Is the Stock Market Cheap?

[samclem]

But does it tell you when to get into the market I don't think so...

You could build in a mechanical trigger for out and into the market based on P/E10, I've seen such systems.
Do they work? Probably in backtesting, with the right tweaks and data sets.

I'm sticking with my high equity allocation. The dividends will hopefully keep coming even if the share prices go down.

 

[Texas Proud]

You could build in a mechanical trigger for out and into the market based on P/E10, I've seen such systems.
Do they work? Probably in backtesting, with the right tweaks and data sets.

I'm sticking with my high equity allocation. The dividends will hopefully keep coming even if the share prices go down.

That is what I am also doing ... I can only put so much in the market any year, so it really doesn't matter if it is currently overvalued or not... I still like the long term direction and hope it continues..

 

[bondi688]

Agree with the general feel of ziggy29. Donoff, if the market had really taken a big tumble today, I would have added some of the MF I hold in Vanguard.

 

[MichaelB]

The bottom line is that 10% earnings growth with 3-4% long-term revenue growth is simply not sustainable in the long term, but most corporate HQs aren't as interested in the long term as they are with this quarter.

Years ago I spent a day shadowing a senior exec, former "CEO in running" that hadn't made it but still had a key job. It was close to quarter end and he was going through forecasts and phone calls like noting I'd ever seen or imagined. During a short break I asked just how important the quarterly number was, relative to other measures. He looked at me like I was out of my mind, got a strange and scary look on his face, and said "I'd sell my mother for another 25 cents a share this quarter". That was it for the day for me. They locked me out for the rest because they were covering inside info. He meant it.

 

[haha]

In a way, yes.... a tool like PE/10 or even PE does not mean much if it can not be used to make informed decisions...

Since we have history, it would be interesting to see if there were any predictive value to this number... or is it just some interesting calculations that someone makes... with no meaning....

The whole genesis and point of PE10 is that it does predict returns- not over the next year, but over the next 10 or more.

Most of the criticisms are uninformed attempts by people trying to justify higher prices for assets

It is very interesting book to read, and there is much misunderstanding and misinformation about it out there in yackety-yackville.

Ha

 

[MichaelB]

The whole genesis and point of PE10 is that it does predict returns- not over the next year, but over the next 10 or more.

Most of the criticisms are uninformed attempts by people trying to justify higher prices for assets

It is very interesting book to read, and there is much misunderstanding and misinformation about it out there in yackety-yackville.

Ha

Agree. That, and it is also criticized for not being a timing tool, when it never was intended for that.

 

[photoguy]

The whole genesis and point of PE10 is that it does predict returns- not over the next year, but over the next 10 or more.

This is very clear when you look at the graphs Schiller produced in his papers. The first is 1-year price growth / PE10 and the second is 10-year price growth / PE10. (Taken from http://cowles.econ.yale.edu/P/cd/d12b/d1295.pdf )

The other thing to note is that there is a huge variation on returns. So it's not good as a timing signal even for 10 year predictions. I think he mentions in his paper that the R^2 is something like 30%.

 

[photoguy]

Here's a better graph showing PE10 vs returns (the schiller graphs above was price growth and I don't think it included total returns from dividends). The graph is from scatter – Mebane Faber Research – Stock Market and Investing Blog .

Note that around PE10 = 20 the "forecasted returns" range from -5% to 10% which is a huge range of returns. So if the next decade turns out closer to the -5% return I'd say with hindsight that yes the market was overvalued in 2013. On the other hand, if we get returns closer to 10%, then I'd say it was undervalued.

I also recall reading a paper comparing PE10 to PE1 and some other measures (possibly including forward estimates of earnings). I think PE10 edged the others out but not overwhelming so. I wish I could find the paper...

 

[clifp]

Here's a better graph showing PE10 vs returns (the schiller graphs above was price growth and I don't think it included total returns from dividends). The graph is from scatter – Mebane Faber Research – Stock Market and Investing Blog .

