Valuation-based market timing with PE10 can improve returns?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc »

Bongleur wrote:>Only of the goal is to target returns.

If the goal is to target risk you would raise it.
>

But you only take risk in the first place in order to achieve some required return.
That entirely misses the point.

You looked a low P/E10 and took a target return and made an adjustment to AA based on fixed expected return.

Another perfectly rational person could look at the exact same situation and make an to their AA based on a fixed willingness to take on risk.

P/E10 does not thus, in general, say very much about what one should, in general, do with AA.
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: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

Roy wrote:
louis c wrote: In the mid-nineties, was not the PE 10 ratio (>25) already saying the market was overvalued? If you had followed PE 10 you would have been under-weight stocks during one of the biggest 5-year bull market runs of the 20th century. PE 10 eventually peaked above 40. While you would have been under-weight during the subsequent decline, the starting point would have been lower having been under-weight the previous five years.
Precisely, Louis. And Irrational Exuberance was then sounded in '96. Anyone who went underweight, or out, missed three more years of stellar returns. Historical valuations levels can reset as they did then—moving the bar. Presumably the bar can be moved lower too. One can have some instances where PE/10 suggests good buying opportunities and the market go yet lower for an extended time. And then one's time horizon must be
considered; it is not a one-size-fits-all solution. Perhaps it has some utility
at valuation extremes. Maybe.

As I said elsewhere, there are also enormous psychological challenges to
implementing a strategy based purely on PE/10 (or whatever). This might
be the biggest practical problem. One would not only be rebalancing during scary times (something many avoid doing due to fear), but adding yet
greater amounts of equity; it is easy to say buy low in calm moments and hard to do in the cauldron—especially beyond rebalancing—with fresh dollars.
The reverse is true during booms, as Louis points out.

Good plans, especially conservative ones that truly accounted for market downturns, did fine.
Reiterating:


letsgobobby



Joined: 18 Sep 2009
Posts: 2024

Posted: Sat Jul 30, 2011 7:12 pm    Post subject:
No. Stocks fell to 1996 levels by the nadir in 2009. Pe10 correctly signaled that large caps were a bad investment beginning in the mid 90 s. I didn't miss a darn thing but volatility while the buy and holder missed 7% annual gains in bonds. Ignoring the message pe10 was sending in 1995-96 cost an imvestor dearly.

Just because something is psychologically hard doesn't mean it doesn't work. If it is hard then few will do it meaning there may be an inefficiency to exploit.
Bongleur
Posts: 2276
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur »

>Another perfectly rational person could look at the exact same situation and make an to their AA based on a fixed willingness to take on risk.
>

You think Traders are "perfectly rationale?" ;0
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.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Post by Roy »

letsgobobby wrote:Pe10 correctly signaled that large caps were a bad investment beginning in the mid 90 s. I didn't miss a darn thing but volatility while the buy and holder missed 7% annual gains in bonds. Ignoring the message pe10 was sending in 1995-96 cost an imvestor dearly.

Just because something is psychologically hard doesn't mean it doesn't work. If it is hard then few will do it meaning there may be an inefficiency to exploit.
The folks that got hurt are the ones that believed stocks were less risky and over committed to them. As in any period, greed is the problem, not failure to heed PE 10 signals—or any other timing mechanisms. A well-diversified, conservative portfolio (comprising 40% stocks and 60% Intermediate Treasuries) did just fine over a very long period, right through the 90s—and the two Bears that followed. Even something as boring as 40% TSM (Large Cap dominated) plus 60% IT Treasuries did fine— and coasted right through the '01-'02 Bear.

Here is how that simple mix did through the period you describe (and this assumes LC dominated equities—using just rebalancing):

2002 0.11%
2001 0.14%
2000 4.19%
1999 7.41%
1998 15.67%
1997 17.77%
1996 9.54%
1995 26.58%

Making something harder than it needs to be doesn't make it easier for anyone—especially those who believe they can spot and exploit inefficiencies. Even Shiller has been wishy-washy when asked in interviews about how he was investing at around the March lows—he said he was waiting for the PE to go even lower (like PE 10). Then, 2 years later, he said he was adding money back then. Sure.


Added: The period that continued through the "lost decade" was also fine for the above buy and hold strategy, even one that used TSM as the equities.

2010 14.61%
2009 10.46%
2008 -6.82%
2007 8.18%
2006 8.09%
2005 3.78%
2004 7.05%
2003 13.96%

Sensible buy-and-hold portfolios that included some amounts of international, or slice/dice, did even better. All one needed was a reasonable plan and rebalancing for a successful, long ride. While a 60/40 portfolio had a bumpier ride, it also had a good run if held for the entire period.
Last edited by Roy on Sun Jul 31, 2011 5:29 am, edited 2 times in total.
User avatar
thewatcher
Posts: 120
Joined: Sun Jun 19, 2011 3:22 am
Location: UK

Post by thewatcher »

letsgobobby wrote;
Stocks fell to 1996 levels by the nadir in 2009. Pe10 correctly signaled that large caps were a bad investment beginning in the mid 90 s. I didn't miss a darn thing but volatility while the buy and holder missed 7% annual gains in bonds. Ignoring the message pe10 was sending in 1995-96 cost an imvestor dearly.
Well, yeah, if you cherry-pick the absolute nadir! How about 6 months or a couple of years later?

