Where are we in the stock market cycle?

[audreyh1]

S&P500 crossed 2100! OMG!

 

[Totoro]

With the Sp500 reaching all time highs again, I am amazed how it just keeps going up. It has not been a smooth ride. There have been been bumps in the road. Yet, not that I am asking for a correction or even worse, I can't help but wonder, where are we? I am not a market timer, but I do feel a sense of responsibility to myself if I can ,to understand if the markets are overvalued, fully valued , or if earnings are good, undervalued? Now I know there will be those who say it does not matter. To those who say that, you are correct, to a point. But to get more knowledge we have to ask why things are the way they are. So I am asking.

I tend to look at several things:

• Shiller P/E
• World GDP vs. total world market cap
• Bond yields vs. earnings yield
• Inflation
• Unemployment numbers
• Some annual reports
• Reports of major sectors in trouble. Hindsight 20/20 and all, but the 2008 implosion was not a new problem. It was relatively well known in 2007. Same thing with tech in 2001.

I'm wondering about one big indicator right now: the real yield on stocks.

Everytime it went negative in the past 100 years or so big trouble was ahead, with only a few months warning.

Real yield is E/P - inflation. If you take the 10 year E/P (3.7%), inflation is around 1%, so we are still far away from the cliff according to that measure.

Another one is long term corporate bond yields (VCLT vanguard for example), which are around 4.1% right now. That is higher than the E/P ratio, not a good sign. Stocks are supposed to have a premium vs. bonds as they are junior in rank.

In my view the US market is moderately overvalued, but not extremely so given inflation and bond rates. Not enough deviation to make a bet on it though. Edit: to put a number on it: 20% or so.

Europe still has some ways to go to recover (in valuation and in actual performance). Have no opinion on emerging (Especially China).

What will happen, nobody knows.

 

[audreyh1]

I do notice that some intermediate bond funds are yielding around the same as the S&P500, and that's pretty unusual. Usually they yield more - in recent decades, that is.

 

[Ticker]

I am not a market timer, but I do believe in taking profits. I weigh the
current price/profit against the likelihood of further appreciation vs. further
risk. You don't have to be a long term holder of everything in your portfolio.

this morning I sold a winning position in the snp 500. I'll buy it back again sometime later this year when it drops back to a level I am more comfortable with, and one that has better profit potential/risk for my tastes. Meanwhile, I'll sleep better.

 

[Options]

"There are two kinds of investors, be they large or small: Those who don't know where the market is headed and those who don't know that they don't know."

- William Bernstein

 

[utrecht]

I am not a market timer, but I do believe in taking profits. I weigh the
current price/profit against the likelihood of further appreciation vs. further
risk. You don't have to be a long term holder of everything in your portfolio.

this morning I sold a winning position in the snp 500. I'll buy it back again sometime later this year when it drops back to a level I am more comfortable with, and one that has better profit potential/risk for my tastes. Meanwhile, I'll sleep better.

The second statement is pretty much the very definition of a market timer.

 

[NW-Bound]

... The real question is when is Fed going to raise rates. I very much doubt that they have balls to raise rates...

I'm 92.4% certain that JY doesn't have any... ...

I recall the couple of scenes in "Crocodile Dundee" where Hogan wanted to be sure, and did a manual check.

 

[razztazz]

The second statement is pretty much the very definition of a market timer.

Not necesarilly. "Market timer" connotes a person sitting there using either a calendar or a moving average or related metric who simply buys/sells based on that talisman/method.

Ticker just sounds like a prudent investor not slaved to either mechanical "timing" or buy and hold.

 

[imoldernu]

W2R's visual depiction of the market triggered some deep thought about the subject. So real...
So if I'm the guy in the middle seat, in the third car... (hands are not raised to show I'm not afraid)... what do I do now? There was the long ride up, enjoying the scenery, listening to the chain gears struggling to pull the cars to the top. The view is great.
Now I look down... it's long way down, and it's going to be a faster ride down, than the ride up. There are a lot of people on this ride along with me.
We can all see where this is going.

