Joined: Sat Mar 10, 2007 12:15 am
Location: West Coast

Re: Valuation-based market timing with PE10 can improve retu

Post by BlueEars » Mon Apr 09, 2012 3:45 pm

chaz wrote:Bogleheads abhor timing.

You are talking about the pure ones I presume.

mosu
Posts: 170
Joined: Wed Jan 26, 2011 4:54 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by mosu » Mon Apr 09, 2012 3:48 pm

.....
Last edited by mosu on Thu Mar 12, 2015 9:25 pm, edited 1 time in total.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Rodc » Mon Apr 09, 2012 3:54 pm

I note that 50:50 yielded about 1/7th the gains of 100% timed.


I did not read the paper, but are you saying it says timing increases returns by 700%?
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
CaliJim
Posts: 2725
Joined: Sun Feb 28, 2010 8:47 pm
Location: San Fran Bay Area

Re: Valuation-based market timing with PE10 can improve retu

Post by CaliJim » Mon Apr 09, 2012 4:11 pm

Rodc wrote:
I note that 50:50 yielded about 1/7th the gains of 100% timed.


I did not read the paper, but are you saying it says timing increases returns by 700%?


yes - over a hundred and something year period. the data is quite astounding. the power of compound interest. small difference in return compounding over a long time.
-calijim- | | For more info, click this

User avatar
wade
Posts: 539
Joined: Fri Sep 17, 2010 12:38 am
Location: Main Line
Contact:

Re: Valuation-based market timing with PE10 can improve retu

Post by wade » Mon Apr 09, 2012 5:15 pm

jbaron wrote:I could not access the paper either, so not of much use to me.


The article has been discussed already, so I just wished to point out it was published and thank everyone. I do understand the published version will be hard to find for free unless you have access to a university subscription.

A free "working paper" version of the finalized article can be downloaded from here:

http://ideas.repec.org/p/pra/mprapa/29448.html

And thank you, CaliJim.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Rodc » Tue Apr 10, 2012 7:28 am

CaliJim wrote:
Rodc wrote:
I note that 50:50 yielded about 1/7th the gains of 100% timed.


I did not read the paper, but are you saying it says timing increases returns by 700%?


yes - over a hundred and something year period. the data is quite astounding. the power of compound interest. small difference in return compounding over a long time.


Ah.

Yes under the microscope of 100 years of compounding even noise looks huge.

However, 100 year and 700% means a gain of about 2% per year, which is a little more modest than 700%. :)

Nothing like that has ever been shown in the real world to the best of my knowledge. I would suggest that thinking timing will improve returns by 2% going forward does not pass the simplest "Does this make any sense?" filter.

Added, I'll soften that last comment, I am currently still reacting to the 700% claim (if it ever applies it would be someone who lump sum invested and waited 100 years before spending money). Two percent is not totally crazy, though I remain deeply skeptical that fine tuned back tested strategies will work any think like that well going forward.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

nolabogle
Posts: 103
Joined: Sat May 05, 2012 4:44 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by nolabogle » Thu Sep 06, 2012 8:45 pm

I recently came across this thread and read through its numerous well thought and argued posts/replies. I have a question for Professor Pfau or other knowledgeable members. Prof. Pfau's research utilizes the PE10. Currently PE10 is 23 (http://www.multpl.com/shiller-pe/). Its been above the long-term mean since approximately August 2009. Prof Pfau's research suggests that this PE10 value would dictate a relative less equity weighted portfolio. (I believe?) TSM has returned approximately 50% cumulative since then. A relative lesser equity portfolio as guided by your theory/research would have partially missed out on these returns following the 2008 markey downturn.

It seems to me that equities are relatively fairly valued currently and have been since approximately June 2010 based on current PE. PE10 seems to be still skewed upward by valuations throughout the 2000's. Thus an investor may partially missed much of these returns based on PE10 investment driven decisions. My question is where does current PE (vs PE10) fit into Prof. Pfau's theory and research? Current PE valuations seem to dictate a relatively more heavily equity position currently based on his evidence based valuation research.

P.S. for Boglehead purists, IMHO its ok to discuss valuations as Mr. Bogle himself can be found here discussing valuations with Morningstar. http://www.morningstar.com/cover/videoc ... ?id=397707

Moderator Note: This post is a continuation of a 5-month old thread.

Bongleur
Posts: 2037
Joined: Fri Dec 03, 2010 10:36 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Bongleur » Thu Sep 06, 2012 10:13 pm

The question to ask is how fast might the PE10 drop to a number where buying in at that level, later on, would make up for the time out of the market. While out of equities, you are making a little money in bonds at much lower risk (although at the moment the government ZIRP policy is distorting that market, in order to induce people to take the risks inherent in equity markets).

A person interested in capital preservation over growth might be happy making zero real in bonds for a while, until the risk:return ratio of equities is more favorable.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

User avatar
jeffyscott
Posts: 6578
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Re: Valuation-based market timing with PE10 can improve retu

Post by jeffyscott » Fri Sep 07, 2012 7:19 am

nolabogle wrote:PE10 seems to be still skewed upward by valuations throughout the 2000's.


