Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
siamond
Posts: 4239
Joined: Mon May 28, 2012 5:50 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by siamond » Mon Jun 26, 2017 12:07 pm

XdUzHa3NtSeIkBkIGPVn wrote:
siamond wrote:
Top99% wrote:I have read most of Larry's books and have tremendous respect for him but I am not willing to bet my entire financial future on a Larry style portfolio. So, I need some total market and alternatives mixed in to sleep at night.
Same here. It is true that said portfolio backtests surprisingly well (at least in the US - I tried and showed outcomes in a past thread), but the reasoning lacks a strong rationale, its past performance might be more luck than anything (i.e. a fortunate confluence of events), and today's situation seems to speak especially against it. I would be mildly curious to see if it would have worked well for investors located in other countries, which I don't believe has been demonstrated, but would certainly NOT bet my life's savings on such approach. I do use an SCV and an EM tilt, but added to a TSM + Total Int'l backbone...
It has been demonstrated on a large enough sample of US equities that it's statistically very unlikely to be "chance". The value premium also has been found to exist on every country's stock market that we have data on.
Yes, the (small) value premium comes with strong historical evidence, domestically (US) and internationally, I fully agree with you here.

This wasn't my point though. My point was about the portfolio combination of a LOT of bonds (e.g. 70%) and some especially erratic asset classes (e.g. SCV, or EM - Larry sometimes uses Int'l Small too), the 'black swan' effect that Larry seems to bet on to provide a proper engine of growth in the portfolio on the long-term, to counter-balance the low expected return and inflation risk of bonds. Well, I don't know how one can bet on black swans, this is a severe contradiction in itself.

Again, I would be intellectually curious to see how such combination worked in other countries, but in any case, the theoretical rationale behind such portfolio seems terribly weak to me. It's a creative idea, mind you, but I would never, never, make such a bet.

User avatar
Top99%
Posts: 334
Joined: Sat Apr 22, 2017 9:30 am
Location: Austin, TX

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Top99% » Mon Jun 26, 2017 12:29 pm

XdUzHa3NtSeIkBkIGPVn wrote:[It has been demonstrated on a large enough sample of US equities that it's statistically very unlikely to be "chance". The value premium also has been found to exist on every country's stock market that we have data on..
I am definitely convinced there has been a value premium and I *think* there will be one in the future but I have my doubts it will be large enough going forward to allow a 30/70 Larry portfolio to offer comparable returns to a 60/40 portfolio. That being said, Larry's work has been instrumental in convincing me to have a strong value tilt. As for short term bond/CDs and TIPs, I don't recall the former doing all that well in the 70s net of inflation and the latter haven't been battle tested in my opinion. So, my approach is kind of akin to "Trust But Verify": "Tilt but don't bet the farm". I think all assets are risky so diversification is a must.
Adapt or perish

WhiteMaxima
Posts: 1462
Joined: Thu May 19, 2016 5:04 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by WhiteMaxima » Mon Jun 26, 2017 1:03 pm

PE vs long-term interest rate. I don't see peak yet unless Fed raise interest too much and too fast. 2% interest? I will say no and invest in S&P for 1.8% dividend and enjoy the capital gain. If the interest reaches 4% then I will get worried. Why 4%? Did we talk 4% not 2%?

User avatar
willthrill81
Posts: 6146
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by willthrill81 » Mon Jun 26, 2017 1:21 pm

Da5id wrote:
HomerJ wrote:
Da5id wrote:
HomerJ wrote: I'm 7 years from retirement, and I'm 50/50. In a few years, I may drop to 40/60. I don't care about CAPE. I would be 50/50 right now, regardless of CAPE. It's a terrible measurement tool for timing the market. People like to say it's the "best" of a bunch of bad tools, but it's still a terrible tool. A screwdriver is the "best" tool for driving in a nail, if all you have is a screwdriver and a saw, but it's still a terrible tool for that job.
Just curious, who are the people who say it is useful for timing the market? Not Shiller, its creator. In fact, I'm not aware of many (any?) reputable sources who argue for that. The base you can say about CAPE is that it gives some indication of likely dispersion of returns over next 10 years, and even then it doesn't account for a good part of those returns. So who are these people you refer to?
Every person who suggests changing your AA because valuations are "high".

Not a small, hidden group. I'm surprised you're unaware of them on the Internet and on these boards.
OK, you mean random posters rather than organizations like Vanguard or generally respected figures. Fine. I also see people arguing for gold etc.

I've yet to see a coherent CAPE based market timing scheme that anyone actually can make a good case for, presumably because there isn't such a viable scheme. I see people like Bogle suggesting that returns will be depressed going forward due to CAPE/valuations, but that isn't a call to market timing.
Even people like Michael Kitces and Larry Swedroe say that, statistically, CAPE has been the best predictor of stock returns, and I don't think this specific point is really disputed much by anyone. Larry Swedroe has explicitly stated on this forum that he would recommend that if CAPE is predicting lower market returns in the future that perhaps an investor should take on more stocks so as to get their needed returns.

But there are two issues that critics have pointed out with using CAPE as a forecasting tool. The first is that for the last ~25 years, CAPE has been a poor predictor of market returns, significantly worse than in years prior. CAPE has been above its historical average for that entire period with the exception of a few months in 2008-2009, yet average market returns have only been less than 1% lower than historically (9.34% CAGR from 1992 to now for TSM). So this begs the question as to whether CAPE will be a good predictor of market returns going forward.

The second issue is that even though CAPE has been the best statistical predictor of market returns, that does not necessarily mean that it is good enough predictor to use for any practical reason. At its historic peak predictive power, which is roughly nine years into the future, CAPE has explained about 30% of the variance in market returns (the square of the correlation). That means that at its best, a majority of the variance in market returns has still been unexplained. This also means that there has been wide variation in market returns for a given CAPE value. Kitces found that 30 year returns were only impacted up or down by about 1% based on historic high or low CAPE values, with significant variation in the actual returns. Many would argue that that difference is so small as to not be worth concerning oneself over.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
Top99%
Posts: 334
Joined: Sat Apr 22, 2017 9:30 am
Location: Austin, TX