Note that around PE10 = 20 the "forecasted returns" range from -5% to 10% which is a huge range of returns. So if the next decade turns out closer to the -5% return I'd say with hindsight that yes the market was overvalued in 2013. On the other hand, if we get returns closer to 10%, then I'd say it was undervalued.

Interesting stuff. But even though the range of -5% to 10% is is large it is much smaller than the normal standard deviation of market returns.

BTW are those real or nominal returns?

It also makes something easier figure out for instance if we ever see TIPS bonds with 3.5+% real returns again, then assuming you can make your retirement work with 3.5% returns you should jump at the chance.

I know I will.

 

[photoguy]

Interesting stuff. But even though the range of -5% to 10% is is large it is much smaller than the normal standard deviation of market returns.

Agree. We're definitely on the higher side for pe10 with corresponding expectations.

BTW are those real or nominal returns?

Schiller's graphs are real price changes. The graph from Mebane Faber was taken from Schiller's data (which has CPI) so I think it's real as well. Although Mebane doesn't explicitly state so (it's from a blog post) the numbers seem consistent with other real returns I have seen.

It also makes something easier figure out for instance if we ever see TIPS bonds with 3.5+% real returns again, then assuming you can make your retirement work with 3.5% returns you should jump at the chance.

I think the failing to jump into TIPS/ibonds when they had 3%+ real returns was one of the biggest mistakes I made.

 

[Texas Proud]

The whole genesis and point of PE10 is that it does predict returns- not over the next year, but over the next 10 or more.

Most of the criticisms are uninformed attempts by people trying to justify higher prices for assets

It is very interesting book to read, and there is much misunderstanding and misinformation about it out there in yackety-yackville.

Ha

I have not read the book.... but there are two things that can make PE/10 go down... lower price or higher earnings....

You can see that the PE/10 is going up and down.... probably along with the business cycles... but does it really predict the returns over the next 10 years

Photoguy says it is between minus 5% to 10%... that seems like a normal 10 year change in the market anyhow...

When I get some time.... I will try and read up a bit more...

 

[donheff]

The graphs are interesting. Clearly an association between PE10 value and 10 year performance but also clearly not enough to predict anything useful for the next few years, including what PE10 will be then. I'm going to stick with my AA and keep my eye out for those 3.5% TIPS. My .83/PE10 spreadsheet cell will stay as an interesting reminder of what Schiller thinks the appropriate tweeks on business as usual might be.

If PE10 gets way out in marginland it would make sense to change AA but as always, when to do so becomes the impossible question.

 

[CCdaCE]

Morningstar's tool thinks "The Market" is fairly valued today at 1.01.

Market Fair Value by Sector, Industry, Super-Sector, Index | Morningstar

Who ya gonna believe?

Says Energy stocks and Basic Materials are undervalued if you're bottom fishing sector-wise, but, that wasn't really the question, I guess.

-CC

 

[EvrClrx311]

Another problem with this view is that companies now can grow fast... Google and Facebook were not companies 10 years ago... Google is worth over $250 billion, Facebook over $64 billion...

I've often thought what affect if any this has on these historical models. 80 years ago companies needed profits and products and infrastructure, just like they still do today, however on a different scale. You can produce software for cheap and sell for massive profits... as one example. I'm hesitate to assume that P/E just deserve to be higher moving forward than average historical figures suggest, but is is tempting to try and make such assumptions around the data. At least when trying to predict things will get better

What I find at a crossroads is the P/E data suggesting we have a lot more average to bad times ahead... when looking at the fact that we're exiting the worst 13-14 year period in market history (suggesting that VERY good times are coming soon). Can't both be right.

 

[haha]

What I find at a crossroads is the P/E data suggesting we have a lot more average to bad times ahead... when looking at the fact that we're exiting the worst 13-14 year period in market history (suggesting that VERY good times are coming soon). Can't both be right.