I do not believe this is correct. PE10 is not the average of the PE over the last 10 years as this comment implies. It takes the current price and divides it by the average earnings over the last 10 years...it's P/E10 or Pcurrent/E10, not (P/E)10. Thus in order to be skewed high due to past data, it would be low past earnings that would have to be the cause of the skewing.

While the low recession earnings may be a factor, other indications are that current corporate earning are unusually high. Perhaps this is a permanently high plateau, but if not then it could be that the thing that is skewed is the current PE, while for PE10 the current high earnings are offset by the past low recession earnings.
press on, regardless - John C. Bogle

nolabogle
Posts: 103
Joined: Sat May 05, 2012 4:44 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by nolabogle » Fri Sep 07, 2012 9:03 am

I do not believe this is correct. PE10 is not the average of the PE over the last 10 years as this comment implies. It takes the current price and divides it by the average earnings over the last 10 years...it's P/E10 or Pcurrent/E10, not (P/E)10. Thus in order to be skewed high due to past data, it would be low past earnings that would have to be the cause of the skewing.


Thank you for the clarification; I was unaware of this.

While the low recession earnings may be a factor, other indications are that current corporate earning are unusually high. Perhaps this is a permanently high plateau, but if not then it could be that the thing that is skewed is the current PE, while for PE10 the current high earnings are offset by the past low recession earnings.


Thank you for the further insight into this topic.

User avatar
Dick Purcell
Posts: 520
Joined: Tue Oct 26, 2010 1:58 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Dick Purcell » Fri Sep 07, 2012 9:35 am

Somewhere in those 13+ pages of earlier discussion in this thread, the point is made that virtually all of that discussion fails to address the investor purpose – best probabilities for a particular plan's future net real dollar needs and goals.

With re-ignition of this thread, this point should be made again.

Higher PE10 means lower return-rate prospects ahead. This may mean that for sufficient growth potential for best prospects for the particular plan's future dollar goals, it’s best to increase the equity allocation.

And it’s likely that the most important response is adjustment not of allocation but of the cash flow plan. Higher PE10 means lower return-rate prospects ahead means increase investments or reduce withdrawals

Dick Purcell

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Valuation-based market timing with PE10 can improve retu

Post by 555 » Fri Sep 07, 2012 6:53 pm

nolabogle wrote:Moderator Note: This post is a continuation of a 5-month old thread.

Actually the Original Post was made on Sat May 28, 2011 4:52 pm, but then on Thu Feb 23, 2012 8:24 pm the OP editted it to literally say
fredflinstone wrote:.....

viewtopic.php?f=10&t=75585

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by DRiP Guy » Fri Sep 07, 2012 10:38 pm

Editing posts shortly after posting in order to correct issues the OP notes with grammar, clarity, spelling or minor omissions, seems fine to me. I certainly do that myself.

But I do wish posters would not edit posts that are already long-standing, and in particular not completely obliterate them, since it can ruin the context of the subsequent dialog.

Oh well. I suppose people can have their reasons. [shrug]

Bongleur
Posts: 2037
Joined: Fri Dec 03, 2010 10:36 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Bongleur » Fri Sep 07, 2012 10:52 pm

So the first responder should always quote the entire OP...
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

User avatar
DRiP Guy
Posts: 2237
Joined: Tue Feb 20, 2007 4:54 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by DRiP Guy » Fri Sep 07, 2012 10:58 pm

Bongleur wrote:So the first responder should always quote the entire OP...


Nope, just the relevant part he finds important to contextualizing his own reply.

(IMHO, of course!)


But we all get lazy, and sometimes don't quote at all....

User avatar
zaboomafoozarg
Posts: 1849
Joined: Sun Jun 12, 2011 12:34 pm

Re: Valuation-based market timing with PE10 can improve retu

Post by zaboomafoozarg » Sat Sep 08, 2012 12:11 am

OP fail.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Rodc » Sat Sep 08, 2012 7:58 am

Dick Purcell wrote:Somewhere in those 13+ pages of earlier discussion in this thread, the point is made that virtually all of that discussion fails to address the investor purpose – best probabilities for a particular plan's future net real dollar needs and goals.

With re-ignition of this thread, this point should be made again.

Higher PE10 means lower return-rate prospects ahead. This may mean that for sufficient growth potential for best prospects for the particular plan's future dollar goals, it’s best to increase the equity allocation.

And it’s likely that the most important response is adjustment not of allocation but of the cash flow plan. Higher PE10 means lower return-rate prospects ahead means increase investments or reduce withdrawals

Dick Purcell


Agree.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

letsgobobby
Posts: 10616
Joined: Fri Sep 18, 2009 1:10 am

Re: Valuation-based market timing with PE10 can improve retu

Post by letsgobobby » Sat Sep 08, 2012 8:35 am

No, that would end up being performance chasing like going all in on stocks in January 2000. High PE10 should cause one to expect lower future returns, I agree with that. But the solution in that moment is to save more money, not to take exponentially more risk just when the risks are highest. In this way we see that extremes of PE10 affect not only investment decisions but personal spending/budgeting decisions as well.