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Top99% » Mon Jun 26, 2017 1:38 pm

willthrill81 wrote: The second issue is that even though CAPE has been the best statistical predictor of market returns, that does not necessarily mean that it is good enough predictor to use for any practical reason. At its historic peak predictive power, which is roughly nine years into the future, CAPE has explained about 30% of the variance in market returns (the square of the correlation). That means that at its best, a majority of the variance in market returns has still been unexplained. This also means that there has been wide variation in market returns for a given CAPE value. Kitces found that 30 year returns were only impacted up or down by about 1% based on historic high or low CAPE values, with significant variation in the actual returns. Many would argue that that difference is so small as to not be worth concerning oneself over.
I guess for me is all I find actionable from the CAPE being high is to be a bit more cautious with my withdrawal rate and avoid betting the farm on any one factor (Beta, Size,Value, etc.) or asset class. I certainly don't think it can be used to time the market any more than the fact the length of the current bull market in US equities is closing in on a record can be used to time the next bear market.
Adapt or perish

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Mon Jun 26, 2017 1:46 pm

willthrill81 wrote:Kitces found that 30 year returns were only impacted up or down by about 1% based on historic high or low CAPE values, with significant variation in the actual returns.
From your link:
.... the chart below shows the 30-year average annual compound growth rates for every starting year going back to 1871; time periods with starting valuations in the top 20% (a Shiller CAPE above 22) are colored red .... the average of the high valuation zones is only 5.5%, while the lower valuation zones produced a 7.5% average return.
So top quintile CAPEs produced 30 year returns around 1% below average, if the average is 6.5% which sounds about right.

Of course today, we are in the top 1 percentile of CAPE for that time period. There's just 1929 all by itself.

User avatar
XdUzHa3NtSeIkBkIGPVn
Posts: 39
Joined: Thu Jul 16, 2015 4:12 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by XdUzHa3NtSeIkBkIGPVn » Mon Jun 26, 2017 2:20 pm

siamond wrote:
XdUzHa3NtSeIkBkIGPVn wrote:
siamond wrote:One more thing, I don't know what is the current CAPE for SCV, but the PE of the corresponding Vanguard fund (VSIAX) is 43x. While the PE of the S&P 500 Vanguard fund (VFIAX) is 24x. Meaning that the SCV sub-category appears to be way more over-valued than the S&P 500...
The PE of VSIAX is 17.29. It wouldn't be value if it wasn't priced lower than the market average :wink:
Er, no. First, you're quoting the price to *prospective* earnings, and probably got this number from Morningstar here. Morningstar does all sorts of strange things with their PE computation (see more details on the corresponding Wiki page), and the worst sin of all is to use prospective (aka forward) earnings, which is basically analysts guessing future earnings with all their biases enabled...

I was quoting the VSIAX Vanguard number, which can be found here (check the portfolio tab). Vanguard, unsurprisingly, does it much better than Morningstar, and provides price to *past* earnings (trailing twelve months), hence something factual.

Next, as to the fact that the SCV PE is much higher than the S&P 500 PE, well, SCV is made of small-caps while S&P 500 is made of large-caps blend (mixing value and growth). So we have two factors at play here, value and size. Now let's compare apple to apple, using Vanguard's numbers.

S&P 500 (Large Cap Blend) vs. Large Cap Value: VFIAX PE is 24, while VVIAX PE is 21, which makes sense.

Small Cap Blend vs. Small Cap Value: NAESX PE is 69, while VSIAX is 43, which makes sense. And amazingly enough, the PE for Small-Cap Growth (VSGAX) is a mind-boggling 259. :shock:
JRB22 wrote:On this last point-- Isn't "value" investing based on selecting funds with relatively low PE ratios? If so, how is it that a broad index of "value" based funds has such a remarkably higher PE than the S&P 500?
Same answer as above. Was an interesting point!
Thank you for pointing out the use of Prospective P/E. I personally don't find it so outrageous, and use the reverse (the prospective earnings yield) as a somewhat optimistic way to estimate future growth.

However, the P/E listed by Vanguard looks like bad data. If true, that would be way above market average valuations, and you'd hear talk of a "small-cap bubble".

The Trailing Twelve Month P/E listed by Schwab for the ETFs based on the same indeces, VBR and VBK, is just 18.77 and 26.71, respectively, as of 05/31/2017. I haven't been able to verify from another source, however, so if someone find another number, that might confirm which one is correct.

User avatar
Taylor Larimore
Advisory Board
Posts: 27525
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"The Larry Portfolio" ?

Post by Taylor Larimore » Mon Jun 26, 2017 2:39 pm

Bogleheads:

I'm not clear. What is the "Larry Portfolio?"

Please give the exact source from Larry Swedroe.

Thank you and best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
siamond
Posts: 4239
Joined: Mon May 28, 2012 5:50 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by siamond » Mon Jun 26, 2017 2:54 pm

XdUzHa3NtSeIkBkIGPVn wrote:Thank you for pointing out the use of Prospective P/E. I personally don't find it so outrageous, and use the reverse (the prospective earnings yield) as a somewhat optimistic way to estimate future growth.
I don't have a pointer readily available, but Forward PEs have been demonstrated as quite ludicrous numerous times, and few people on this board give them any credit. Fact is most of those analysts have a major conflict of interest (and no crystal ball)... Apologies, can't back up my claim off the top of my head though.
XdUzHa3NtSeIkBkIGPVn wrote:However, the P/E listed by Vanguard looks like bad data. If true, that would be way above market average valuations, and you'd hear talk of a "small-cap bubble".

The Trailing Twelve Month P/E listed by Schwab for the ETFs based on the same indices, VBR and VBK, is just 18.77 and 26.71, respectively, as of 05/31/2017. I haven't been able to verify from another source, however, so if someone find another number, that might confirm which one is correct.
Well, Vanguard operates those funds, so... they would know better than anybody else. Did you check the Wiki page I pointed you too? I did most of the research for it, actually (nagging Morningstar, Vanguard, Prof. Shiller, etc), and the more I took a dive in PE computation procedures, the more I shook my head. Even when solely focusing on TTM PE, the differences are striking, and it became clear that one can only compare numbers coming from the same source, otherwise, it is just apples and oranges. There isn't "one correct number", there are multiple methodologies providing different numbers. More head-scratching details in this thread.

About the TTM numbers from Schwab, I don't know exactly where you found them, but I can guess where they are coming from. Morningstar displays the prospective PE, but they also maintain their own version of the TTM PE in their databases (access through Morningstar Direct or similar paying contract). Using YCharts, one can actually display those TTM values from M*, except that YCharts calls it "Weighted Average PE Ratio", just to confuse readers... Then I found the exact numbers you quoted (18.77 and 26.71, as of May 31st). While I was at it, I looked at the S&P 500 with this precise metric, and it's 21.24. Sigh, go figure. All this being said, I find the Morningstar methodology rather dubious, and Vanguard owns those funds and they should know better... So I much prefer looking at the Vanguard numbers.