What to do? Gotta decide... Ride it down with everybody else? Get off before the ride down?

C'mon... seriously... Should I try to climb over my broker, who is sitting beside me? Even if I could, where would I go? Could I get off, quickly, and then carefully climb down the tracks or the frame? Then, even if I could, how would I catch up with the cars when they reach bottom, and start the next climb?

Most likely I'd reason that what goes down, must come back up... and stay in my seat... and like the rest of the riders... ride it down. When it reaches bottom again, it will go back up. Again, a slower ride up. Ooo.. and what if that trip down was the end of the ride?

Yup... overthinking again ...but the analogy was so clear. Maybe the picture was taken in 2008... and that was not the highest point on the ride. The car went down for a quick thrill... came back up and was now at the peak. What if this is the peak, now... and the ride ends at the bottom of the hill.

Unfortunately, I don't have time to buy another ticket.

 

[Lsbcal]

Of course, in real equity markets the picture includes many variations on that top with plenty of turbulence. And there is no end to the ride for the market. You age on that ride and eventually die in harness.

It's a game you have to play. Even sitting on the ground and watching the equity ride is part of that game.

 

[CaliforniaMan]

IIRC, there are number of people here who retired into a big down turn. They seem to have adjusted, rolled with the punches and have lived to fight another day.

My guess is that there is also a number who retired into a big down turn who did not adjust well, they are just not the ones posting here.

 

[CaliforniaMan]

I do not get the sense that this is a market where the good times will roll on forever. I am still a little mindful of the 2008/2009 market crash. I do feel like there is no set pattern to market up and downs. The stock market reacts on news and events and earnings of companies, and all those are unpredictable. Still, it would be nice if there was a pattern, but that is what makes it interesting and challenging and rewarding because there does not seem to be one.

Have a friend I see occasionally and talked to again a few weeks ago. We chatted a bit about the markets and he mentioned how unusually high the stock market is. We first met around 2009 and he is quite successful in his work, but as far as I know he doesn't own any stocks, everything is in bank savings. The interesting thing is that he has made that exact same observation every single year I have known him. There was never a time when the stock market was not too high, at a price that was supported by fundamentals, the economy, etc. If the market had been going down it proved how bad things are, and it will go down more. If it had been going up, well it has to go down to more reasonable levels.

I just always tell him that I don't know what it will do, but that I just never change my allocation. Over the past few years he seems to look at my view with a little less disdain, however if the market tanks again, I am sure he will tell me it was coming.

One poster mentioned that they had a technique they used to see if the stock market is overvalued or not. The problem is for anyone who knows a bit about multivariate statistics and models, knows how incredibly easy to make a model that predicts the past. And it can even sound sensible so it is easy to fool yourself that it is real. The only problem is that while we know the past, it is that damn future that is unpredictable, and the butterfly effect on top of unpredictable occurrences can cause small disturbances to propagate quickly throughout the market.

So in the end really nobody knows what stage of the cycle we are at, or even if there is a cycle. But history does tell us something. Adding a little risk can buy a bit more return over time than those savings accounts. that is all we really know. So you place your bet, and take your chances.

 

[Lsbcal]

...
One poster mentioned that they had a technique they used to see if the stock market is overvalued or not. The problem is for anyone who knows a bit about multivariate statistics and models, knows how incredibly easy to make a model that predicts the past. And it can even sound sensible so it is easy to fool yourself that it is real. The only problem is that while we know the past, it is that damn future that is unpredictable, and the butterfly effect on top of unpredictable occurrences can cause small disturbances to propagate quickly throughout the market.

So in the end really nobody knows what stage of the cycle we are at, or even if there is a cycle. But history does tell us something. Adding a little risk can buy a bit more return over time than those savings accounts. that is all we really know. So you place your bet, and take your chances.