Anyone investing in stocks does so with the knowledge that they are volatile and that this volatility is the risk you take to capture the long-term higher expected return. So saying, look stocks are volatile and you would have lost money at the lowest point of the troughs is hardly the same as saying you shouldn't invest in stocks or even that using PE10 or any other metric is justified.

The other thing with PE10 is that if you are using it to compare value vs bonds then you must also quote the bond yield at the relevent time. For example, investing in equities when the cyclically adjusted earnings yield (inverse of PE10) is 4.3% is very different when bonds are yielding 3% than when they yield 6%.

This chart give CAEP vs long bonds and the spread between them;

http://imageshack.us/photo/my-images/19 ... 01105.png/

TW
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

Roy using 40/60 is disingenuous. Using 80/20 or 70/30 is a better comparator because aggressive pe10 tracking can only be rationally used by those with long time horizons, those who would otherwise be counseled to be largely in stocks.

Bond yields might matter but at historically high pe10 bond yields have been both very high and very low and in both cases pe10 was equally predictive. Maybe when pe10 is high and yields are high bonds are the obvious choice; but when pe10 is high and yields are low the conclusion should be that returns in general will be low.

I can't say why shiller didn't follow his own model. In march 09 pe10 was around 12 or 1/4 below its long term mean. There was no other time in us history that stocks had fallen 50% from their highs and had below average pe10 and hadn't delivered excellent long term returns. It was the best buying opportunity in at least 15 years. It was psychologically hard but nonetheless quite rewarding.
User avatar
NoRoboGuy
Posts: 733
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Post by NoRoboGuy »

Based on the fact that PE 10 seeks to approach long term allocation from a valuation standpoint versus a timing one, I think it is reasonable to consider PE 10 potentially useful for adjusting allocations.

While I would consider PE 10 to be a 'value timing strategy' as opposed to a 'market timing strategy,' markets can and do stay overvalued or undervalued for extended periods. For this reason one should be skeptical of PE 10 having high enough predictive value to outweigh simple annual rebalancing to a fixed allocation.

Edit: Agree that PE 10 is of value at extreme valuation periods, infrequent as they are.
There is no free lunch.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Post by Roy »

letsgobobby wrote:Roy using 40/60 is disingenuous. Using 80/20 or 70/30 is a better comparator because aggressive pe10 tracking can only be rationally used by those with long time horizons, those who would otherwise be counseled to be largely in stocks.

I can't say why shiller didn't follow his own model. In march 09 pe10 was around 12 or 1/4 below its long term mean. There was no other time in us history that stocks had fallen 50% from their highs and had below average pe10 and hadn't delivered excellent long term returns.
Disingenuous? As pointed out, you have cherry-picked and delimited throughout. I used the most basic combination you could get with no stock diversification beyond domestic TSM (your "bad investment" Large Caps) throughout the entire period embraced by your opinion. It shows that a dead-simple strategy that accounted for inevitable market Bears worked well. Just because you don't like the outcome, doesn't invalidate the data or the buy-and-hold concept. Many folks seem to overestimate their taste for risk. Maybe that is the lesson that needs be learned.

Note that even the time-honored 60/40 did fine, if it used Treasuries as the fixed component throughout that 1995-2010 period—taking risk primarily on the equity side. And it did so with nothing more than rebalancing.

Adding International diversification improved things further. International Slice/Dicers did even better (as Larry says all the time about the "lost decade"). The main point is really about incorporating future market declines into one's plan and seeing it through. A broad-based claim that the buy-and-hold approach doesn't work is absurd.

The point with Shiller was that even he—the man who invented this—could not pull a trigger on what people now see, ex ante, was the "right move"; how much less a normal investor, who would then also have to create some arbitrary system for implementing what PE/10 was telling him? Failure to account for the added stress in these of hypothetical implementations is a big mistake.
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

Roy wrote:
letsgobobby wrote:Roy using 40/60 is disingenuous. Using 80/20 or 70/30 is a better comparator because aggressive pe10 tracking can only be rationally used by those with long time horizons, those who would otherwise be counseled to be largely in stocks.

I can't say why shiller didn't follow his own model. In march 09 pe10 was around 12 or 1/4 below its long term mean. There was no other time in us history that stocks had fallen 50% from their highs and had below average pe10 and hadn't delivered excellent long term returns.
Disingenuous? As pointed out, you have cherry-picked and delimited throughout. I used the most basic combination you could get with no stock diversification beyond domestic TSM (your "bad investment" Large Caps) throughout the entire period embraced by your opinion. It shows that a dead-simple strategy that accounted for inevitable market Bears worked well. Just because you don't like the outcome, doesn't invalidate the data or the buy-and-hold concept. Many folks seem to overestimate their taste for risk. Maybe that is the lesson that needs be learned.

Note that even the time-honored 60/40 did fine, if it used Treasuries as the fixed component throughout that 1995-2010 period—taking risk primarily on the equity side. And it did so with nothing more than rebalancing.

Adding International diversification improved things further. International Slice/Dicers did even better (as Larry says all the time about the "lost decade"). The main point is really about incorporating future market declines into one's plan and seeing it through. A broad-based claim that the buy-and-hold approach doesn't work is absurd.