This being said, I share your puzzlement about the Vanguard SCG number (PE of 259). I mean, sometimes, earnings are temporarily so dismal that the PE metric becomes meaningless (hence the CAPE methodology), but still, this does seem quite weird.

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: "The Larry Portfolio" ?

Post by triceratop » Mon Jun 26, 2017 3:10 pm

Taylor Larimore wrote:Bogleheads:

I'm not clear. What is the "Larry Portfolio?"

Please give the exact source from Larry Swedroe.

Thank you and best wishes.
Taylor
Taylor,

Here is a description of the class of portfolios on portfoliocharts: The Larry Portfolio

In particular it references Reducing the Risk of Black Swans by Swedroe and Grogan. The idea is that the following risk factors have historically been rewarded: market (beta), value, size, and term (for bonds). You diversify across these risks and obtain a portfolio that is less exposed to any single risk factor. A typical portfolio might be 70% intermediate term treasuries, 15% US SCV, 10% International SV, and 5% EM Value.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
Taylor Larimore
Advisory Board
Posts: 27525
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: "The Larry Portfolio" ?

Post by Taylor Larimore » Mon Jun 26, 2017 4:23 pm

triceratop wrote:
Taylor Larimore wrote:Bogleheads:

I'm not clear. What is the "Larry Portfolio?"

Please give the exact source from Larry Swedroe.

Thank you and best wishes.
Taylor
Taylor,

Here is a description of the class of portfolios on portfoliocharts: The Larry Portfolio

In particular it references Reducing the Risk of Black Swans by Swedroe and Grogan. The idea is that the following risk factors have historically been rewarded: market (beta), value, size, and term (for bonds). You diversify across these risks and obtain a portfolio that is less exposed to any single risk factor. A typical portfolio might be 70% intermediate term treasuries, 15% US SCV, 10% International SV, and 5% EM Value.
Tricertop:

There must be a mistake. I have most of Larry's books. In Larry's 1998 book, The Only Guide to a Winning Investment Strategy, Larry lists 4 portfolios. His "Moderate Portfolio" contains:

U.S. Stocks:
Large-Cap 10%
Large-cap value 10%
Small-cap 5%
Small-cap value 5%
Real Estate 5%

International Stocks:
Large-cap 10%
Small-cap 10%
Emerging markets 5%

Fixed income:
U.S. 5-year gov't 10%
U.S. 1-year gov't 20%
Global Fixed income 10%

I was head of the Finance Division of South Florida Small Business Administration but I cannot understand most of the (11) charts in your post that were copyrighted this year?
David Swensen, Yale Chief Investment Officer: "As a general rule of thumb, the more complexity that exists in a Wall Street creation, the faster and farther investors should run."
Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
siamond
Posts: 4239
Joined: Mon May 28, 2012 5:50 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by siamond » Mon Jun 26, 2017 4:39 pm

Taylor, there is no mistake. This specific type of portfolio (70% bonds, plus some high-risk stock classes) was discussed at length last year in this thread, with Larry being actively involved.

As to the type of charts displayed by PortfolioCharts, yes, they are fairly novel but actually surprisingly intuitive. You do need to read more on this wonderful site to get the hang of it though, maybe starting from this page.

Random Walker
Posts: 3205
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Random Walker » Mon Jun 26, 2017 4:47 pm

Top99% and Siamond,
You both said you're not willing to bet on the Larry Portfolio extreme SV tilt. I bet you know how he responds to that issue. He would say that you are making a big concentrated bet on a single factor, market beta. SV is diversifying across factors that have low correlations to each other: market, size, value. When one thinks about diversifying acros factors, risk factors, sources of return, the typical 60/40 portfolio with 90% of its risk in market beta looks like the highly concentrated bet! (Of course it's also the cheapest and simplest way to invest)

Dave

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Mon Jun 26, 2017 4:55 pm

Taylor,

It might be in this book: “Reducing the Risk of Black Swans: Using the Science of Investing to Capture Returns with Less Volatility.”

Mentioned by Larry here: http://www.etf.com/sections/index-inves ... nopaging=1

User avatar
siamond
Posts: 4239
Joined: Mon May 28, 2012 5:50 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by siamond » Mon Jun 26, 2017 5:06 pm

Random Walker wrote:Top99% and Siamond,
You both said you're not willing to bet on the Larry Portfolio extreme SV tilt. I bet you know how he responds to that issue. He would say that you are making a big concentrated bet on a single factor, market beta. [...]
Yes, as I indicated in my previous post (the answer to Taylor), there was a long thread where the topic was discussed in a 'lively manner' with Larry, and sure enough, you're right, he mentioned the point. He didn't convince me, nor did I convince him (as usual!), but this was an interesting discussion. :P

Note that personally, I don't use a 3-funds portfolio, I use tilts towards SV, EM, IS, and more. I just prefer to ground my portfolio on the basis of TSM (and Int'l TSM) and add tilts, not bet the farm on such market corners. The 'factor diversification' story seems rather suspicious to me (we could find many other ways to slice and dice the market), but I do acknowledge that the past trajectory of such portfolio is quite remarkable (so far and in the US, at least). In retrospect, maybe Larry did convince me that my tilts are a tad too timid...

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: "The Larry Portfolio" ?

Post by triceratop » Mon Jun 26, 2017 5:31 pm

Taylor Larimore wrote:
triceratop wrote:
Taylor Larimore wrote:Bogleheads:

I'm not clear. What is the "Larry Portfolio?"

Please give the exact source from Larry Swedroe.

