You Can Market Time At Valuation Extremes

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
reformed.trader
Posts: 76
Joined: Sat Apr 15, 2017 11:14 pm

Re: You Can Market Time At Valuation Extremes

Post by reformed.trader » Sun Feb 17, 2019 2:00 pm

Just use CAPE or any other of your favorite valuation metrics as a cross sectional tools, rather than a market timing tool.

User avatar
gmaynardkrebs
Posts: 1660
Joined: Sun Feb 10, 2008 11:48 am

Re: You Can Market Time At Valuation Extremes

Post by gmaynardkrebs » Sun Feb 17, 2019 2:02 pm

mariezzz wrote:
Sun Feb 17, 2019 1:29 pm
gmaynardkrebs wrote:
Sun Feb 17, 2019 8:32 am
mariezzz wrote:
Sun Feb 17, 2019 12:45 am
Somewhere on bogleheads is an old thread which mentioned that missing just a few of the top earnings days a year would massively affect your personal earnings ... if you try to market time, you likely miss those days, was part of the discussion.
Except that over whatever period of time you are out, wouldn't you miss all the down days as well?
The points made in the thread indicated that the logic you seem to be following was wrong. (You could just stay out of the market all together if your ultimate goal was to miss the best & worst days in the market.)

The arguments presented in the thread I referred to was that long term, you're better off getting in & staying in ... and weathering the ups &, downs, all of them - in other words, market timing doesn't work as a strategy long-term. That's the entire premise of this forum, in case that wasn't apparent.

You seem to think it can work, from other responses here.

(But there's a lot of research that suggests a subset of people are more strongly influenced by their wins (when they tried to market time & it seemed to work) than by their losses (when it didn't work). This is the same mechanism which tends to drive gamblers across the board. This unequal assessment of outcomes lead some to believe (based on anecdotal evidence) that it can work.)
No, I don't practice or believe in"market timing." I was merely questioning the logic behind the "missing the best days" argument. It might be valid if big up days are both rarer and of greater magnitude than down days, but only over shorter periods. Over long periods. since the ending price must be the net of all the up days vs down days, being out or in here or there shouldn't matter very much. Your going to catch all or most of both ups and downs. FWIW, I did look at a couple of threads, but didn't find the one you probably have in mind.

comeinvest
Posts: 290
Joined: Mon Mar 12, 2012 6:57 pm

Re: You Can Market Time At Valuation Extremes

Post by comeinvest » Sun Feb 17, 2019 8:35 pm

Why not invest internationally or globally, and the problem is solved? (If the problem is whether or not to stay invested at current U.S. valuations, or how to reduce exposure to extreme valuations and the associated potentially higher risk while keeping your assets invested at all times.) Most markets currently trade close to, or below, their own long-term averages, and most trade at valuations well below that of U.S. markets.
Several options here:
- invest in total world stock market
- tilt towards markets with lower valuations using rules based on relative valuations (this has been discussed in other threads, and defeated by some, while supported by others)
- split your assets across global markets according to ratios that you set, re-balance periodically, and stay the course
I realize the second effectively represents market timing to some extent, but the others don't.

Valuethinker
Posts: 38961
Joined: Fri May 11, 2007 11:07 am

Re: You Can Market Time At Valuation Extremes

Post by Valuethinker » Mon Feb 18, 2019 3:11 am

comeinvest wrote:
Sun Feb 17, 2019 8:35 pm
Why not invest internationally or globally, and the problem is solved? (If the problem is whether or not to stay invested at current U.S. valuations, or how to reduce exposure to extreme valuations and the associated potentially higher risk while keeping your assets invested at all times.) Most markets currently trade close to, or below, their own long-term averages, and most trade at valuations well below that of U.S. markets.
Several options here:
- invest in total world stock market
- tilt towards markets with lower valuations using rules based on relative valuations (this has been discussed in other threads, and defeated by some, while supported by others)
- split your assets across global markets according to ratios that you set, re-balance periodically, and stay the course
I realize the second effectively represents market timing to some extent, but the others don't.
European valuations e.g. of banks trading at 0.3x book. Reflect lower prospects of growth and ongoing financial problems.

US valuations reflect the high growth of the tech sector.