The point with Shiller was that even he—the man who invented this—could not pull a trigger on what people now see, ex ante, was the "right move"; how much less a normal investor, who would then also have to create some arbitrary system for implementing what PE/10 was telling him? Failure to account for the added stress in these of hypothetical implementations is a big mistake.
it is disingenuous to compare PE10 timing to a 40/60 policy because no investor for whom a 40/60 AA represents the correct need to take risk would embrace aggressive PE10 timing. Only a young investor could, for example, be 90% TSM if PE10 were 5. A 70 year old could not take such risk. Most young investors are advised to be 80/20 or 70/30, not 40/60. And of course we are only talking about the TSM portion of one's portfolio; PE10 of the S&P500 says nothing about the prospects of bonds or international stocks or anything else for that matter.

There is no broad-based claim that B&H doesn't work. I have certainly said no such thing. What I've said is that PE10 tracking grants higher returns with less volatility - it is more efficient than B&H investing as it relates to the perhaps artificial dichotomy of "S&P/other" portfolios where "other" may be bonds or international stocks or whatever other asset class you wish. S&P500 was dead money since the mid-90s through 2009. PE10 told you that. I followed it and so did some others and why Shiller didn't and why it was hard for you or anyone else is irrelevant to me.
Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Post by Roy »

letsgobobby wrote:I didn't miss a darn thing but volatility while the buy and holder missed 7% annual gains in bonds. Ignoring the message pe10 was sending in 1995-96 cost an imvestor dearly.
Bobby, as has been shown—from 1995 on—none of that was true for a person whose buy and hold portfolio was well-designed to meet market downturns, be they 40/60 or 60/40. It did not cost them dearly. And certainly such a buy-and-holder did not miss out on the gains in bonds. For those who chose to be massively invested in stocks, or advised to be such, that was indeed their problem. And whether anyone should "embrace aggressive PE 10 timing" is questionable for a number of reasons. Finally, regarding what you or others on investing forum boards claim to have done is irrelevant to me.
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

Just FYI, PE10 is down to 20.7 as of today's close.

http://www.multpl.com/
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
DP
Posts: 659
Joined: Thu Apr 17, 2008 5:19 pm

Post by DP »

Hi,
I've seen it said that a bear market is a decline in valuations (PE) as much or more so than a decline in price. In looking at the chart the downtrend in PE from the 2000 highs is very clear. Quite different than a chart of the price: http://finance.yahoo.com/echarts?s=%5EG ... ff;source=;

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

Post by Rodc »

Bongleur wrote:>Another perfectly rational person could look at the exact same situation and make an to their AA based on a fixed willingness to take on risk.
>

You think Traders are "perfectly rationale?" ;0
In either case person A sold stocks because he estimated expected returns were now higher, and person B bought stocks because he estimated expected risk was now lower. That was the set up. Neither is more or less rational than the other, or more active or less active.

I on the other hand took the average and neither bought or sold so I in a contest of Bogleheadedness, I win. :)
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.
Bongleur
Posts: 2276
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur »

the "Crestmont PE Ratio" methodology claims to have an improved predictive value. They present reasons why PE & PE10 can give bad signals at key cycle turning points:

http://advisorperspectives.com/dshort/u ... -Ratio.php

I have asked Wade to take a look at their article & see what he thinks:

http://www.bogleheads.org/forum/viewtop ... 93#1135793
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
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

pwm112 wrote:Did it ever occur to you that, even if earnings skyrocketed, PE10 would still look poor because it would take 10 years to flush out the old earnings data?
Yes. That is actually the point.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
mosu
Posts: 170
Joined: Wed Jan 26, 2011 4:54 pm

Post by mosu »

.....
Last edited by mosu on Thu Mar 12, 2015 9:45 pm, edited 1 time in total.
User avatar
tadamsmar
Posts: 9077
Joined: Mon May 07, 2007 12:33 pm

Post by tadamsmar »

Consider this plan for picking the stocks to invest in:

1. Put an equal amount of weight on earnings from one month 10 years ago (back in July 2001 as of now) as you do on earnings from last month.

2. Put zero weight on every other piece of available information about the company except 10 years of trailing earnings. But wait, you can quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

if the market falls by half in the next six months, PE10 will be roughly 10 or maybe 12, at which time it will be signaling buy per this model.
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

pwm112 wrote:
fredflinstone wrote:
pwm112 wrote:Did it ever occur to you that, even if earnings skyrocketed, PE10 would still look poor because it would take 10 years to flush out the old earnings data?
Yes. That is actually the point.
Then why are you posting the day by day changes to PE10? As I said, if you believe in the validity of 10 year averaged earnings, then you shouldn't be buying stocks for a long time.

There's no need to check the oven because the cake hasn't been in there long enough to be anywhere close to finished.
wow, I didn't realize I was posting daily PE10 numbers. can you please point me to where those were posted?
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

letsgobobby wrote:if the market falls by half in the next six months, PE10 will be roughly 10 or maybe 12, at which time it will be signaling buy per this model.
correct; I think that points to the limitations of the model. Surely, the expected real returns on bonds should be taken into account as well. To see why, consider this simple thought experiment:

Scenario A: 30-year TIPS yield is 1%; PE10 ratio is 10
Scenario B: 30-year TIPS yield is 10%; PE10 ratio is 10


Assume for the sake of argument and simplicity that all your assets are in a tax-deferred retirement account.