User avatar
jeffyscott
Posts: 6578
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Re: Valuation-based market timing with PE10 can improve retu

Post by jeffyscott » Sat Sep 08, 2012 8:52 am

Also, there are two different scenarios to consider.

One is you have substantial assets invested prior to the run up in PE. In this scenario you basically got something like 10 years of expected returns in just 3 years. In addition the run up would increase your stock allocation and the correct response would be to at a minimum sell and rebalance.

The other scenario is you are in accumulation and have many years left to do this. In this situation, rather than putting even more in stocks because the future returns are lower, you could instead accumulate some assets in safer places and wait (hope) for a better price to buy at. This also implies you must have an alternate plan in case what you hope for does not occur.

But if you decide that the response to high prices is to buy even more stocks, then you also would need to have an alternate plan in case your hopes are not realized in that case. I assume what would be hoped for is that we have achieved a permanent high plateau of stock prices or perhaps a near term collapse in stock prices, before you have put too much in at high prices?
press on, regardless - John C. Bogle

letsgobobby
Posts: 10616
Joined: Fri Sep 18, 2009 1:10 am

Re: Valuation-based market timing with PE10 can improve retu

Post by letsgobobby » Sat Sep 08, 2012 10:42 am

"Stock prices have reached what looks like a permanently high plateau."

- Irving Fisher, prominent American economist, 3 days before the 1929 stock market crash

User avatar
NoRoboGuy
Posts: 723
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: Valuation-based market timing with PE10 can improve retu

Post by NoRoboGuy » Sat Oct 20, 2012 12:09 pm

jeffyscott wrote:One is you have substantial assets invested prior to the run up in PE. In this scenario you basically got something like 10 years of expected returns in just 3 years. In addition the run up would increase your stock allocation and the correct response would be to at a minimum sell and rebalance.


This is how I viewed the PE10 debate...at first. Then I noted that Wade in this subject study does take into account a 50/50 allocation with rebalancing to make the timing comparisons. Another point is most investors can not stomach changing between a 0/100 and a 100/0 allocation, so he also makes the comparison using 80/20 to 20/80, 70/30 to 30/70, and so on. In every case, the PE10 valuation approach which Jack Bogle himself views as a credible measure of valuation, appears sound.

I am interested to know if the reliability of the valuation measure can be improved though the combination of other measures like Tobin's Q. The trick is to do so without data mining. I think Wade did an excellent job of handling that aspect with the PE10 study.
There is no free lunch.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: Valuation-based market timing with PE10 can improve retu

Post by Rodc » Sat Oct 20, 2012 2:57 pm

louis c wrote:
I am interested to know if the reliability of the valuation measure can be improved though the combination of other measures like Tobin's Q. The trick is to do so without data mining. I think Wade did an excellent job of handling that aspect with the PE10 study.


http://home.comcast.net/~rodec/finance/ ... mingV2.pdf

Probably not. I looked at something similar a few years back with Tobin's q. Worked in the early years (data used to build and test q) but not in later years (independent data).

One can always fine tune the back testing to make something work. But if you have to fine tune on back tested data, you are on very shaky ground.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

User avatar
William Million
Posts: 544
Joined: Wed May 05, 2010 4:41 am
Location: A Deep Mountain

Re: Valuation-based market timing with PE10 can improve returns?

Post by William Million » Sun Mar 13, 2016 5:35 am

Has the PE10 crowd thrown in the towel yet?

Quark
Posts: 1045
Joined: Sun Nov 01, 2015 5:32 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by Quark » Sun Mar 13, 2016 6:25 am

William Million wrote:Has the PE10 crowd thrown in the towel yet?

Unlikely. What's happened recently that should cause them to do so?

User avatar
William Million
Posts: 544
Joined: Wed May 05, 2010 4:41 am
Location: A Deep Mountain

Re: Valuation-based market timing with PE10 can improve returns?

Post by William Million » Sun Mar 13, 2016 6:57 am

Quark wrote:
William Million wrote:Has the PE10 crowd thrown in the towel yet?

Unlikely. What's happened recently that should cause them to do so?


Failing, due to PE10 adherence, to exploit the bull market run since this thread started in 2011 would do it for me.

CAPE was about 23 at that time, indicating an over-valued market. Anyone who eased up on equities due to the "overvalued market" would've missed some pretty good years: 2012 +16%, 2013 +32%, 2014 +14%.

Under the best of circumstances, PE10 is a poor valuation metric. Using it in isolation of other valuation metrics is crazy.

User avatar
Rx 4 investing
Posts: 735
Joined: Sat Apr 25, 2015 11:03 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by Rx 4 investing » Sun Mar 13, 2016 1:15 pm

William Minton wrote: “ Failing, due to PE10 adherence, to exploit the bull market run since this thread started in 2011 would do it for me. CAPE was about 23 at that time, indicating an over-valued market. Anyone who eased up on equities due to the "overvalued market" would've missed some pretty good years: 2012 +16%, 2013 +32%, 2014 +14%.Under the best of circumstances, PE10 is a poor valuation metric. Using it in isolation of other valuation metrics is crazy.”