Thank you and best wishes.
Taylor
Taylor,

Here is a description of the class of portfolios on portfoliocharts: The Larry Portfolio

In particular it references Reducing the Risk of Black Swans by Swedroe and Grogan. The idea is that the following risk factors have historically been rewarded: market (beta), value, size, and term (for bonds). You diversify across these risks and obtain a portfolio that is less exposed to any single risk factor. A typical portfolio might be 70% intermediate term treasuries, 15% US SCV, 10% International SV, and 5% EM Value.
Tricertop:

There must be a mistake. I have most of Larry's books. In Larry's 1998 book, The Only Guide to a Winning Investment Strategy, Larry lists 4 portfolios. His "Moderate Portfolio" contains:

U.S. Stocks:
Large-Cap 10%
Large-cap value 10%
Small-cap 5%
Small-cap value 5%
Real Estate 5%

International Stocks:
Large-cap 10%
Small-cap 10%
Emerging markets 5%

Fixed income:
U.S. 5-year gov't 10%
U.S. 1-year gov't 20%
Global Fixed income 10%
That is not a Larry portfolio. I wasn't referring to The Only Guide to a Winning Investment Strategy; however, the previously mentioned Reducing the Risk of Black Swans book by Swedroe does indeed have a Larry portfolio. In this post Larry notes that it was the NY Times that coined "The Larry Portfolio", it wasn't an act of self-aggrandizement. ; )
I was head of the Finance Division of South Florida Small Business Administration but I cannot understand most of the (11) charts in your post that were copyrighted this year?
David Swensen, Yale Chief Investment Officer: "As a general rule of thumb, the more complexity that exists in a Wall Street creation, the faster and farther investors should run."
Best wishes.
Taylor
Those charts are not related to what the Larry Portfolio is defined as, but instead to help (or not) understand its past performance in various market conditions. This link is the same page but for your Three-Fund Portfolio. I think you'll agree the "Copyright 2017" on those charts doesn't invalidate the many reasons for investors to consider the Three Fund portfolio as an investment allocation.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

Avo
Posts: 1099
Joined: Wed Jun 11, 2008 2:21 am
Location: California

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Avo » Mon Jun 26, 2017 5:51 pm

Random Walker wrote:[Larry] would say that you are making a big concentrated bet on a single factor, market beta. SV is diversifying across factors that have low correlations to each other: market, size, value. When one thinks about diversifying acros factors, risk factors, sources of return, the typical 60/40 portfolio with 90% of its risk in market beta looks like the highly concentrated bet! (Of course it's also the cheapest and simplest way to invest)
I've always had a hard time with the notion that market beta is just another "factor", like size or value.

To me, it makes sense that equities are fundamentally different beasts than bonds. I don't see how anyone could deny this. This is the equity "factor".

The efficient market hypothesis (EMH) also makes sense to me. There are a lot of investors with a lot of cumulative knowledge; it seems extremely plausible that the market price is the best available estimate of the true value of a stock.

But in order for factors such as value and size to exist, EMH must be wrong.

I can accept that EMH might be wrong for behavioral/psychological reasons, but it sure seems like a much less robust argument, and that these factors might well disappear with increased knowledge on the part of more and more investors, and increased access to inexpensive trading.

But none of this will ever turn a stock into a bond. So I believe that beta is fundamentally different than the other "factors".

OK, what did I get wrong here??

Random Walker
Posts: 3205
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Random Walker » Mon Jun 26, 2017 6:02 pm

Avo,
Why do you think that the presence of size and value factors requires EMH to be wrong? Seems to me that if one considers them to be purely risk factors with no behavioral component, then they fit with EMH.

Dave

Random Walker
Posts: 3205
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Random Walker » Mon Jun 26, 2017 6:04 pm

Avo,
Also, do you understand that the size and value risk factors are based on long-short portfolios, and thus independent of market beta?

Dave

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by triceratop » Mon Jun 26, 2017 6:12 pm

Avo wrote:
Random Walker wrote:[Larry] would say that you are making a big concentrated bet on a single factor, market beta. SV is diversifying across factors that have low correlations to each other: market, size, value. When one thinks about diversifying acros factors, risk factors, sources of return, the typical 60/40 portfolio with 90% of its risk in market beta looks like the highly concentrated bet! (Of course it's also the cheapest and simplest way to invest)
I've always had a hard time with the notion that market beta is just another "factor", like size or value.

To me, it makes sense that equities are fundamentally different beasts than bonds. I don't see how anyone could deny this. This is the equity "factor".

The efficient market hypothesis (EMH) also makes sense to me. There are a lot of investors with a lot of cumulative knowledge; it seems extremely plausible that the market price is the best available estimate of the true value of a stock.

But in order for factors such as value and size to exist, EMH must be wrong.

I can accept that EMH might be wrong for behavioral/psychological reasons, but it sure seems like a much less robust argument, and that these factors might well disappear with increased knowledge on the part of more and more investors, and increased access to inexpensive trading.

But none of this will ever turn a stock into a bond. So I believe that beta is fundamentally different than the other "factors".

OK, what did I get wrong here??
(emphasis mine)

Stocks and bonds are fundamentally different beasts. But you wouldn't contend that treasuries have the same risks and rewards as investment-grade / junk corporates (or fallen angels, a further complication to the idea of a monolithic "bond" asset class), would you? That is the difference between the term factor and the additional default factor. People are perfectly content to expect a default premia. Does that contradict the EMH too?

If you agree with me on that point, it's not such a large leap to understanding that different kinds of stocks (e.g. sorted on size, value) have different risk exposures.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
XdUzHa3NtSeIkBkIGPVn
Posts: 39
Joined: Thu Jul 16, 2015 4:12 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by XdUzHa3NtSeIkBkIGPVn » Mon Jun 26, 2017 7:18 pm

siamond wrote:
XdUzHa3NtSeIkBkIGPVn wrote:However, the P/E listed by Vanguard looks like bad data. If true, that would be way above market average valuations, and you'd hear talk of a "small-cap bubble".

The Trailing Twelve Month P/E listed by Schwab for the ETFs based on the same indices, VBR and VBK, is just 18.77 and 26.71, respectively, as of 05/31/2017. I haven't been able to verify from another source, however, so if someone find another number, that might confirm which one is correct.
Well, Vanguard operates those funds, so... they would know better than anybody else. Did you check the Wiki page I pointed you too? I did most of the research for it, actually (nagging Morningstar, Vanguard, Prof. Shiller, etc), and the more I took a dive in PE computation procedures, the more I shook my head. Even when solely focusing on TTM PE, the differences are striking, and it became clear that one can only compare numbers coming from the same source, otherwise, it is just apples and oranges. There isn't "one correct number", there are multiple methodologies providing different numbers. More head-scratching details in this thread.
Thank you for the links. I think you're right, the Vanguard numbers are probably the more reasonable ones, considering how Morningstar butchers the math on fund P/Es.

I think the much higher numbers from Vanguard reflect the impact of companies with negative earnings. And yes, it looks like Schwab just sources from M*, including data M* doesn't even show in their website.