If those are the two choices, would you hold the same % in stocks in Scenario A as in Scenario B?
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

if the TIPS yield is 10% I would assume that is because people believe the US will default. I'm not sure what I would do in that situation!
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

OK, let's try again with a more realistic example:

Scenario A: 30-year TIPS yield is 1%; PE10 ratio is 20
Scenario B: 30-year TIPS yield is 3%; PE10 ratio is 20

to my mind, stocks look a lot less attractive in Scenario B (equity risk premium of 2 percent) than in Scenario A (equity risk premium of 4 percent). Am I wrong?
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
Sammy_M
Posts: 1896
Joined: Sun Nov 25, 2007 8:30 am

Post by Sammy_M »

fredflinstone wrote:OK, let's try again with a more realistic example:

Scenario A: 30-year TIPS yield is 1%; PE10 ratio is 20
Scenario B: 30-year TIPS yield is 3%; PE10 ratio is 20

to my mind, stocks look a lot less attractive in Scenario B (equity risk premium of 2 percent) than in Scenario A (equity risk premium of 4 percent). Am I wrong?
I don't get why you must link the two.

A.1. PE10 is 20. That's a pretty normal based on historical standards. Tells me to stick to my normal stock allocation. I agree PE10 at 40 would be another story.

A.2. 30-year TIPS yield is 1%. That's pretty low based on historical standards. Tells me rolling 90-day T-Bills might be a better strategy. Both protect me against inflation.

B.1. PE10 is 20. Same as above.

B.2. 30-year TIPS yield is 3%. That's pretty decent based on historical standards. Tells me it's probably smart to stop rolling treasuries and lock in. If 4%, it tells me to take some away from equities since 4% is good enough for me to meet my financial objectives.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc »

fredflinstone wrote:OK, let's try again with a more realistic example:

Scenario A: 30-year TIPS yield is 1%; PE10 ratio is 20
Scenario B: 30-year TIPS yield is 3%; PE10 ratio is 20

to my mind, stocks look a lot less attractive in Scenario B (equity risk premium of 2 percent) than in Scenario A (equity risk premium of 4 percent). Am I wrong?
Whether you make the numbers realistic or not the point is valid.

Surely one should not set up an allocation between 2 or 3 or more assets by looking at only one asset.
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
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

Sammy_M wrote:
fredflinstone wrote:OK, let's try again with a more realistic example:

Scenario A: 30-year TIPS yield is 1%; PE10 ratio is 20
Scenario B: 30-year TIPS yield is 3%; PE10 ratio is 20

to my mind, stocks look a lot less attractive in Scenario B (equity risk premium of 2 percent) than in Scenario A (equity risk premium of 4 percent). Am I wrong?
I don't get why you must link the two.

A.1. PE10 is 20. That's a pretty normal based on historical standards. Tells me to stick to my normal stock allocation. I agree PE10 at 40 would be another story.

A.2. 30-year TIPS yield is 1%. That's pretty low based on historical standards. Tells me rolling 90-day T-Bills might be a better strategy. Both protect me against inflation.

B.1. PE10 is 20. Same as above.

B.2. 30-year TIPS yield is 3%. That's pretty decent based on historical standards. Tells me it's probably smart to stop rolling treasuries and lock in. If 4%, it tells me to take some away from equities since 4% is good enough for me to meet my financial objectives.
Interesting. I'd rather have more money in bonds in Scenario B than in Scenario A. Why? Because the opportunity cost of investing in stocks is three times higher in Scenario B than in Scenario A.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
jeffyscott
Posts: 9139
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Post by jeffyscott »

tadamsmar wrote:Consider this plan for picking the stocks to invest in:

1. Put an equal amount of weight on earnings from one month 10 years ago (back in July 2001 as of now) as you do on earnings from last month.

2. Put zero weight on every other piece of available information about the company except 10 years of trailing earnings. But wait, you can quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?
Consider this plan for picking the stocks to invest in:

1. Put zero weight on every piece of available information about the company. But wait, you can't quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?

:)
The two greatest enemies of the equity fund investor are expenses and emotions. ― John C. Bogle
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

tadamsmar wrote:Consider this plan for picking the stocks to invest in:

1. Put an equal amount of weight on earnings from one month 10 years ago (back in July 2001 as of now) as you do on earnings from last month.

2. Put zero weight on every other piece of available information about the company except 10 years of trailing earnings. But wait, you can quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?
you raise an interesting point. Perhaps a weighted average earnings would perform better than unweighted, i.e., recent earnings should count more. Professor Pfau: another idea for you to research! There is something to be said for keeping things simple, though.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
Sammy_M
Posts: 1896
Joined: Sun Nov 25, 2007 8:30 am

Post by Sammy_M »

fredflinstone wrote:Interesting. I'd rather have more money in bonds in Scenario B than in Scenario A. Why? Because the opportunity cost of investing in stocks is three times higher in Scenario B than in Scenario A.
Equities are the primary growth vehicle in a porfolio. The role of bonds (TIPS, short bonds) is to preserve value on an inflation adjusted basis. If TIPS yields are very high, then that changes the story, but those times are rare.