The market is not an idiot. It rates these stocks fairly given what it knows

comeinvest
Posts: 290
Joined: Mon Mar 12, 2012 6:57 pm

Re: You Can Market Time At Valuation Extremes

Post by comeinvest » Tue Feb 19, 2019 8:56 pm

Valuethinker wrote:
Mon Feb 18, 2019 3:11 am

European valuations e.g. of banks trading at 0.3x book. Reflect lower prospects of growth and ongoing financial problems.

US valuations reflect the high growth of the tech sector.

The market is not an idiot. It rates these stocks fairly given what it knows
Generally agreed, although I'm not sure if "high growth" justifies high valuations. However, the problem of whether or not to stay fully invested at current high U.S. valuations, is solved by e.g. 1/3 U.S. 1/3 int'l developed 1/3 EM.

User avatar
gmaynardkrebs
Posts: 1660
Joined: Sun Feb 10, 2008 11:48 am

Re: You Can Market Time At Valuation Extremes

Post by gmaynardkrebs » Tue Feb 19, 2019 9:43 pm

comeinvest wrote:
Tue Feb 19, 2019 8:56 pm
Valuethinker wrote:
Mon Feb 18, 2019 3:11 am

European valuations e.g. of banks trading at 0.3x book. Reflect lower prospects of growth and ongoing financial problems.

US valuations reflect the high growth of the tech sector.

The market is not an idiot. It rates these stocks fairly given what it knows
Generally agreed, although I'm not sure if "high growth" justifies high valuations. However, the problem of whether or not to stay fully invested at current high U.S. valuations, is solved by e.g. 1/3 U.S. 1/3 int'l developed 1/3 EM.
I'd say ex-US is more like you're at least not paying above the sticker price. But you're not getting $100 above invoice either. I have upped ex-US lately, but with little relish. More like, where else do I go?

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 12:45 pm

I'm glad to see that my comments stirred up the discussion in this thread again. I'm not surprised to see the critical nature of the views reacting to my points. CAPE/8 in December wasn't a sudden discovery. I was somewhat aware of the problems, but I had not personally calculated a CAPE/8 and wasn't sure of the magnitude of the difference until that point. I fundamentally don't want to be out of the market, do believe in buy-and-hold, and don't believe in timing IN GENERAL. But, just as in life, I'm not going to pay twice the value for an item. If stocks are at extreme valuations, I am uncomfortable with them. Today, I am just as uncomfortable with them, even though I have more faith in the CAPE/8 number (at somewhere around 3 points lower than the actual CAPE).

If I had been in investing mode in 1996 -- I was a kid -- I'd have been just as nervous about the CAPE/10. However, the other metric I appreciate is the Buffet indicator TMC/GDP. In 1996, it did not exceed 90%. Given the divergent signals, I don't think I would have moved anything. Today, the Buffet indicator is at almost 140%. These metrics are flashing the same signal: the market is near its 100+ year extremes in valuation, exceeded only twice.

Now, suppose you had a standing rule not to be invested above a CAPE of 25 (an upper bound indicating the start point of excessive valuations). You'd have pulled out in 1996, and reinvested between June and December of 2002. The S&P 500 annualized return during that time was just 5.0% (div's reinvested). I don't have time now to do the research, but I'm fairly confident that money markets did better. Perhaps bonds did as well. The point is that even if you had made the decision to leave in '96, it probably didn't hurt you. AND! That period preceded THE most significant anomaly/bubble in US market history.

Regarding criticisms of my approach, I'm not denying my luck at all. In fact, I directly stated it. I had an emotional response when I saw Shiller and Buffet indicators align at valuations higher than any point in history. I think I saw Shiller at 32-34 and Buffet at 144-ish. Right now, I see them at 30 & 139. Something's wrong, or something fundamentally has changed in the way investors are acting. Unless there's a crash, I'm going to be shifting some money, this week. My taxable accounts will remain invested, because I don't want to pay short term gains on the money I just made. My retirement accounts are going to bear this adjustment for the proportion of the portfolio I need to achieve some measure of security against this craziness. I'm thinking something like 50-60% towards money markets and/or bond funds.
Last edited by Chris42163 on Thu Feb 21, 2019 9:51 pm, edited 1 time in total.

marcopolo
Posts: 2399
Joined: Sat Dec 03, 2016 10:22 am

Re: You Can Market Time At Valuation Extremes

Post by marcopolo » Thu Feb 21, 2019 12:57 pm

Chris42163 wrote:
Thu Feb 21, 2019 12:45 pm
I'm glad to see that my comments stirred up the discussion in this thread again. I'm not surprised to see the critical nature of the views reacting to my points. Christmas eve wasn't a sudden discovery. I was somewhat aware of the problems, but I had not calculated a CAPE/8 and wasn't sure of the magnitude of the difference. I fundamentally don't want to be out of the market, do believe in buy-and-hold, and don't believe in timing IN GENERAL. But, just as in life, I'm not going to pay twice the value for an item. If stocks are at extreme valuations, I am uncomfortable with them. Today, I am just as uncomfortable with them, even though I have more faith in the CAPE/8 number (at somewhere around 3 points lower than the actual CAPE).