This would probably be easier if people used the earnings yield, rather than P/Es. No problem with zero or negative earnings, you can do weighted-averages, it's a reasonable estimate of returns, etc.

freyj6
Posts: 206
Joined: Thu Jun 19, 2014 2:09 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by freyj6 » Mon Jun 26, 2017 7:46 pm

I don't think a large adjustment like this ever makes sense.

Going from 60/40 or whatever to 30/70 when CAPE hits 30 is like garaging your car when gas is over $3 but driving 20 miles a day when it's $2.90.

I do somewhat agree with the general sentiment though.

Personally I've made some adjustments down from the 100% equity I held before that. If CAPE goes toward 35, I'll probably gradual edge toward 50/50. If it moves toward 40, I'll probably end up somewhere near the Larry portfolio.

Big, hasty jumps smell like emotional overreaction to me. Small, incremental changes based on valuation, I think, make sense.

:)

AlohaJoe
Posts: 3847
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by AlohaJoe » Mon Jun 26, 2017 7:53 pm

Avo wrote:The efficient market hypothesis (EMH) also makes sense to me. There are a lot of investors with a lot of cumulative knowledge; it seems extremely plausible that the market price is the best available estimate of the true value of a stock.

But in order for factors such as value and size to exist, EMH must be wrong.

I can accept that EMH might be wrong for behavioral/psychological reasons, but it sure seems like a much less robust argument, and that these factors might well disappear with increased knowledge on the part of more and more investors, and increased access to inexpensive trading.

But none of this will ever turn a stock into a bond. So I believe that beta is fundamentally different than the other "factors".

OK, what did I get wrong here??
Phrases like "EMH" are sort of weasel-words (because, when pressed, people start talking about weak- or strong- or semi-strong- variants of EMH) but my understanding is that outside of a small number of (admittedly) extremely influential people who made their names in the 1970s (Fama & Sharpe being probably the most prominent), academics don't really subscribe to the EMH any more. In essence, that's why Stiglitz in 2001 and Shiller in 2013 won their Nobel Prizes in Economics -- as "official" recognition of the death of EMH. From the official Nobel description of why Shiller won his prize, "Robert Shiller's conclusion was therefore that the market is inefficient."

From Andrew Ang's book Asset Management he says "active managers have never believed in truly efficient markets, and now academics no longer believe in them either."
Today, economists do not believe in perfectly efficient markets. In fact, markets cannot be efficient in their pure form. The modern notion of market near-efficiency is developed by Sanford Grossman and Joseph Stiglitz (1980), which forms part of the collection of papers for which Stiglitz was awarded his Nobel Prize in 2001. Grossman and Stiglitz describe a world in which markets are nearly efficient, and in doing so they address a conundrum that arises from the costless information assumption of the CAPM. Suppose that it is costly to collect information and to trade on that information, as it is in the real world. Then, if all information is in the price already, why would anyone ever invest in gathering the information? But if no one invests in gathering the information, how can information be reflected in security prices so that markets are efficient? It is then impossible that markets be efficient in their pure form.

Grossman and Stiglitz develop a model in which markets are near-efficient. Active managers search for pockets of inefficiency, and in doing so cause the market to be almost efficient.
I don't quite follow what you mean by "you can't turn a stock into a bond" so I might be misunderstanding. But, as stated, this is clearly false. (At best maybe you can argue "you can't turn a stock into a US short- or intermediate-term government bond".) Long bonds are famously "just like stocks" in every risk measurement you can think of. This board is full of posts telling people that "high-yield corporate bonds have too much equity risk". Or that "emerging market government bonds have too much emerging market equity risk". So the line between "stock" and "bond" is clearly not a bright line. If bonds can be stock-like then it isn't clear at all why stocks can't be bond-like. (Plus you get into weird arguments about how certain bonds aren't really bonds...which seems like abusing terms; clearly you mean something other than "bond" when you say "bond".) And that's (essentially) what the low-volatility anomaly argument and the "easy access to diversification has driven up valuations" arguments have been about.

Random Walker
Posts: 3205
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Random Walker » Mon Jun 26, 2017 8:59 pm

Big, hasty jumps smell like emotional overreaction to me. Small, incremental changes based on valuation, I think, make sense.
I basically agree. But the market makes big jumps. If one is really in tune with ability, willingness, need for risk and their own changes in need, willingness, ability, then after a big run up over say last 9 years, I can see rational dramatic once in a lifetime decrease in % equity.

Dave

User avatar
siamond
Posts: 4239
Joined: Mon May 28, 2012 5:50 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by siamond » Mon Jun 26, 2017 10:45 pm

XdUzHa3NtSeIkBkIGPVn wrote:I think the much higher numbers from Vanguard reflect the impact of companies with negative earnings.
Yes, exactly. And obviously, numerous small companies have negative earnings. I'm still puzzled by the SCG number, but otherwise, it does seem to make sense. Would be interesting to see the numbers at the end of June to see if the updated value stays in the same range.
XdUzHa3NtSeIkBkIGPVn wrote:And yes, it looks like Schwab just sources from M*, including data M* doesn't even show in their website.
M* has plenty of data they don't show on public Web pages. I subscribed to a free trial of Morningstar Direct while I was investigating those matters, and of course, M* monetizes all this additional data (including a bunch of historical snapshots). Unfortunately, they do NOT archive enough information to be able to redo the PE (or the CAPE) math for a given fund using another methodology than theirs... :annoyed
XdUzHa3NtSeIkBkIGPVn wrote:This would probably be easier if people used the earnings yield, rather than P/Es. No problem with zero or negative earnings, you can do weighted-averages, it's a reasonable estimate of returns, etc.
Yup, I had the very exact same thought when I studied the matter. And when you think about the Vanguard methodology, this is basically what they do (since a harmonic average is the reciprocal of a weighed average of the E/P of individual companies). Also, when speaking of expected returns, E/P is (imho) a much more useful quantity than P/E (or CAPE) and the dubious reasoning about the existence of a historical mean.