A valuation-agnostic investor who allocated to the canonical “60/40” buy-and-hold portfolio is likely to be disappointed by the following results from the Portfolio visualizer backtesting tool. ( results may vary based on stock-bond allocations.)

First, a “refresher” on how Dr. Shiller believes his Shiller P/E (CAPE 10) ratio should be used by valuation-conscious investors…

…” I think the bottom line that we were giving – and maybe we didn’t stress or emphasize it enough – was that it’s continual. It’s not a timing mechanism, it doesn’t tell you...wait until it goes all the way down to a P/E of 7, or something... But actually, the lesson there is that if you combine that with a good market diversification algorithm, the important thing is that you never get completely in or completely out of stocks. The lower CAPE is, as it gradually gets lower, you gradually move more and more in." In other words, don't dump stocks and hide in cash because the CAPE is at (fill the blank). Rather, buy less, be cautious, and expect lower returns for years to come."


1. The last decade (i.e. 2006-2015). The Shiller P/E (CAPE 10) was conceived to suggest a possible “range” of stock return outcomes for the next decade.

--The investment period from 2006-2015 covered the last 10 years.

--On January 1, 2006, the CAPE reading was at 26.47. (Source: multpl.com)

Results from the portfoliovisualizer.com back-test tool under “timing models” (link), using their Shiller P/E valuation corridors, shows that the valuation-based investor would have beaten the valuation-agnostic 60/40 investor on many important metrics. Perhaps of interest…the Shiller PE timing model beat 60/40 on “final balance” , "CAGR" and “Sharpe ratio (risk-adjusted) ” .

Code: Select all

Model portfolio              Amt Invested       Timespan         Final Balance            CAGR     Sharpe Ratio

Shiller P/E                 $10,000                  ’06-’15               $22,963            8.67%      0.91

60/40                      $10,000                 ’06-’15                $20, 193           7.28%      0.65


Source: https://www.portfoliovisualizer.com/tes ... sisResults

2. The poster's "would do it for me" period from 2011-2015.

--On January 1, 2011, the CAPE reading was at 22.98. (above the 50 year mean of 19.8).

--Although P/E 10 is not ideally intended for forecasts less than 10 years, nonetheless the concept is supported by the iron law of valuation, i.e stocks are a claim on an expected stream of future cash flows, so the current price of a stock index moves opposite to the expected future return on that stock index.

Again, results from the portfolio visualizer back-test shows that the valuation-based investor would have beaten the valuation-agnostic 60/40 investor on many important metrics. Again, looking at few, the Shiller PE timing model beat 60/40 on “final balance” , "CAGR", and “Sharpe ratio (risk-adjusted) ” .

Code: Select all

Model portfolio              Amt Invested       Timespan         Final Balance                  CAGR      Sharpe Ratio

Shiller P/E                $10,000                  ’11-’15              $15,903              9.72%       1.78

60/40                      $10,000                  ’11-’15               $15,529             9.20%     1.34                                   

https://www.portfoliovisualizer.com/tes ... sisResults

Comments:

--The more stock risk an investor takes, the higher the expected return. Interestingly, the 100% stock portfolio did out-perform during the shorter-time span ('11-15), but did not out-perform over last decade--when considering both final balance and Sharpe score.

--The Shiller P/E timing model offers promise to the the conservative retirement investor who is interested in risk-adjusted return.

Image


Also for consideration...


An alternate method for the buy-and-hold investor is to consider utilizing an often over-looked financial planning concept …

A buy-and-hold investor with no particular view about market conditions or future returns, and who wishes to have a fairly predictable amount of wealth at some future date, should hold a portfolio with a “duration” that is roughly equal to the investment horizon.

This concept might be more helpful for a pre-retiree closer to his/her need to spend a portion of their nest egg (a "bucket") . The pre-retiree has a shorter runway and faces the imminent future of no more paychecks, compared to a younger --with a long-time horizon ---who will be able to D-C-A (dollar-cost-average) to mitigate today’s long “duration” of stocks.

--The approximate “modified duration” of the S & P 500 based on today’s Price /Dividend ratio is currently 46 years.

--A buy-and-hold investor who desires a reasonable amount of certainty around their future balance needed for spending ten (10) years might consider 22% (or less) in stocks, and the balance in intermediate-term bonds and cash. (10 years time horizon /46 years duration = 22%)

FYIs…

--The duration of cash = 0 years;
--The duration of Vanguard’s Total Bond Market index fund is 5.8 years.
--The current approx. duration of the S & P 500 is 46 years.

At present, the canonical 60/40 portfolio has a duration of 29.3 years.

(0.6 x 46) + ( 0.30 x 5.8) + ( 0.1 x 0) = 27.6 + 1.74 + 0 = 29.3

Good luck out there. :moneybag
“Everyone is a disciplined, long-term investor until the market goes down.” – Steve Forbes

User avatar
William Million
Posts: 544
Joined: Wed May 05, 2010 4:41 am
Location: A Deep Mountain

Re: Valuation-based market timing with PE10 can improve returns?