If I had been in investing mode in 1996 -- I was a kid -- I'd have been just as nervous about the CAPE/10. However, the other metric I appreciate is the Buffet indicator TMC/GDP. In 1996, it did not exceed 90%. Given the divergent signals, I don't think I would have moved anything. Today, the Buffet indicator is at almost 140%. These metrics are flashing the same signal: the market is near its 100+ year extremes in valuation, exceeded only twice.

Now, suppose you had a standing rule not to be invested above a CAPE of 25 (an upper bound indicating the start point of excessive valuations). You'd have pulled out in 1996, and reinvested between June and December of 2002. The S&P 500 annualized return during that time was just 5.0% (div's reinvested). I don't have time now to do the research, but I'm fairly confident that money markets did better. Perhaps bonds did as well. The point is that even if you had made the decision to leave in '96, it probably didn't hurt you. AND! That period preceded THE most significant anomaly/bubble in US market history.

Regarding criticisms of my approach, I'm not denying my luck at all. In fact, I directly stated it. I had an emotional response when I saw Shiller and Buffet indicators align at valuations higher than any point in history. I think I saw Shiller at 32-34 and Buffet at 144-ish. Right now, I see them at 30 & 139. Something's wrong, or something fundamentally has changed in the way investors are acting. Unless there's a crash, I'm going to be shifting some money, this week. My taxable accounts will remain invested, because I don't want to pay short term gains on the money I just made. My retirement accounts are going to bear this adjustment for the proportion of the portfolio I need to achieve some measure of security against this craziness. I'm thinking something like 50-60% towards money markets and/or bond funds.

In your previous post you made a big deal about the realization you had on Christmas eve about how CAPE10 was not the right metric, maybe its CAPE8, and there are other adjustments, so it is really at 22, so you jumped back in. OK, so why are you now going back to using CAPE10 (without any adjustments) and wringing your hands about it being at 30? Should all those brilliant insights you had on Christmas Eve still apply?
Once in a while you get shown the light, in the strangest of places if you look at it right.

User avatar
HomerJ
Posts: 13338
Joined: Fri Jun 06, 2008 12:50 pm

Re: You Can Market Time At Valuation Extremes

Post by HomerJ » Thu Feb 21, 2019 1:21 pm

Chris42163 wrote:
Thu Feb 21, 2019 12:45 pm
I'm glad to see that my comments stirred up the discussion in this thread again. I'm not surprised to see the critical nature of the views reacting to my points. Christmas eve wasn't a sudden discovery. I was somewhat aware of the problems, but I had not calculated a CAPE/8 and wasn't sure of the magnitude of the difference. I fundamentally don't want to be out of the market, do believe in buy-and-hold, and don't believe in timing IN GENERAL. But, just as in life, I'm not going to pay twice the value for an item. If stocks are at extreme valuations, I am uncomfortable with them. Today, I am just as uncomfortable with them, even though I have more faith in the CAPE/8 number (at somewhere around 3 points lower than the actual CAPE).

If I had been in investing mode in 1996 -- I was a kid -- I'd have been just as nervous about the CAPE/10. However, the other metric I appreciate is the Buffet indicator TMC/GDP. In 1996, it did not exceed 90%. Given the divergent signals, I don't think I would have moved anything. Today, the Buffet indicator is at almost 140%. These metrics are flashing the same signal: the market is near its 100+ year extremes in valuation, exceeded only twice.

Now, suppose you had a standing rule not to be invested above a CAPE of 25 (an upper bound indicating the start point of excessive valuations). You'd have pulled out in 1996, and reinvested between June and December of 2002. The S&P 500 annualized return during that time was just 5.0% (div's reinvested). I don't have time now to do the research, but I'm fairly confident that money markets did better. Perhaps bonds did as well. The point is that even if you had made the decision to leave in '96, it probably didn't hurt you. AND! That period preceded THE most significant anomaly/bubble in US market history.