User avatar
Top99%
Posts: 334
Joined: Sat Apr 22, 2017 9:30 am
Location: Austin, TX

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Top99% » Tue Jun 27, 2017 7:26 am

Random Walker wrote:Top99% and Siamond,
You both said you're not willing to bet on the Larry Portfolio extreme SV tilt. I bet you know how he responds to that issue. He would say that you are making a big concentrated bet on a single factor, market beta. SV is diversifying across factors that have low correlations to each other: market, size, value. When one thinks about diversifying acros factors, risk factors, sources of return, the typical 60/40 portfolio with 90% of its risk in market beta looks like the highly concentrated bet! (Of course it's also the cheapest and simplest way to invest)

Dave
Dave,

Our portfolio has a lot of Larry influence in it but we are closer to 50% CDs/intermediate bond funds with the other 50% consisting of a fairly even mix of alternatives, broad market index funds and small value. So, our portfolio is certainly Larry influenced but not full Larry.
Adapt or perish

jdilla1107
Posts: 601
Joined: Sun Jun 24, 2012 8:31 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by jdilla1107 » Tue Jun 27, 2017 8:54 am

It's easy to see what is going on with the CAPE by looking at two graphs:

Here is the current PE of the sp500:

http://www.multpl.com/

Here is the PE10 (cape)

http://www.multpl.com/shiller-pe/

The big reason the cape is high is because of the huge PE spike in 2008-2009. Once we get through that period, it's going to drop. Perhaps we should base our crystal balls on PE7 or PE5 instead. :twisted:

People get surprised seeing the PE spike in 2008, because it didn't include a run up in prices. People always forget the E in PE. For example, the PE could fall by earnings increasing, with no change in price.

skjoldur
Posts: 144
Joined: Thu Sep 25, 2014 3:11 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by skjoldur » Tue Jun 27, 2017 10:51 am

jdilla1107 wrote: The big reason the cape is high is because of the huge PE spike in 2008-2009. Once we get through that period, it's going to drop. Perhaps we should base our crystal balls on PE7 or PE5 instead. :twisted:

People get surprised seeing the PE spike in 2008, because it didn't include a run up in prices. People always forget the E in PE. For example, the PE could fall by earnings increasing, with no change in price.
Here is a forecast of CAPE as 2008-2009 rolls out under a range of assumptions (and beyond):
https://earlyretirementnow.com/2017/03/22/cape-fear/

roflwaffle
Posts: 439
Joined: Mon Mar 02, 2015 10:08 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by roflwaffle » Tue Jun 27, 2017 11:25 pm

CAPE is on the high side, but what does the CA(After Tax)PE look like? If we're using an index over time, we should either make sure the majority of other factors that could affect that index are more or less stable, or that we're adjusting/tweaking our index relative to how much those factors change.

selters
Posts: 608
Joined: Thu Feb 27, 2014 9:26 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by selters » Wed Jun 28, 2017 1:52 am

Avo wrote: But in order for factors such as value and size to exist, EMH must be wrong.
Not if the excess return of size, value and momentum comes from increased risk.

selters
Posts: 608
Joined: Thu Feb 27, 2014 9:26 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by selters » Wed Jun 28, 2017 2:02 am

jdilla1107 wrote:
The big reason the cape is high is because of the huge PE spike in 2008-2009.
I read somewhete a couple of years ago that the impact og 2008-2009 on CAPE is often overstated. They could be wrong, I don't know. Is there a tool to find PE8 or PE5 somewhere?

NibbanaBanana
Posts: 224
Joined: Sun Jan 22, 2017 10:34 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by NibbanaBanana » Wed Jun 28, 2017 7:56 am

siamond wrote:
XdUzHa3NtSeIkBkIGPVn wrote:Thank you for pointing out the use of Prospective P/E. I personally don't find it so outrageous, and use the reverse (the prospective earnings yield) as a somewhat optimistic way to estimate future growth.
I don't have a pointer readily available, but Forward PEs have been demonstrated as quite ludicrous numerous times, and few people on this board give them any credit. Fact is most of those analysts have a major conflict of interest (and no crystal ball)... Apologies, can't back up my claim off the top of my head though.
XdUzHa3NtSeIkBkIGPVn wrote:However, the P/E listed by Vanguard looks like bad data. If true, that would be way above market average valuations, and you'd hear talk of a "small-cap bubble".

The Trailing Twelve Month P/E listed by Schwab for the ETFs based on the same indices, VBR and VBK, is just 18.77 and 26.71, respectively, as of 05/31/2017. I haven't been able to verify from another source, however, so if someone find another number, that might confirm which one is correct.
Well, Vanguard operates those funds, so... they would know better than anybody else. Did you check the Wiki page I pointed you too? I did most of the research for it, actually (nagging Morningstar, Vanguard, Prof. Shiller, etc), and the more I took a dive in PE computation procedures, the more I shook my head. Even when solely focusing on TTM PE, the differences are striking, and it became clear that one can only compare numbers coming from the same source, otherwise, it is just apples and oranges. There isn't "one correct number", there are multiple methodologies providing different numbers. More head-scratching details in this thread.

About the TTM numbers from Schwab, I don't know exactly where you found them, but I can guess where they are coming from. Morningstar displays the prospective PE, but they also maintain their own version of the TTM PE in their databases (access through Morningstar Direct or similar paying contract). Using YCharts, one can actually display those TTM values from M*, except that YCharts calls it "Weighted Average PE Ratio", just to confuse readers... Then I found the exact numbers you quoted (18.77 and 26.71, as of May 31st). While I was at it, I looked at the S&P 500 with this precise metric, and it's 21.24. Sigh, go figure. All this being said, I find the Morningstar methodology rather dubious, and Vanguard owns those funds and they should know better... So I much prefer looking at the Vanguard numbers.

This being said, I share your puzzlement about the Vanguard SCG number (PE of 259). I mean, sometimes, earnings are temporarily so dismal that the PE metric becomes meaningless (hence the CAPE methodology), but still, this does seem quite weird.
I read through all your posts so far siamond and they are very helpful. Thanks. This whole thing kind of argues for using another metric for valuation. One that is a little more reliable. I just got finished reading "The Investor's Manifesto" by William Bernstein. He emphasizes that it is very valuable to deploy the Gordon Equation for estimating future returns. That's dividend yield plus dividend growth rate for stocks. That's it! It's that simple. And that's exactly what John Bogle says to use in "Bogle on Mutual Funds". (Current yield minus default rate for bonds BTW. Bogle just uses current yield but he might have just been talking treasuries.) I don't think there can be any argument about the dividend yield. The dividend growth rate? Who knows. But it might be pretty close to what it's been. IIRC Bernstein said it was between 1.5% and 2%. So there you go. About 4% for stocks. The last paper I read from VG, they were calling for 5-6% returns for stocks. Low and slow, rates and growth.