Post by William Million » Fri Mar 25, 2016 10:04 pm

Someone might make good money in a PE10 contrarian fund.

User avatar
Rx 4 investing
Posts: 735
Joined: Sat Apr 25, 2015 11:03 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by Rx 4 investing » Sat Mar 26, 2016 1:41 pm

William Minton wrote: “ Someone might make good money in a PE10 contrarian fund.”

As Michael Kitces has written, the Shiller PE 10 is a lousy short-term timing mechanism. But he wonders if the Nobel committee split the prize between Shiller and Eugene Fama because while the market can’t be predicted in the short-term, the P/E 10 can be helpful with long-term forecasts.

Backtests from the Portfolio Visualizer’s P/E 10 model offer some promising results in terms of final amounts, CAGRs, and Sharpe ratios …

--Since the market bottom in 2009, and through December 2015 (7 years) , the model has beaten the 60/40 portfolio.

--In the last 30 years (1985-2015), the model beat the 60/40 portfolio.

--In the last 40 years (1975 -2015), which might cover a typical retirement investing time horizon for those who start investing at age 25 and retire at 65, the PE 10 model beat the 60/40 portfolio.

--Benjamin Graham preferred a minimum of 50 years to develop a comfort level with results. Over the period 1965-2015, the PE 10 timing model beat the 60/40 portfolio.

https://www.portfoliovisualizer.com/tes ... sisResults

IMO: The canonical 60/40 "balanced" portfolio is a reasonable comparison for the Shiller P/E 10 timing model in my view. The 60/40 portfolio is stock-heavy, and thus points in the right direction for a long-term, retirement investor.

How did they arrive at the stock allocations algorithm for the PV timing tool? I don’t honestly know. For fair and balanced, other BH posters with sophisticated software tools, and strong statistical knowledge, have posted their results from various P/E 10 algorithms at the BH website over the past few years. And depending on the stock ranges used, results are variable.

The PV timing model algorithm is generally in keeping with how Dr. Shiller thinks P/E 10 should be used for investment allocation decisions. As Sam Ro wrote in a BI article a few years back, “ You're Not Listening To Robert Shiller If His CAPE Ratio Has You Scared Of Stocks.” Here’s Dr. Shiller’s own words from an interview…

“John Campbell, who’s now a professor at Harvard, and I presented our findings first to the Federal Reserve Board in 1996, and we had a regression, showing how the P/E ratio predicts returns. And we had scatter diagrams, showing 10-year subsequent returns against the CAPE, what we call the cyclically adjusted price earnings ratio. And that had a pretty good fit. So I think the bottom line that we were giving – and maybe we didn’t stress or emphasize it enough – was that it’s continual. It’s not a timing mechanism, it doesn’t tell you – and I had the same mistake in my mind, to some extent — wait until it goes all the way down to a P/E of 7, or something.
But actually, the lesson there is that if you combine that with a good market diversification algorithm, the important thing is that you never get completely in or completely out of stocks. The lower CAPE is, as it gradually gets lower, you gradually move more and more in. So taking that lesson now, CAPE is high, but it’s not super high. I think it looks like stocks should be a substantial part of a portfolio. “


http://www.businessinsider.com/robert-s ... pe-2013-11

Good luck in the years ahead.
“Everyone is a disciplined, long-term investor until the market goes down.” – Steve Forbes

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Mon Mar 28, 2016 1:12 pm

Rx 4 investing wrote:Backtests from the Portfolio Visualizer’s P/E 10 model offer some promising results in terms of final amounts, CAGRs, and Sharpe ratios …

--Since the market bottom in 2009, and through December 2015 (7 years) , the model has beaten the 60/40 portfolio.

--In the last 30 years (1985-2015), the model beat the 60/40 portfolio.

--In the last 40 years (1975 -2015), which might cover a typical retirement investing time horizon for those who start investing at age 25 and retire at 65, the PE 10 model beat the 60/40 portfolio.

--Benjamin Graham preferred a minimum of 50 years to develop a comfort level with results. Over the period 1965-2015, the PE 10 timing model beat the 60/40 portfolio.

https://www.portfoliovisualizer.com/tes ... sisResults


That PE10 model back-test from Portfolio Visualizer is junk. Please see the thread about it, and please stop posting it as "proof" of anything.

https://www.bogleheads.org/forum/viewtopic.php?f=10&t=186920

User avatar
Rx 4 investing
Posts: 735
Joined: Sat Apr 25, 2015 11:03 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by Rx 4 investing » Wed Mar 30, 2016 9:44 am

Homer J wrote: “That PE10 model back-test from Portfolio Visualizer is junk. Please see the thread about it, and please stop posting it as "proof" of anything.”

You seem to believe an investor can completely dismiss starting valuations with minimal impact on long-term investing results? You reference the musings of website bloggers and amateur investors who argue that a Nobel prize-winning model is so flawed that it can be dismissed? :oops:

Here is more evidence that starting valuation does matter, and it has big influence on final amounts. If your notion of valuation-agnostic stock investing were to have any merit, an investor who ignored valuation and committed a lump sum--- with a one year head start--- should have more money today than an investor late to game. Let’s put that to the test.