I believe no one really knows which way stocks are headed. When stocks get to extreme valuation levels, one way or the other, perhaps the odds are stacked in a certain direction and we should be willing to alter our allocations accordingly. I'll grant that.

Right now we have PE10 levels in a relatively normal range. Nobody knows where stocks are headed. So stay the course.

Right now we have TIPS yields at a very low level. Short bonds offer a very viable alternative without the real interest rate risk.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc »

fredflinstone wrote:
tadamsmar wrote:Consider this plan for picking the stocks to invest in:

1. Put an equal amount of weight on earnings from one month 10 years ago (back in July 2001 as of now) as you do on earnings from last month.

2. Put zero weight on every other piece of available information about the company except 10 years of trailing earnings. But wait, you can quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?
you raise an interesting point. Perhaps a weighted average earnings would perform better than unweighted, i.e., recent earnings should count more. Professor Pfau: another idea for you to research! There is something to be said for keeping things simple, though.
Since over long periods earnings tend to increase, and the goal is to smooth out the ups and downs, a more appropriate model is to fit a curve (if you think earning, like price, tend to increase at some yearly percentage rate, plus of minus noise, the proper curve is exponential)), rather than taking an average.

Consider a simple case of increasing linearly, plus noise. You have 11 regularly space data points. The average is then an estimate of the value at point #6 (middle of the data series). If E is generally increasing linearly then P/E10 would be an estimate of the value 5 years ago, not today. If growth is exponential it would more recent, but still well in the past.

Using an exponential or even linear fit to smooth noise makes rather more sense, but seems not to be used.
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
HomerJ
Posts: 15148
Joined: Fri Jun 06, 2008 12:50 pm

Post by HomerJ »

letsgobobby wrote:S&P500 was dead money since the mid-90s through 2009. PE10 told you that. I followed it and so did some others and why Shiller didn't and why it was hard for you or anyone else is irrelevant to me.
Maybe if you retired in the mid-90s this is true... Your hypothetical "young investor" who cares about your PE10 model certainly didn't drop a bunch of money in the S&P 500 in the mid 90s and never touch it again...

Continuing to put new money in through all the valleys and peaks paints a very different picture than just saying oh, the S&P 500 is the exact same place it was 12 years ago.

And I like how you stopped in 2009... Talk about cherry-picking.

PE10 may offer some value, but a simple 50/50 portfolio does just fine.
User avatar
DRiP Guy
Posts: 2241
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy »

jeffyscott wrote: Consider this plan for picking the stocks to invest in:

1. Put zero weight on every piece of available information about the company. But wait, you can't quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?

:)
Yes, it certainly does, if the average return of all the companies you buy is a return you can live with, and you intend to simply buy a little piece of all of them.
User avatar
jeffyscott
Posts: 9139
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Post by jeffyscott »

DRiP Guy wrote:
jeffyscott wrote: Consider this plan for picking the stocks to invest in:

1. Put zero weight on every piece of available information about the company. But wait, you can't quite do that. Ignore everything about the company except whether it has gone out of existence.

Anyone think that is a good investing plan?

:)
Yes, it certainly does, if the average return of all the companies you buy is a return you can live with, and you intend to simply buy a little piece of all of them.
Well you can also buy a little piece of them all, while also allowing P/E10 to influence how much of your total assets you put in the stock market.

I've allowed valuations in general to influence me, such that I began this latest decline with 38% in stocks rather than what would be my sort of "neutral" position of 50%.
The two greatest enemies of the equity fund investor are expenses and emotions. ― John C. Bogle
User avatar
DRiP Guy
Posts: 2241
Joined: Tue Feb 20, 2007 4:54 pm

Post by DRiP Guy »

jeffyscott wrote:I've allowed valuations in general to influence me, such that I began this latest decline with 38% in stocks rather than what would be my sort of "neutral" position of 50%.
Different strokes!

8-)
letsgobobby
Posts: 12073
Joined: Fri Sep 18, 2009 1:10 am

Post by letsgobobby »

rrosenkoetter wrote:
letsgobobby wrote:S&P500 was dead money since the mid-90s through 2009. PE10 told you that. I followed it and so did some others and why Shiller didn't and why it was hard for you or anyone else is irrelevant to me.
Maybe if you retired in the mid-90s this is true... Your hypothetical "young investor" who cares about your PE10 model certainly didn't drop a bunch of money in the S&P 500 in the mid 90s and never touch it again...

Continuing to put new money in through all the valleys and peaks paints a very different picture than just saying oh, the S&P 500 is the exact same place it was 12 years ago.

And I like how you stopped in 2009... Talk about cherry-picking.

PE10 may offer some value, but a simple 50/50 portfolio does just fine.
I didn't cherry pick 2009. I picked 2009 because that's when PE10 finally got below it's long term mean, and ultimately reached an attractive level. From that point forward it doubled (well, not any longer after today).