As an aside comment: That book by Bernstein is really good. Recommended highly.

azanon
Posts: 1964
Joined: Mon Nov 07, 2011 10:34 am
Location: Little Rock, AR

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by azanon » Wed Jun 28, 2017 8:01 am

sjwoo wrote:How about switching to the Larry Portfolio right about now? Which of course we know to be:

15% Small Cap Value
15% Emerging Market
70% Intermediate Term Bonds

- Sung
So you'd suggest to have 50% of your equities in US stock, despite being so aware of US stock valuations? Why? You can get about half that CAPE with a total international fund. What I'm getting at, is that is you believe there's validity to the problem you're pointing out, then your suggested solution isn't logical, or near logical enough. I found 3 different sites that use CAPE valuation to estimate future return of US stock, and those estimates range from 0% to -1.4% real. So why own any?

scone
Posts: 1454
Joined: Wed Jul 11, 2012 4:46 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by scone » Wed Jun 28, 2017 9:25 am

Consider that all this stuff about market timing, CAPE ratios, Larry Portfolio, etc. is simply about your "gut" fear of a crash. And as Larry says over and over, your "gut" tends to make poor decisions when it comes to investing.

You are experiencing fear. Pure and simple. If the fear is threatening to make you do something stupid with your portfolio, then take action now to reduce the fear. Suggestions: stop reading financial news, even Bogleheads; increase your savings rate and put that money into a very low risk vehicle; review your rebalancing plan.

OTOH, if starting this thread is really for infotainment purposes, then please carry on. Bogleheads are lots of fun to hang out with. :D
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 9:47 am

azanon wrote:I found 3 different sites that use CAPE valuation to estimate future return of US stock, and those estimates range from 0% to -1.4% real.
Are any of the 3 worth sharing?

I like RA and AQR, both of which predict above 0% for US. GMO doesn't give methodology but I still look at their predictions. Those 3 are linked here: http://www.raddr-pages.com/forums/viewt ... f=2&t=7047 and I've been planning to make a similar thread here at Bogleheads.

User avatar
HomerJ
Posts: 11843
Joined: Fri Jun 06, 2008 12:50 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by HomerJ » Wed Jun 28, 2017 9:57 am

azanon wrote:I found 3 different sites that use CAPE valuation to estimate future return of US stock, and those estimates range from 0% to -1.4% real.
Shiller himself (the guy who INVENTED CAPE) predicted 0% 10-year real return based on CAPE in 1996. Instead we got like 6% real over the next 10 years.

Swedroe and other experts calculated 4.5% expected real returns based on valuations in 2010. Instead we've gotten like 12% real since 2010.

Nobody knows enough to predict anything accurately. And certainly not after looking at just ONE metric. There are too many variables.

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by triceratop » Wed Jun 28, 2017 10:19 am

HomerJ wrote:
azanon wrote:I found 3 different sites that use CAPE valuation to estimate future return of US stock, and those estimates range from 0% to -1.4% real.
Shiller himself (the guy who INVENTED CAPE) predicted 0% 10-year real return based on CAPE in 1996. Instead we got like 6% real over the next 10 years.

Swedroe and other experts calculated 4.5% expected real returns based on valuations in 2010. Instead we've gotten like 12% real since 2010.

Nobody knows enough to predict anything accurately. And certainly not after looking at just ONE metric. There are too many variables.
I imagine Shiller didn't realize that in 2006 we would be in the middle of a massive speculative bubble propped up by fraud and other illegal activity. Is his model worth discarding because of this failure? Draw your own conclusion.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 12:00 pm

HomerJ wrote:Nobody knows enough to predict anything accurately.
That's true, but if you're buying a business with average growth, and pay triple what it earned over the last decade after adjusting for inflation, then you shouldn't expect much return on your investment.

If you buy the US market today with CAPE of 30, you shouldn't expect much of a return. 1/CAPE is just a guess, yield + growth is just a guess. Better a guess then nothing.

Hey maybe you'll get lucky and someday a fool will pay 4 times what the business earned over the last decade.

Thesaints
Posts: 1619
Joined: Tue Jun 20, 2017 12:25 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Thesaints » Wed Jun 28, 2017 12:07 pm

lazyday wrote:
HomerJ wrote:Nobody knows enough to predict anything accurately.
That's true, but if you're buying a business with average growth, and pay triple what it earned over the last decade after adjusting for inflation, then you shouldn't expect much return on your investment.
Only if the business did not change in the past 10 years.
Apple Inc. 10 years ago was in the personal computers business, today they do phones.
Also, business may even stay the same, but the environment may change. Look at the interest rates.

That is to say that CAPE today is vastly overrated. In the good old times it is true that averaging earnings over ten years would smooth out temporary glitches, but today the dynamics is quite different.
Why should I give any importance to netflix earnings in 2008 and actually give them the very same importance as earnings in 2016 ? It makes no sense at all.

Da5id
Posts: 2059
Joined: Fri Feb 26, 2016 8:20 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Da5id » Wed Jun 28, 2017 12:17 pm

Thesaints wrote: That is to say that CAPE today is vastly overrated. In the good old times it is true that averaging earnings over ten years would smooth out temporary glitches, but today the dynamics is quite different.
Why should I give any importance to netflix earnings in 2008 and actually give them the very same importance as earnings in 2016 ? It makes no sense at all.
I don't use CAPE for market timing, or anything much, though it contributes to my low target initial SWR (< 3%, which I'm already at). I think people considering retirement now might consider working a bit longer if their SWR is at 4% or above and they don't have lots of options to go back to work if things get ugly, but that is as far as I'd go there.

That said, your arguments sound like some famous last words, "This time it is different". Maybe it is, and maybe it isn't, but it feels like rationalization. I'd not be shocked by a serious correction. Or by stocks going up, but slowly. I'd be most surprised if after a long bull market we go off on a series of double digit gaining years, but not shocked by that either. Who knows, maybe stocks have reached a "permanently high plateau"...

User avatar
HomerJ
Posts: 11843
Joined: Fri Jun 06, 2008 12:50 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by HomerJ » Wed Jun 28, 2017 12:22 pm

triceratop wrote:
HomerJ wrote:There are too many variables.
I imagine Shiller didn't realize that in 2006 we would be in the middle of a massive speculative bubble propped up by fraud and other illegal activity. Is his model worth discarding because of this failure? Draw your own conclusion.
Oh, the excuse is that something happened that wasn't expected?

Weird, that's what EVERY SINGLE MODEL that has ever failed uses as an excuse.