Here are the results from Portfolio Visualizer of a $10,000 investment made into Vanguard’s Total Stock Market ETF (VTI) …

Code: Select all

Fund      Initial amount  (lump)       Time Period    Starting P/E 10    Final Balance         CAGR

VTI         $10,000              2008-16 (8yrs)           24.02              $15,936            5.87%

VTI        $10,000               2009-2016  (7yrs)       15.17             $25,286             13.82%   


If your point is that a valuation-agnostic investor who purchased in 2008 at P/E 10 of 24 has made some money, then the evidence says that is true. On the other hand, the risk-averse investor who was mindful of valuations, and patiently waited for a better entry point a year later at P/E 10 of 15.2, has made a lot more. Investing at a lower valuation level, and side-stepping a 40% draw-down, has seen money compound at a faster rate.

The current Shiller PE Ratio is at 26.13, and with a Nobel prize and 100+ years of market history backing it, a risk-averse investor should be mindful that the pattern is likely to repeat. Stock earnings, the mother’s milk of stocks, are cyclical, just like the business cycle.

Carry on , and stay calm. :beer
“Everyone is a disciplined, long-term investor until the market goes down.” – Steve Forbes

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Wed Mar 30, 2016 11:00 am

I specifically address the PE10 "model" on Portfolio Visualizer that is constructed using ex-post-facto data, and setting PE10 inflection points that get one into stocks at exactly the bottom of 2009 (not half-way down, right at the bottom).

You keep bringing it up showing how that model beats buy and hold. But that model is junk. It's fantasy.

You may be correct that valuations matter in a general sense, but that Portfolio Visualizer "model" is deeply deeply flawed. You have been convinced that valuations allow one to market-time successfully by websites presenting false data. This should bother you.
Last edited by HomerJ on Wed Mar 30, 2016 1:54 pm, edited 2 times in total.

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Wed Mar 30, 2016 11:10 am

Rx 4 investing wrote:If your point is that a valuation-agnostic investor who purchased in 2008 at P/E 10 of 24 has made some money, then the evidence says that is true. On the other hand, the risk-averse investor who was mindful of valuations, and patiently waited for a better entry point a year later at P/E 10 of 15.2, has made a lot more.


Sure, buying at lower valuations is highly likely to result in better returns. I agree with you there.

But "patiently waiting for a better entry point" based on valuations is market-timing. Your above statement looks great if we have a time-machine or could see the future in 2008.

But in reality, someone following Schiller's model would have been "patiently waiting" from 1992-2009. Not one year... 17 years. And again from 2011 to 2016. While watching the market nearly double in 5 years, even starting from "high" valuations.

1992-2016 has not followed the CAPE model (created in 1988) very well at all. Investing at high valuations of 20+, even 25+ have paid off just fine over the past 24 years.

When you have a model that is based on 80 years of data, and then the next 24 years don't fit inside the model very well at all, I find it hard to believe that one wouldn't wonder a little bit if the model might have a flaw. There's a ton of independent variables to consider, and the model addresses only one.

edge
Posts: 3221
Joined: Mon Feb 19, 2007 7:44 pm
Location: Great Falls VA

Re: Valuation-based market timing with PE10 can improve returns?

Post by edge » Wed Mar 30, 2016 11:33 am

Just seems like people looking for certainty where there is none.

PE/10 is constantly shifting and there is nowhere near enough data. Over its history in the U.S. the average has continually risen and is non-adjusted for material accounting and behavioral changes in the market (e.g. dividend yield shifts and buy back changes). Relying on an old 'mean' in this scenario is illogical. Granted, paying attention to the absolute ends of the tails makes sense, at least to me. But P/E does a fine job of that without the need for multi-year smoothing.

rgs92
Posts: 1433
Joined: Mon Mar 02, 2009 8:00 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by rgs92 » Wed Mar 30, 2016 11:37 am

So what does the CAPE model say you should do now as far as setting your asset allocation?

User avatar
Rx 4 investing
Posts: 735
Joined: Sat Apr 25, 2015 11:03 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by Rx 4 investing » Wed Mar 30, 2016 11:54 am

Homer J wrote: "You have been convinced that valuations allow one to market-time successfully by websites presenting false data. This should bother you."

I am not bothered in the least by valuation-based approaches . The notion was put forth many years ago by one of the most successful value investors ever, Benjamin Graham. His student, Warren Buffett, reminds small investors to re-read Chapters 8 and 20 in The Intelligent Investor at least once annually. Here are some supportive passages from Chapter 8 "The Investor and Market Fluctuations".

"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: the way of timing and the way of pricing...by pricing we mean the endeavor to buy stocks when they are quoted below their fair value and to sell them when they rise above such value. A less ambitious form of pricing is the simple effort to make sure that when you buy you do not pay too much for your stocks. This may suffice for the defensive investor, whose emphasis is on long-pull holding; but as such it represents an essential minimum of attention to market levels."

Further...