FWIW, today the PE10 will probably be around 19.8 or something. Looking at this graph, I think anything 12-18 or so predicts pretty much nothing. At which point - another -9% or so - I will up my AA from 55/45 to 60/40.

http://www.bogleheads.org/forum/viewtopic.php?p=746593
Bongleur
Posts: 2276
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur »

http://hussmanfunds.com/wmc/wmc110815.htm

August 15, 2011
Two One-Way Lanes on the Highway to Hell

John P. Hussman, Ph.D.
SNIP

It is important to recognize that the S&P 500 is presently only about 13% below its April peak, and the word "only" deserves emphasis. Our valuation impressions align fairly well with those of Jeremy Grantham at GMO, who puts fair value for the S&P 500 "no higher than 950" - a level that we would still associate with prospective 10-year total returns only modestly above 8% annually. I would consider investors to be very fortunate if the market does not substantially breach that level in the coming 12-18 months. Wall Street continues its servile attachment to forward operating earnings, seemingly unconscious that the perceived "norms" for the resulting P/E are artifacts of a bubble period. The fact is that historical periods of overvaluation and poor subsequent long-term returns correspond to forward operating P/E multiples anywhere above 12, while secular buying opportunities such as 1950, 1974 and 1982 map to forward operating multiples of only 5 or 6 (based on the strong correlation but downward-biased level of forward operating P/E ratios, when compared with multiples based on normalized earnings - see Chutes and Ladders for a graphic).

The composite of recession warning evidence we observe here (year-over-year GDP growth of just 1.6%, S&P 500 below its level of 6 months earlier, widening credit spreads versus 6 months earlier, yield curve spread at 2.2%, Purchasing Managers Index at 50.9, year-over-year nonfarm payroll growth below 1%) falls into a Recession Warning Composite that has been observed in every recession since 1950, and has never been observed except during or immediately preceding a recession.
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
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

the difference between TIPS yields and the equity yield is fairly large. I feel comfortable overweighting equities here. And, yes, I agree that a recession is quite possible.

Look at it this way. Chevron is trading a PE1 of 8.4. Even if its earnings get cut in half, it will still have an earnings yield (1/PE10) of about 6 percent. By comparison, intermediate-duration treasuries have a negative expected yield. Which would you rather own?
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
Bongleur
Posts: 2276
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur »

I'd prefer to buy it AFTER the price gets cut in half...
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
docneil88
Posts: 927
Joined: Mon Apr 30, 2007 3:39 pm
Location: Taxable

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

Post by docneil88 »

Image

The "Bonds" data above refers to long-term government bonds (for the US from 1926 onward, it refers to the Ibbotson Associates Long Government Bond Index). Looking at the above chart, the geometric mean real return for stocks in the US in the 20th century was 6.7% vs. 1.6% for long-term US government bonds, for a 20th century equity risk premium (ERP) of 5.1% (= 6.7 - 1.6).

Today, 8/14/11, with the S&P 500 at 1,179, the inverse of PE10 is 4.9%, which I consider to be about the expected 20-year real return. The 20-year TIPS real yield is 0.6%. So the current expected ERP over the next 20 years is 4.3%, which is the difference between 4.9% and 0.6%. That 4.3% is 0.8 of a percentage point worse than the 20th century average. So, based on this analysis, the expected ERP for US stocks is not especially attractive.

For perspective, when the inverse of PE10 hit 7.5% on March 1, 2009, very near the bear-market bottom, the 20-year TIPS real yield was 2.5%, for an ERP of 5.0%, which just 0.1 below the 20th century ERP. That did turn out to be a great buying opportunity. (Sources: http://www.multpl.com/table?f=m , http://online.wsj.com/mdc/public/page/2 ... stcalendar .) Best, Neil (Edit above in dark blue.)
Last edited by docneil88 on Mon Aug 15, 2011 2:56 pm, edited 1 time in total.
User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

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

Post by bob90245 »

docneil88 wrote:Image

The "Bonds" data above refers to long-term government bonds (the Ibbotson Associates Long Government Bond Index from 1926 forward). Looking at the above chart, the geometric mean real return for stocks in the US in the 20th century was 6.7% vs. 1.6% for long-term US government bonds, for a 20th century equity risk premium (ERP) of 5.1% (= 6.7 - 1.6).
I wonder if using long-term US govt. bonds is the commonly accepted asset class as "risk free" in the ERP calculation. Especially since annual volatility is relatively higher than other terms. Given it's lower volatility, I thought short-term treasury bills would be the appropriate asset class as "risk free" in the ERP calculation.
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

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

Post by Rodc »

docneil88 wrote:Image

The "Bonds" data above refers to long-term government bonds (the Ibbotson Associates Long Government Bond Index from 1926 forward). Looking at the above chart, the geometric mean real return for stocks in the US in the 20th century was 6.7% vs. 1.6% for long-term US government bonds, for a 20th century equity risk premium (ERP) of 5.1% (= 6.7 - 1.6).

Today, 8/14/11, with the S&P 500 at 1,179, the inverse of PE10 is 4.9%, which I consider to be about the expected 20-year real return. The 20-year TIPS real yield is 0.6%. So the current expected ERP over the next 20 years is 4.3%, which is the difference between 4.9% and 0.6%. That 4.3% is 0.8 of a percentage point worse than the 20th century average. So, based on this analysis, the expected ERP for US stocks is not especially attractive.