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 12:24 pm

Thesaints wrote:Only if the business did not change in the past 10 years.
Apple Inc. 10 years ago was in the personal computers business, today they do phones.
If you expect earnings to grow faster in the future than in the past, that boosts expected return and could justify paying more. But I wouldn't want to make that bet.
Also, business may even stay the same, but the environment may change. Look at the interest rates.
Yes, with low interest rates, it can still make sense to own stocks. In my first post in this thread, I did say that stocks don't seem so extremely expensive if you look at the risk premium (over bonds). I still own stocks, but don't expect them to return much.

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by triceratop » Wed Jun 28, 2017 12:26 pm

HomerJ wrote:
triceratop wrote:
HomerJ wrote:There are too many variables.
I imagine Shiller didn't realize that in 2006 we would be in the middle of a massive speculative bubble propped up by fraud and other illegal activity. Is his model worth discarding because of this failure? Draw your own conclusion.
Oh, the excuse is that something happened that wasn't expected?

Weird, that's what EVERY SINGLE MODEL that has ever failed uses as an excuse.
Actually, models don't make excuses. Humans interpret the failures and try to understand if there was anything wrong with the model. But this model is about fundamentals and fair value; outright fraud is not a factor one should consider so pillorying the model for that reason seems opportunistic, at least to this reader, as if you will reject any model with an outlier or which does not fitting the data entirely.

By the way, Larry doesn't make point forecasts alone, he makes clear that it is the center of a very wide possible distribution of returns. Those who understand probability will draw the correct conclusion.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

Thesaints
Posts: 1619
Joined: Tue Jun 20, 2017 12:25 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Thesaints » Wed Jun 28, 2017 12:30 pm

Da5id wrote:
Thesaints wrote: That is to say that CAPE today is vastly overrated. In the good old times it is true that averaging earnings over ten years would smooth out temporary glitches, but today the dynamics is quite different.
Why should I give any importance to netflix earnings in 2008 and actually give them the very same importance as earnings in 2016 ? It makes no sense at all.
I don't use CAPE for market timing, or anything much, though it contributes to my low target initial SWR (< 3%, which I'm already at). I think people considering retirement now might consider working a bit longer if their SWR is at 4% or above and they don't have lots of options to go back to work if things get ugly, but that is as far as I'd go there.

That said, your arguments sound like some famous last words, "This time it is different". Maybe it is, and maybe it isn't, but it feels like rationalization. I'd not be shocked by a serious correction. Or by stocks going up, but slowly. I'd be most surprised if after a long bull market we go off on a series of double digit gaining years, but not shocked by that either. Who knows, maybe stocks have reached a "permanently high plateau"...
Until ~1985 CAPE has oscillated around 15. Past that it has oscillated around 25. Have returns decreased ?
Some time it is different and there are plenty of examples

And I do agree: until interest rates will stay this low, stock returns on average will be subdued.
Last edited by Thesaints on Wed Jun 28, 2017 12:37 pm, edited 1 time in total.

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 12:37 pm

Thesaints wrote:Until ~1985 CAPE has oscillated around 15. Past that it has oscillated around 25. Have returns decreased ?
Some time it is different and there are plenty of examples
Simple predictions like yield+growth do not assume that CAPE will return towards 15, they assume that CAPE will stay the same.

So we might predict 3.5% return. To get a 6.5% return, that requires either faster growth than in the past, or an even higher CAPE than today's.

Thesaints
Posts: 1619
Joined: Tue Jun 20, 2017 12:25 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Thesaints » Wed Jun 28, 2017 12:42 pm

lazyday wrote:
Thesaints wrote:Until ~1985 CAPE has oscillated around 15. Past that it has oscillated around 25. Have returns decreased ?
Some time it is different and there are plenty of examples
Simple predictions like yield+growth do not assume that CAPE will return towards 15, they assume that CAPE will stay the same.

So we might predict 3.5% return. To get a 6.5% return, that requires either faster growth than in the past, or an even higher CAPE than today's.
You realize that CAPE still has one year with no earnings inside, don't you ? I though we agreed on it not being a very relevant parameter

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 12:48 pm

Thesaints wrote:You realize that CAPE still has one year with no earnings inside, don't you ? I though we agreed on it not being a very relevant parameter
I don't follow. What's the significance of the missing most recent earnings data?

I don't agree that CAPE isn't very relevant, I think it's important how much you pay for earnings.

User avatar
HomerJ
Posts: 11843
Joined: Fri Jun 06, 2008 12:50 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by HomerJ » Wed Jun 28, 2017 12:51 pm

triceratop wrote:Those who understand probability will draw the correct conclusion.
The correct conclusion, in my opinion, is that 4.5% plus or minus 8% is basically the same thing as "No one knows enough to predict anything accurately"

We don't even know if we're talking about a normal distribution here.

Look, I'm not saying I know the answer... I would just like to see a little bit of humility and doubt when actual results are so far off from predictions.

If you have a model that predicts a flood every hundred years, and you get 3 floods in 15 years, sure it's possible the model is correct, and we just got unlucky.

But I can't imagine why a person wouldn't wonder, just a little, if maybe, just maybe, the model is missing a variable (or if a variable in the model has changed since the model was created).
Last edited by HomerJ on Wed Jun 28, 2017 12:56 pm, edited 1 time in total.

Thesaints
Posts: 1619
Joined: Tue Jun 20, 2017 12:25 am

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by Thesaints » Wed Jun 28, 2017 12:54 pm

lazyday wrote:
Thesaints wrote:You realize that CAPE still has one year with no earnings inside, don't you ? I though we agreed on it not being a very relevant parameter
I don't follow. What's the significance of the missing most recent earnings data?

I don't agree that CAPE isn't very relevant, I think it's important how much you pay for earnings.
There were "no earnings" in 12 months across 2008-09. Those are still in the CAPE formula. I understand paying for the sins of our fathers, but I'm not gonna calculate my P/E based on their earnings and my prices.

lazyday
Posts: 3297
Joined: Wed Mar 14, 2007 10:27 pm

Re: Yes, This Is Market Timing: If the CAPE Goes Above 30, Bring on the Larry

Post by lazyday » Wed Jun 28, 2017 12:57 pm

Thesaints wrote:There were "no earnings" in 12 months across 2008-09. Those are still in the CAPE formula. I understand paying for the sins of our fathers, but I'm not gonna calculate my P/E based on their earnings and my prices.
OK I see what you mean now.

If you remove that year from CAPE, then you should also remove the worst year from every 10 year span when calculating average CAPE.

Post Reply