" Our recommended policy has made provision for changes in the proportion of common stocks to bonds in the portfolio, if the investor chooses to do so, according as the level of stock prices appears less or more attractive by value standards."



And on market fluctuations and the investor's portfolio...

"It is for these reasons of human nature, even more than by calculation of financial gain or loss, the we favor some kind of mechanical method for varying the proportion of bonds to stocks in the investor's portfolio. The chief advantage, perhaps, is that such a formula will give him something to do. As the market advances he will from time to time make sales out of his stock holdings, putting the proceeds into bonds; as it declines he will reverse the procedure. These activities will provide some outlet for his otherwise too-pent-up energies. If he is the right kind of investor, he will take added satisfaction from the thought that his operations are exactly opposite from those of the crowd."


The current P/E 10 is 26. Even if you argue that the mean P/E 10 of the last 50 years is roughly 20, how does one rationalize having the same portfolio allocation to stocks when the current reading is 30% higher? The data from PV that shows the investor who invested at lower than P/E 24 is richer today, even having started a year later.

Maybe it's the concept of "cyclical" , i.e. the "C" in C-A-P-E , that you refuse to assign any weight? Granted, CAPE is a horrible short-term timing tool, but it has fairly decent predictive value for 10-18 year time horizons. Do you really believe from your studies and life's experiences from the US and global economies that they are not cyclical in nature ? :confused
“Everyone is a disciplined, long-term investor until the market goes down.” – Steve Forbes

randomguy
Posts: 4885
Joined: Wed Sep 17, 2014 9:00 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by randomguy » Wed Mar 30, 2016 12:12 pm

Rx 4 investing wrote:
The current P/E 10 is 26. Even if you argue that the mean P/E 10 of the last 50 years is roughly 20, how does one rationalize having the same portfolio allocation to stocks when the current reading is 30% higher? The data from PV that shows the investor who invested at lower than P/E 24 is richer today, even having started a year later.


Average of the last 20 years is 27. Does that mean the market is undervalued today?

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Wed Mar 30, 2016 1:52 pm

Rx 4 investing wrote:The data from PV that shows the investor who invested at lower than P/E 24 is richer today, even having started a year later.


The data from PV is ex-post-facto. No one in the past would have considered a CAPE of 24 "low" enough to invest. They would have waited for CAPE below 20 AT LEAST (and probably waited for CAPE below 15).

But CAPE has been above 20 for 95% of the past 24 years.

You can't take the average from the past 50 years and use it to model what someone would have invested in 25 years ago. Someone from 25 years ago correctly calculated the average to be much lower. 20 was a very high PE. 25 was ridiculously high. The only time CAPE got above 25 before 1988 was right before the Great Depression. No one (including Schiller or Benjamin Graham) would have thought it was a good idea to invest at 23 PE.

But Portfolio Visualizer website does because they know how it turned out. Back-testing this way is junk. The data you are relying on is no good.

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Wed Mar 30, 2016 2:05 pm

Rx 4 investing wrote:The current P/E 10 is 26. Even if you argue that the mean P/E 10 of the last 50 years is roughly 20, how does one rationalize having the same portfolio allocation to stocks when the current reading is 30% higher?


Because the current P/E 10 in 1996 was 25 which was even higher than 30% above the average at the time. And the market has never been as low as it was in 1996. Why are you so certain that a PE10 of 26 is doom, when we have 20 years of data showing the opposite?

There will be another crash. It's okay to be certain about that. But we don't know when. Valuations have failed to predict crashes in a timely manner for the past 24 years. Why do you believe that today is different? Why do you think a crash Is imminent based on valuations?

I'm not saying you are wrong. But I'm saying you shouldn't be so certain you are right. The data contradicts your model.

User avatar
Rx 4 investing
Posts: 735
Joined: Sat Apr 25, 2015 11:03 am

Re: Valuation-based market timing with PE10 can improve returns?

Post by Rx 4 investing » Wed Mar 30, 2016 3:31 pm

Random guy asked: “Average of the last 20 years is 27. Does that mean the market is undervalued today?”


In the last 20 years, CAPE has been in a range from 15 to 44. Shiller recommends that stock allocations be adjusted periodically based on CAPE level, but to never get completely in nor completely out of stocks. Again, from the Shiller interview in B.I. …

“.. if you combine (CAPE) with a good market diversification algorithm, the important thing is that you never get completely in or completely out of stocks. The lower CAPE is, as it gradually gets lower, you gradually move more and more in. So taking that lesson now, CAPE is high, but it’s not super high. I think it looks like stocks should be a substantial part of a portfolio.”


The Portfolio Visualizer back-test tool uses a “dynamic” model between stocks and bonds, in keeping with Shiller’s notion quoted above, and uses the following algorithm…

-PE10 >= 22 - 40% stocks, 60% bonds

-PE 10 14 <= PE10 < 22 - 60% stocks, 40% bonds

-PE10 < 14 - 80% stocks, 20% bonds

The PV model would currently suggest a 40% stock allocation for a valuation-based investor ( CAPE 10 at 26 as of 3-30-16) . Jack Bogle doesn’t recommend dramatic moves in stock allocations either. ..