For perspective, when the inverse of PE10 hit 7.5% on March 1, 2009, very near the bear-market bottom, the 20-year TIPS real yield was 2.5%, for an ERP of 5.0%, which just 0.1 below the 20th century ERP. That did turn out to be a great buying opportunity. (Sources: http://www.multpl.com/table?f=m , http://online.wsj.com/mdc/public/page/2 ... stcalendar .) Best, Neil
I'm not at all sure that when using a simple analysis that shows a fraction of one percent difference that the difference is statistically significant.
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
docneil88
Posts: 927
Joined: Mon Apr 30, 2007 3:39 pm
Location: Taxable

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

Post by docneil88 »

bob90245 wrote:I wonder if using long-term US govt. bonds is the commonly accepted asset class as "risk free" in the ERP [Equity Risk Premium] calculation. Especially since annual volatility is relatively higher than other terms. Given it's lower volatility, I thought short-term treasury bills would be the appropriate asset class as "risk free" in the ERP calculation.
You're right Bob, T-bills are generally used in ERP calculations. So let's use T-Bills in the above table: 20th century ERP = (6.7% [20th century real returns on US equities] - 0.9% [20th century real returns on US T-bills]) = 5.8.

Today, 8/16/11, with the S&P 500 at 1,193, EP10 (the inverse of PE10) is 4.9% and 3-month T-Bill yields 0.0. So, current ERP = (4.9% - 0.0%) = 4.9%. That 4.9% is 0.9 of a percentage point worse than the 20th century average. So, based on this analysis, the expected ERP for US stocks is not especially attractive.

And now, I'm wondering if things aren't worse than the above analysis indicates. If we want some indication of expected 10-year real return for stocks, shouldn't we also factor in a reversion to the mean of 0.9% of the real T-Bill rate over 10 years? If so, then EP10 would have to increase 0.9% points (or 18% [5.8/4.9 - 1] higher than it is now) by the end of that 10-years just to maintain the current ERP. That's a drag of 1.7% annualized per year on returns over that 10 years if the real interest rate on T-Bills revert to its mean. This would throw cold water on the common view that low real interest rates help to justify higher market PEs. Comments welcome. Best, Neil
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

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

Post by fredflinstone »

docneil88 wrote:
bob90245 wrote:I wonder if using long-term US govt. bonds is the commonly accepted asset class as "risk free" in the ERP [Equity Risk Premium] calculation. Especially since annual volatility is relatively higher than other terms. Given it's lower volatility, I thought short-term treasury bills would be the appropriate asset class as "risk free" in the ERP calculation.
You're right Bob, T-bills are generally used in ERP calculations. So let's use T-Bills in the above table: 20th century ERP = (6.7% [20th century real returns on US equities] - 0.9% [20th century real returns on US T-bills]) = 5.8.

Today, 8/16/11, with the S&P 500 at 1,193, EP10 (the inverse of PE10) is 4.9% and 3-month T-Bill yields 0.0. So, current ERP = (4.9% - 0.0%) = 4.9%. That 4.9% is 0.9 of a percentage point worse than the 20th century average. So, based on this analysis, the expected ERP for US stocks is not especially attractive.

And now, I'm wondering if things aren't worse than the above analysis indicates. If we want some indication of expected 10-year real return for stocks, shouldn't we also factor in a reversion to the mean of 0.9% of the real T-Bill rate over 10 years? If so, then EP10 would have to increase 0.9% points (or 18% [5.8/4.9 - 1] higher than it is now) by the end of that 10-years just to maintain the current ERP. That's a drag of 1.7% annualized per year on returns over that 10 years if the real interest rate on T-Bills revert to its mean. This would throw cold water on the common view that low real interest rates help to justify higher market PEs. Comments welcome. Best, Neil
I don't understand why one would want to compare the equity yield, which is generally considered a real yield, with a nominal yield. The apples-to-apples comparison is equity yield vs TIPS yield.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
docneil88
Posts: 927
Joined: Mon Apr 30, 2007 3:39 pm
Location: Taxable

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

Post by docneil88 »

fredflinstone wrote:I don't understand why one would want to compare the equity yield, which is generally considered a real yield, with a nominal yield. The apples-to-apples comparison is equity yield vs TIPS yield.
Oops. You're right Fred. I remember you like to use the formula EP10 - the real yield on 5-year TIPS = Equity Risk Premium. (EP10 = 1/PE10.)

Using that formula today 8/16/11, ERP = (4.9% - - 0.9%) = 5.8%. And you think that ERP is worthy of a high equity allocation. But aren't you concerned that real yields will revert to their mean of 2% over the next 5-10 years, thus becoming a big drag on the expected real return for equities over the next 5-10 years. If so, wouldn't this throw cold water on the common view that historically low real interest rates help to justify higher market PEs. Best, Neil
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

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

Post by fredflinstone »

docneil88 wrote:
fredflinstone wrote:I don't understand why one would want to compare the equity yield, which is generally considered a real yield, with a nominal yield. The apples-to-apples comparison is equity yield vs TIPS yield.
Oops. You're right Fred. I remember you like to use the formula EP10 - the real yield on 5-year TIPS = Equity Risk Premium. (EP10 = 1/PE10.)