I think you are conflating factual observations with interpretation of them. True, PE10 is what it is, and true that these levels do not suggest unusually high, or even average returns on S&P 500 over the longer term.

And true that we are exiting a bad performance for over a decade. But IMO your idea that this suggests very good returns going forward is flawed. You can stand atop a 5000"foot mountain and with an overall elevation loss of 5000' you should be at sea level. But if you stand atop Mt Rainier and descend 5000', you still have a very long way to go. One could look at the 2000 top as Mt Rainier, or even Denali.

Ha

 

[photoguy]

Here's a really interesting article discussing PE10 and some of it criticisms (and defenses) by Asness:

http://www.aqr.com/Portals/1/ResearchPapers/ShillerPECommentary_AQRCliffAsness.pdf

Some excerpts:

"Ten-year forward average returns fall nearly monotonically as starting Shiller P/E’s increase. Also, as starting Shiller P/E’s go up, worst cases get worse and best cases get weaker (best cases remain OK from any decile, so there is generally hope even if it should not triumph over experience!)."

"If today’s Shiller P/E is 22.2, and your long-term plan calls for a 10% nominal (or with today’s inflation about 7-8% real) return on the stock market, you are basically rooting for the absolute best case in history to playout again, and rooting for something drastically above the average case from these valuations."

"So why do some people dismiss today’s high Shiller P/E, saying it’s not a problem?... They point out that we had two serious earnings recessions recently (though only the tail end of the 2000-2002 event makes it into today’s Shiller P/E), including one that was a doozy following the 2008 financial crisis. They thus feel the final (the right end of the graph) strong earnings number is more relevant than that of the prior 10 years."

"Not surprisingly, if one compares one-year earnings to history, things look much rosier, though the stock market is still not cheap. Instead of a Shiller P/E that is in the 80th percentile versus history right now (expensive), the one-year P/E is in the 54th percentile since 1926 (trivially expensive)"

The last quote is one I think is critical -- basically a 60/40 portfolio can expect a real return of just 2.2%:

"To get the expected real return on stocks you invert the Shiller P/E to get an earnings yield. For bonds we compare 10- year nominal yields on Treasury Bonds to forecasted long-term inflation (here we use nothing fancier than trailing three- year inflation). Take 60% of the stock market number and 40% of the bond market number and you get the above graph, an estimate of the expected long-term forward looking real return on the whole 60/40 portfolio. Needless to say it’s now very low versus history (2.2% real as of September 30, 2012)."

 

[haha]

Here's a really interesting article discussing PE10 and some of it criticisms (and defenses) by Asness:

http://www.aqr.com/Portals/1/ResearchPapers/ShillerPECommentary_AQRCliffAsness.pdf

Some excerpts:

"Ten-year forward average returns fall nearly monotonically as starting Shiller P/E’s increase. Also, as starting Shiller P/E’s go up, worst cases get worse and best cases get weaker (best cases remain OK from any decile, so there is generally hope even if it should not triumph over experience!)."

"If today’s Shiller P/E is 22.2, and your long-term plan calls for a 10% nominal (or with today’s inflation about 7-8% real) return on the stock market, you are basically rooting for the absolute best case in history to playout again, and rooting for something drastically above the average case from these valuations."

"So why do some people dismiss today’s high Shiller P/E, saying it’s not a problem?... They point out that we had two serious earnings recessions recently (though only the tail end of the 2000-2002 event makes it into today’s Shiller P/E), including one that was a doozy following the 2008 financial crisis. They thus feel the final (the right end of the graph) strong earnings number is more relevant than that of the prior 10 years."