Hi CaliforniaMan, you probably are not referring to my post which does mention a timing model but only to say that it is not suggesting a decline. I agree it is not possible to know the future. I do think that it is possible to model when the risks are increasing. For example, one Fed paper models how the yield curve has predicted business declines (and hence equity declines), see The Yield Curve as a Leading Indicator: Some Practical Issues by Estrella and Trubin.

Where I want to quibble a bit is in the blue text above. I've seen similar statements on Bogleheads before. It depends critically on what the model is capable of. For instance, I'm sure it is easy to develop a model that backtests well for a market cycle or two. Some models based on moving averages are often discussed as examples. They seem to filter out really bad declines like 2008 at the expense of whipsaws (unprofitable sell/buy trades). They might do comparatively poorly during equity bull markets like the 1990's and then shine in the next decade.

It is very difficult in my opinion to get a model that backtests well over say the 1920's to the present. The model should give better or equal results then buy/hold over periods of say 10 years as well as the full 90 years, have very few whipsaws, and be easy to implement. Also to be complete the defects in the model should be very clearly understood. For instance, the model might have missed some equity declines not clearly linked to yield curve inversion, or the model might have missed most "corrections" of not worse then -10%.

 

[HFWR]

It is very difficult in my opinion to get a model that backtests well over say the 1920's to the present. The model should give better or equal results then buy/hold over periods of say 10 years as well as the full 90 years, have very few whipsaws, and be easy to implement. Also to be complete the defects in the model should be very clearly understood. For instance, the model might have missed some equity declines not clearly linked to yield curve inversion, or the model might have missed most "corrections" of not worse then -10%.

All that, and a bag of chips...

 

[karluk]

All I can say is that I would be willing to make some modest changes in my asset allocation if Lsbcal's metrics started flashing a "sell" signal. I think I handled the 2007-2009 crash quite well, but I still regret completely disregarding the inversion of the yield curve that preceded it. I read a few articles at the time that suggested it was quite a dire signal indeed, but "buy and hold, ignore the noise" carried the day for me.

I wouldn't be willing to completely abandoned equities based on any sell signal, no matter how reliable it's been in the past. But given another yield curve inversion along with Lbscal's other criteria, I would probably cut back on equities by 5% or 10%.

 

[audreyh1]

As long as you guys don't start talking about Hindenburg Omens, I'll be happy.

 

[audreyh1]

We first met around 2009 and he is quite successful in his work, but as far as I know he doesn't own any stocks, everything is in bank savings. The interesting thing is that he has made that exact same observation every single year I have known him. There was never a time when the stock market was not too high, at a price that was supported by fundamentals, the economy, etc. If the market had been going down it proved how bad things are, and it will go down more. If it had been going up, well it has to go down to more reasonable levels.

I got caught in that trap in 1994. It was my first case of "talking head disease" - in other words, too much CNBC watching. There was a minor correction that year in interest rates. But then it turned out to be an OK year investment wise, and 95 was good too, and the heads kept talking about imminent corrections. I finally understood that no time ever feels safe to invest. You have to pick a plan that works in the long run and figure out how to stick to it. That's when I really learned about asset allocation and rebalancing, and averaging in, etc.

 

[Lsbcal]

All I can say is that I would be willing to make some modest changes in my asset allocation if Lsbcal's metrics started flashing a "sell" signal.

Well I'm not running a newletter and don't plan on issuing any signal. I might discuss the worries I have at the time, but hopefully would not be so presumptuous as to suggest others think the same way.

I handled the 2007-2009 crash quite well, but I still regret completely disregarding the inversion of the yield curve that preceded it. I read a few articles at the time that suggested it was quite a dire signal indeed, but "buy and hold, ignore the noise" carried the day for me.

I remember reading one Boglehead's contention on yield curve inversion being a bad signal. Some posters took great exception to this. He went on to talk about changing his long term buy-hold to a sell at that time (if I remember right). He even bought back in, I think in around Feb 2009. The guy finally gave up on the Boglehead's site after getting a lot of grief and he was a long termer too.