Using that formula today 8/16/11, ERP = (4.9% - - 0.9%) = 5.8%. And you think that ERP is worthy of a high equity allocation. But aren't you concerned that real yields will revert to their mean of 2% over the next 5-10 years, thus becoming a big drag on the expected real return for equities over the next 5-10 years. If so, wouldn't this throw cold water on the common view that historically low real interest rates help to justify higher market PEs. Best, Neil
You are saying the ERP is likely to shrink. I agree. For that to happen, the equity yield must decline or the TIPS yield must rise or some combination of both must occur. Let's consider each possibility in turn.

1) Decline in the equity yield

Since E10 is robust to short-term changes in earnings, the most plausible way for EP10 to fall is via an increase in P. In other words, if you think the equity earnings yield is likely to decline from the current level, that is tantamount to saying that you think the stock market is likely to rise from current levels. This is an argument for overweighting stocks now (i.e. underweighting bonds).

2) Increase in TIPS yield

This is just another way of saying that TIPS prices will decline. Most likely the prices of bonds that aren't indexed for inflation also will decline (i.e. since an increase in TIPS yields is almost certainly going to be associated with an increase in nominal yields). This is an argument for underweighting bonds now (i.e. overweighting equities).

If and when the equity risk premium declines, that will probably be great news for those of us who are current overweight equities. Once it happens, I will reduce my allocation to equities and increase my allocation to bonds.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
docneil88
Posts: 927
Joined: Mon Apr 30, 2007 3:39 pm
Location: Taxable

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

Post by docneil88 »

fredflinstone wrote:2) Increase in TIPS yield

This is just another way of saying that TIPS prices will decline. Most likely the prices of bonds that aren't indexed for inflation also will decline (i.e. since an increase in TIPS yields is almost certainly going to be associated with an increase in nominal yields). This is an argument for underweighting bonds now (i.e. overweighting equities).
Hi Fred, I agree that's an argument for underweighting bonds, but I don't agree that it's an argument for overweighting stocks. There are some big alternatives such as cash, gold, and real estate with a 30-year mortgage.

When interest rates go up, that usually hurts stocks unless the earnings can increase faster than the interest rates go up. Early in an economic cycle, that's quite possible, but later it is less likely. But it's very hard to know when the cycle will turn. Given that we are near historic lows with real interest rates, we are a lot further away from the average cyclical interest rate peak. So, the potential headwind to equites from such an increase in interest rates would be unusually large. Probably nothing to worry about much in the next couple years, but it's something to keep in mind as a long-term investor. Best, Neil
Last edited by docneil88 on Wed Aug 17, 2011 12:06 am, edited 4 times in total.
User avatar
Topic Author
fredflinstone
Posts: 2718
Joined: Mon Mar 29, 2010 7:35 am
Location: Bedrock

Post by fredflinstone »

Thanks, that is a valid point. As it happens, I have a small holding in gold/silver, but in terms of my valuation-based market timing model I only think about stocks and bonds.
Stocks 28 / Gold 23 / Long-term US treasuries 19 / Cash (mainly CDs) 22 / TIPS 8
User avatar
docneil88
Posts: 927
Joined: Mon Apr 30, 2007 3:39 pm
Location: Taxable

Post by docneil88 »

Thanks Fred. Given my thoughts above, I'm now inclined to give more weight to current & historical PE10 than to the current & the historical equity risk premium as a way of gauging whether the stock market is over-valued or under-valued. Best, Neil
User avatar
Liquid
Posts: 301
Joined: Sat Aug 06, 2011 10:17 am

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

Post by Liquid »

Valuation matters, but for many is probably playing with fire.
User avatar
wade
Posts: 544
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 »

Dear all,

The article on which this thread is based (after substantial revisions and a new title) has been finally published:

http://www.tandfonline.com/doi/abs/10.1 ... 011.648317

Thank you all, as I included in the Acknowledgements section of the article:


Though market-timing strategies are specifically
contrary to John Bogle’s investment philosophy,
the author is extremely indebted to suggestions and
reading recommendations provided by countless
users in the threads ‘Any Studies on Long-term
Market Timing?’ and ‘Valuation-based market
timing with PE10 can improve returns?’ at the
Bogleheads Forum.
User avatar
CaliJim
Posts: 3050
Joined: Sun Feb 28, 2010 8:47 pm
Location: California, near the beach

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

Post by CaliJim »

wade wrote:Dear all,
The article on which this thread is based (after substantial revisions and a new title) has been finally published:
http://www.tandfonline.com/doi/abs/10.1 ... 011.648317
Thank you all, as I included in the Acknowledgements section of the article:

Though market-timing strategies are specifically contrary to John Bogle’s investment philosophy,the author is extremely indebted to suggestions and reading recommendations provided by countless users in the threads ‘Any Studies on Long-term Market Timing?’ and ‘Valuation-based market timing with PE10 can improve returns?’ at the Bogleheads Forum.
Excellent paper. I especially like your analysis of the various middle ground strategies: MT 10-90 thru MT 40-60.

I'm not sure I'm ready to go this route - but you have certainly made the data speak very clearly.
-calijim- | | For more info, click this Wiki
Post Reply