"Not surprisingly, if one compares one-year earnings to history, things look much rosier, though the stock market is still not cheap. Instead of a Shiller P/E that is in the 80th percentile versus history right now (expensive), the one-year P/E is in the 54th percentile since 1926 (trivially expensive)"

The last quote is one I think is critical -- basically a 60/40 portfolio can expect a real return of just 2.2%:

"To get the expected real return on stocks you invert the Shiller P/E to get an earnings yield. For bonds we compare 10- year nominal yields on Treasury Bonds to forecasted long-term inflation (here we use nothing fancier than trailing three- year inflation). Take 60% of the stock market number and 40% of the bond market number and you get the above graph, an estimate of the expected long-term forward looking real return on the whole 60/40 portfolio. Needless to say it’s now very low versus history (2.2% real as of September 30, 2012)."

This should not be a surprise to anyone. Yet it will have small effect on ER plans.

Ha

 

[clifp]

This should not be a surprise to anyone. Yet it will have small effect on ER plans.

Ha

While it agree it not surprising and I also agree a not lot of retirement plans will change based on this info. I am curious assuming that you accept P/E 10 is decent forecasting tool for predicting future stock market returns.

What are you going do differently? Now perhaps you don't need to do anything because of age and portfolio size.

But lets say that you had just retired at 55 and you are depending on your portfolio to provide 1/2 your income (SS+ small government pension provide the rest.). You had planned on a 3.5% withdrawal rate and 60/40 portfolio.

Are you going to do anything differently know that the most likely real return of a 60/40 portfolio over the next decade is 2.2%?

 

[audreyh1]

My problem with using PE/10 right now is the massive losses in 2009 distort the measure.

Will stocks suddenly become more valuable in 2019 when that year falls off the PE/10 measurement?

Yeah - load up in 2019!!!

 

[audreyh1]

2.2% real return is good enough. It's the <1% real return that would worry me.

 

[Midpack]

2.2% real return is good enough. It's the <1% real return that would worry me.

+1. I've run all out plans on 0-2%, so 2.2% would be wonderful for us provided the sequence of returns isn't too crazy.

 

[MichaelB]

This should not be a surprise to anyone. Yet it will have small effect on ER plans.

Ha

You may be right.

2.2% real return is good enough. It's the <1% real return that would worry me.

2.2% real return is not enough to satisfy a 4% withdrawal rate over 30 years. Using FIRECalc, plugging in the 2.2% rate and assuming a low rate of volatility (4% SD) it fails 1/3 of the time.

 

[haha]

While it agree it not surprising and I also agree a not lot of retirement plans will change based on this info. I am curious assuming that you accept P/E 10 is decent forecasting tool for predicting future stock market returns.

What are you going do differently? Now perhaps you don't need to do anything because of age and portfolio size.

But lets say that you had just retired at 55 and you are depending on your portfolio to provide 1/2 your income (SS+ small government pension provide the rest.). You had planned on a 3.5% withdrawal rate and 60/40 portfolio.

Are you going to do anything differently know that the most likely real return of a 60/40 portfolio over the next decade is 2.2%?

I agree it is a difficult situation. What I do, and have done, and likely will do in the future, is lean more toward market reflecting and higher volatility securities when PE10 appears to suggest a stronger return, and lower volatility securities or securities that are more likely to respond to other issues than S&P 500 movements when PE10 is high. Also, when interest rates are higher, I will go more into cash if it can be done with moderate or small ltcg cost.

Essentially, in a loose way my allocation is informed by the level of PE10. Incidentally, there are other valid long term market valuation tools that approach valuation from a different angle, but usually more or less in concert with PE10. One is described in Andrew Smithers' book, Valuing Wall Street.

Ha

 

[haha]

You may be right.



2.2% real return is not enough to satisfy a 4% withdrawal rate over 30 years. Using FIRECalc, plugging in the 2.2% rate and assuming a low rate of volatility (4% SD) it fails 1/3 of the time.

What make this even more challenging is that low returns are usually a product of higher volatility, not lower.

IMO it is not a happy prospect.