Some of the people over there are too fanatical IMO, but there are interesting threads and good recommendations too. Unfortunately, the fanatics drive out the alternate opinions over there. Blue rabbits have to be careful on Bogleheads.

Keeping an open mind on investing is a good thing.

I wouldn't be willing to completely abandoned equities based on any sell signal, no matter how reliable it's been in the past. But given another yield curve inversion along with Lbscal's other criteria, I would probably cut back on equities by 5% or 10%.

For me it depends on the conditions. If I'm older and in fine financial condition, maybe I will reduce my equities considerably. Most of us should have an AA that can withstand moderately severe declines. Too many of us would suffer a great deal in a 50% decline and I have no wish to repeat 2008 ... or worse should a recovery like we've had not repeat.

 

[mathjak107]

I got caught in that trap in 1994. It was my first case of "talking head disease" - in other words, too much CNBC watching. There was a minor correction that year in interest rates. But then it turned out to be an OK year investment wise, and 95 was good too, and the heads kept talking about imminent corrections. I finally understood that no time ever feels safe to invest. You have to pick a plan that works in the long run and figure out how to stick to it. That's when I really learned about asset allocation and rebalancing, and averaging in, etc.

i finally put the finishing touches on my plan for descent into retirement land in july.

i was very heavy in bonds and about 30% equities the last few years. bonds did fine and were meeting goal because of capital appreciation.

but they have slowed down considerably so i finished boosting my equity level to 40% .

that includes cash . so the spending down of cash over the first 2 years should give me a rising glide path up to about 50/50 which is the limit of my comfort range.

i have always stayed dynamic and adjusted to the world around me as it changes and will continue to do so as the big picture changes


i think if the party ends at midnight we are at 10 or 11pm both in bonds and perhaps equities although i would stick with my equity funds while finding more appropriate fixed income investments when it is time .

 

[pb4uski]

....I'll buy it back again sometime later this year when it drops back to a level I am more comfortable with, and one that has better profit potential/risk for my tastes. Meanwhile, I'll sleep better.

What if it doesn't drop back but keeps going up with only minor corrections? Quite possible given a recovering economy and being out of the market could have a lot of opportunity cost associated with it.

I know people who bailed back in 2008, never got back in and regret it.

 

[target2019]

What if it doesn't drop back but keeps going up with only minor corrections? Quite possible given a recovering economy and being out of the market could have a lot of opportunity cost associated with it.

I know people who bailed back in 2008, never got back in and regret it.

Here is a 20-year chart for S&P500, with a 200-SMA indicator. It shows that what goes up must come down. Also shows what goes down must go up. If you get out at, say, 2100, what signal would you use to begin re-investing? I know it is possible to make educated guess about this, but I'd be very worried about the timing to get back in. Now if it was OPM, I wouldn't be much worried!

 

[mathjak107]

I am not a market timer, but I do believe in taking profits. I weigh the
current price/profit against the likelihood of further appreciation vs. further
risk. You don't have to be a long term holder of everything in your portfolio.

this morning I sold a winning position in the snp 500. I'll buy it back again sometime later this year when it drops back to a level I am more comfortable with, and one that has better profit potential/risk for my tastes. Meanwhile, I'll sleep better.

the problem with the i will get back in later when things drop for most it requires two very different personality traits in one person.

the person who gets out during a run up to play it save and potentially leaves gains on the table is generally the nervous nellie type.

all justified and okay .

however the personality type to buy back in during or after a bad drop takes nerves and lots of pucker factor to do.

afterall , you bailed just thinking about a drop so how are you going to act when things look so poor around you and the market is falling ?

odds are few can pull this off because they do not have twin personality's .