“Big moves out of stocks should not be done at all,” But…strategic asset allocation…— can be done at very rare times, so rare and so difficult to observe, maybe six (6) times in an investor’s lifetime, three (3) times when the market is stupidly high and three (3) times when stupidly low.”
And...

“…no one can ever be sure of the future path of the financial markets, the tactics I recommend would place severe restrictions on the extent of the allocation changes. Specifically, I would vary the desired strategic balance by no more than 15 percentage points on either side. A portfolio targeted at 50/50 would never have less than 35% in stocks nor more than 65%."


Besides valuation, might time horizon matter to a stock investor? Perhaps. A older research paper by the KC Fed cautioned that stock holding periods shorter than 25 years may not always out-perform gov’t bonds. Here are results from Portfolio Visualizer of $10,000 invested in Vanguard’s Long-Term treasury (VUSTX) vs. Vanguard’s T-M Balanced Fund (VTMFX) from the last 20 years:

Code: Select all

Fund             Time Frame                  Allocation            Final Balance

VUSXT           1995-2015 (20 yrs)           100%                       $ 48,325

VTMFX          1995-2015  (20 years)          100%                     $ 47,261         


Here’s the conclusion from the KC Fed paper:

“ This article confirms the conventional wisdom that in the United States stocks historically have been safer than long-term government bonds for investors with long holding periods. But the article also shows that the conventional wisdom has only been true for investors who held their portfolios for more than 25 years. For practical purposes, that may be too long a holding period for most investors. Over the years, for investors who have held their portfolios for shorter periods, both stocks and bonds were exposed to substantial risks, and stocks did not necessarily outperform government bonds. This implies that in making asset allocation decisions, investors should think carefully about how long they will be able to hold their portfolios undisturbed and how much risk they are willing to bear.”


https://www.kansascityfed.org/publicat/ ... 05shen.pdf
“Everyone is a disciplined, long-term investor until the market goes down.” – Steve Forbes

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Valuation-based market timing with PE10 can improve returns?

Post by HomerJ » Wed Mar 30, 2016 3:47 pm

Rx 4 investing wrote:Besides valuation, might time horizon matter to a stock investor? Perhaps.


You have it exactly backwards. Time horizon, by far, is the most important component for Asset Allocation. Valuation is a tiny component in comparison.

The day I retire, I will be 30/50/20 stocks/bonds/cash, regardless of valuations. It would be extremely foolish of me to be 80/20 stocks/bonds the day I retire just because valuations were at 14.

The stock market can crash at any time. A PE10 of 14 would not guarantee me riches. The stock market could crash from 14 down to valuations in the single-digits just like it has before, and stay super-low for an extended period of time, just like it has stayed super-high for an extended period of time.

On the flip side, if I was 30, I would be 80/20 stocks, regardless of valuations. Even If (when) a crash happens, I would have plenty of time to recover (and new contributions would be buying low). Trying to get in and out of the market, you can end up with far less than if you just stay in, all the way through a cycle. Plus, you don't have to worry about second-guessing yourself. Those 10% average historical gains INCLUDE the crashes and bear markets. You got the 10% without having to time anything.

People who got out in 1992, and watched the market triple over the next 7 years had to wonder if the model they were following was flawed in some way. Even when the crash came in 2000, it never got below where they were in 1992... heck it never got back below where they could have been in 1996. And PE10 never really dropped far enough during that crash to even justify getting back in. They watched the bottom appear, waited patiently for it to drop farther, and then had to sit there as the recovery started without them, and through the entire cycle the stock market remained highly over-valued by CAPE10 metrics.
Last edited by HomerJ on Wed Mar 30, 2016 5:09 pm, edited 1 time in total.

User avatar
Taylor Larimore
Advisory Board
Posts: 26129
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Components for Asset Allocation?

Post by Taylor Larimore » Wed Mar 30, 2016 5:09 pm

Time horizon, by far, is the most important component for Asset Allocation.

Homer:

How would you rank "risk tolerance" and "personal financial situation?"

Thank you and best wishes.

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
HomerJ
Posts: 10379
Joined: Fri Jun 06, 2008 12:50 pm

Re: Components for Asset Allocation?

Post by HomerJ » Wed Mar 30, 2016 5:25 pm

Taylor Larimore wrote:
Time horizon, by far, is the most important component for Asset Allocation.

Homer:

How would you rank "risk tolerance" and "personal financial situation?"

Thank you and best wishes.

Best wishes.
Taylor


Risk tolerance is a variable that depends on time horizon and personal financial situation.

Yes, if you have a ton of money, time horizon and AA doesn't really matter that much.

Risk tolerance, in my mind, is not how you FEEL about "staying the course", but having a plan that can actually withstand a worst case scenario, and there time horizon is very important.

User avatar
Taylor Larimore
Advisory Board
Posts: 26129
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Risk Tolerance

Post by Taylor Larimore » Wed Mar 30, 2016 7:13 pm

Homar J:

Thank you for your reply.

Thinking about your post, I came across this interesting Investopedia article about "risk tolerance":

What is Your Risk Tolerance?

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

Post Reply