Ha

 

[EvrClrx311]

I think you are conflating factual observations with interpretation of them. True, PE10 is what it is, and true that these levels do not suggest unusually high, or even average returns on S&P 500 over the longer term.

And true that we are exiting a bad performance for over a decade. But IMO your idea that this suggests very good returns going forward is flawed. You can stand atop a 5000"foot mountain and with an overall elevation loss of 5000' you should be at sea level. But if you stand atop Mt Rainier and descend 5000', you still have a very long way to go. One could look at the 2000 top as Mt Rainier, or even Denali.

Ha

Just trying to wrap my head around the trends. I definitely get that the future can deviate from the past, but the more past you include the less likely you are to see something new into the future. When looking back 150 years, every period of time where the market has performed as bad as it has the last decade, we've seen a very positive run in the market the next couple decades. What is also true is that in all those periods people thought times were different and stated a number of reasons why the market couldn't/wouldn't get back to its long term 10% yearly return trend.

Market tends to do the opposite of what people expect... particularly near its peaks ("bliss, this bubble will never burst") and troughs ("the economy has changed, we'll never get it back").

There is evidence to support both directions. I think more people are on the side of caution right now because all we've seen the last decade is wild volatility - next up is the drop.

Question is... when do we break this cycle and return to normal (on a historical scale)? On that 10.5% whole market trend we've seen historically... the DOW today would need to be around 40,500 to be as "dangerously" above the long term trend as it was in 2000 (that is what the 100+ year historical average 10.5% return on 11,700 in 2000 would be today). The lost decade... did it really fix anything? Doesn't appear it did, since we still think 15,000 is overpriced, and 40,500 seems looney to even fathom hitting this decade, or the next.

Inflation has a lot to do with this as well. Historically speaking, it has been very low for a while. If you wanted it to revert to it's long term mean you'd have to expect 4-6% levels for the next decade. That would certainly help fuel pushing stocks to double in value between now and 2023. Though the cost of everything would also go up 50-75%.

An interesting read from May of 2000 in Time magazine:
Will the DOW ever hit 50,000?

HASSETT There is no question that if you run through history, Bob is right--that over some periods following peaks in the P/E, there were some bad times. My problem with that exercise is that if you took anything, any metric of how the market is doing, and calculated the average over time, then, of course, when you are above it, then you go down, and when you are below it, you go up.

SHILLER Not necessarily, not with the P/E ratio. You are suggesting there is some spurious--some fallacy in that. There isn't a fallacy here. I look at past historical periods when we had similar leveling, and you know, what comes to mind is 1929. We have had a tripling of the stock market to a record high level in the past five years, and there is only one other time when that has happened, which was '24 to '29. So history doesn't encourage me to think people have suddenly learned something.

GLASSMAN I just think people should know that in 1990 the Dow was at 2,600. In 1995 it [finished] at 5,100. Now, if you are Bob, you would say, "Whoa, we are about to have another crash like 1929." What has it done since the end of 1995? We are now, as we sit here, above 11,000 [the Dow closed at 10,609 last week], so I don't think it is predictive to say simply that if the Dow triples, it is going to crash. In the past 18 years, the Dow has risen by a factor of 14. What we're saying is that something profound is going on.

For what its worth, the DOW at 2,600 in 1990 growing at the long term 10% market trend would put us at 23,000 today (I always forget if the DOW figure includes reinvesting dividends, if not... that is misleading by comparison). If one really thought the 10% average return seen for the last 100 years is to hold up, we could assume the market is cheaper today than it was in 1990. Is 10% the magic number? Or should it be 9%, or 8%... maybe the 20th century was a bubble in itself, and that 10% figure won't hold up in the 21st.

Traveling ahead 100 years in a time machine and looking back at 200 years of stock market data... where do we stand today on the long term trend. Above the line or below? Hard to tell. I think it is safe to say that 2000 was well above whatever line will be drawn, and I think we all hope 2009 was well below it.