 

[UnrealizedPotential]

One thing I have noticed is investors unwillingness to push the SP500 index much past the 2090 range. My guess is they are looking for better earnings to justify higher prices at this level. If so, they should. We have been in the 2090 range quite a bit, it never seems to go much beyond that. I think that is healthy for the market. It tells me that earnings matter to investors.

 

[CaliforniaMan]

Hi CaliforniaMan, you probably are not referring to my post which does mention a timing model but only to say that it is not suggesting a decline. I agree it is not possible to know the future. I do think that it is possible to model when the risks are increasing. For example, one Fed paper models how the yield curve has predicted business declines (and hence equity declines), see The Yield Curve as a Leading Indicator: Some Practical Issues by Estrella and Trubin.

Where I want to quibble a bit is in the blue text above. I've seen similar statements on Bogleheads before. It depends critically on what the model is capable of. For instance, I'm sure it is easy to develop a model that backtests well for a market cycle or two. Some models based on moving averages are often discussed as examples. They seem to filter out really bad declines like 2008 at the expense of whipsaws (unprofitable sell/buy trades). They might do comparatively poorly during equity bull markets like the 1990's and then shine in the next decade.

It is very difficult in my opinion to get a model that backtests well over say the 1920's to the present. The model should give better or equal results then buy/hold over periods of say 10 years as well as the full 90 years, have very few whipsaws, and be easy to implement. Also to be complete the defects in the model should be very clearly understood. For instance, the model might have missed some equity declines not clearly linked to yield curve inversion, or the model might have missed most "corrections" of not worse then -10%.

OK, maybe I overstated a bit with the "incredibly easy" remark. However I do believe it is possible to "predict" the past with quite a bit of precision, certainly enough to beat buy and hold. There are two basic approaches I had in mind. I am sure there are many others but these I am most familiar with.

The first is based on regression techniques (PC and PLS) and the second is searching through some mathematical space to minimize or maximize (or both) some parameters. The seach can be systematic or heck it can even be random, computers are fast and will happily run for hours for you without complaint. (Not so for the programmers of course, but that is a different discussion).

You can use real series, or even random numbers, smooth them to make them look good (and also having the effect of adding additional information). Choose levels or crossovers and with different smoothings with multiple series, and let the computer try millions of combinations. Some will be better than buy and hold, some a lot better. And really you only have to catch about a half dozen big bear markets, maybe a dozen, since 1920 to look pretty good. Try to use as few series as possible for the desired effect.

While it may be not "incredibly easy", I believe it is possible and I have used these techniques a number of times in engineering, not finance, in my past life (yes, I was once an actual w*rker). But please don't call my bluff and ask me to do it now, it is too much like actual w*rk, yes and OK, not "incredibly easy" work.

But I do stand by my statement that it can be done, and can "incredibly easily" fool you into thinking it makes sense, and that you can predict the future as well as you can the past.

I don't know, maybe sometime in the future I will make some test runs, but right now I am enjoying my first year of freedom too much and can't even imagine myself back doing statistical programming.

 

[Lsbcal]

Hi CaliforniaMan, I'm glad we got rid of the "incredibly easy" part.

Another thought, suppose one had a technique that shows good results for sell/buy pairs in many market downturns of the past. Maybe this model is not actually predicting? Maybe it is only showing a way to follow a trend which may grow or shrink as time moves forward. We know that academics now are giving a lot of credit to momentum in the markets. As I understand it there is even some new term added into the French-Fama work to take account of momentum.

So what I'm saying is that a good model does not have to predict the future (which is impossible anyway). It only can indicate a trend to follow forward. If the model is good enough, the trend will either play out to the users benefit or there will be an occasional modest loss (or lost gains) in taking the risk and getting out of the market. Also importantly, even a good model should be expected to have a modest loss but only occasionally.

Also it is important to avoid models which have many sell/buy pairs. Many moving average approaches (filters not models?) suffer from this. One 200 day moving average method I looked at had about 1.6 sell/buy pairs per year. I think a model should aim for maybe one sell/buy every 4 years or so.