Regarding criticisms of my approach, I'm not denying my luck at all. In fact, I directly stated it. I had an emotional response when I saw Shiller and Buffet indicators align at valuations higher than any point in history. I think I saw Shiller at 32-34 and Buffet at 144-ish. Right now, I see them at 30 & 139. Something's wrong, or something fundamentally has changed in the way investors are acting. Unless there's a crash, I'm going to be shifting some money, this week. My taxable accounts will remain invested, because I don't want to pay short term gains on the money I just made. My retirement accounts are going to bear this adjustment for the proportion of the portfolio I need to achieve some measure of security against this craziness. I'm thinking something like 50-60% towards money markets and/or bond funds.
It appears that the "Warren Buffett indicator" wasn't mentioned in mainstream media until a 2001 Forbes article.

As usual, an existing model failed (CAPE), so market-timers have to add in another variable (find one that fits past data), to handle the last exception.

Discovered Rule: "Every October in an election year, the stock market does this!"

Doesn't happen the next time.

New Rule: "Every October in an election year, except when the Red Sox win the World Series, the stock market does this!"

Doesn't happen the next time.

New New Rule: "Every October in an election year, except when the Red Sox win the World Series, OR the apple harvest is bad, the stock market does this!"

It's 100% perfect in backtesting!

And so on.
The J stands for Jay

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 1:25 pm

@marcopolo

On Christmas eve, Cape was lower than today, and CAPE/8 was under 25. I haven't recalculated and for that reason, CAPE/10 is what I'm looking at & referencing again, but I believe I addressed your question in the first paragraph where I stated that CAPE/8 should be roughly 3 points lower than the CAPE/10.

You're coming across a little aggressively -- sarcastic: "brilliant insights" -- and I'm here posting because I'm looking for wisdom and constructive discussion. If you'd like to make it personal, please find another forum, or I'll just ignore you.

User avatar
HomerJ
Posts: 13338
Joined: Fri Jun 06, 2008 12:50 pm

Re: You Can Market Time At Valuation Extremes

Post by HomerJ » Thu Feb 21, 2019 1:25 pm

Chris42163 wrote:
Thu Feb 21, 2019 12:45 pm
I need to achieve some measure of security against this craziness. I'm thinking something like 50-60% towards money markets and/or bond funds.
Welcome to the Bogleheads forum.

I suggest you read some articles on the wiki, so you understand our basic investing philosophy around here.

Market-timing and trying to predict the future based on "signals" is normally frowned upon.

But, of course, we often have discussions about it. Just trying to let you know, as a new member, what kind of forum this is.

It would be like joining a "Ford" forum, and your first few posts are all about the new Chevy trucks that are coming out next year.
The J stands for Jay

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 1:46 pm

HomerJ wrote:
Thu Feb 21, 2019 1:21 pm

It appears that the "Warren Buffett indicator" wasn't mentioned in mainstream media until a 2001 Forbes article.

As usual, an existing model failed (CAPE), so market-timers have to add in another variable (find one that fits past data), to handle the last exception.

Discovered Rule: "Every October in an election year, the stock market does this!"

Doesn't happen the next time.
I appreciate that objection, but I do think there's wisdom in using in metrics that get at value... We're purchasing something. Thus, we should have some idea of the value of our purchase before we determine a price we're willing to pay. Likewise, if someone offers you twice the market value of your car, you should really consider selling. It might be worth more than market value to you. So, perhaps you wouldn't sell. But, I think any investor should have some concept of the value of the equities they own.

The amount of business being done in the united states should be correlated, though not perfectly, to the amount of profits (earnings) businesses generate. It's not only an intuitively reasonable metric, it's simple, it correlates well with the past (as you say), and it comes from a well-regarded expert in the field. That's enough for me to lend it some credibility. The same goes for the CAPE.

Lastly, your objection would dismiss every metric as valueless, regardless of the truth. I don't think that's a wise approach, even if most metrics are bad. Even if they are ALL bad, they may still be valuable for the thoughts and intuitions they provoke.

Thank you for your feedback.

Chris

User avatar
HomerJ
Posts: 13338
Joined: Fri Jun 06, 2008 12:50 pm

Re: You Can Market Time At Valuation Extremes

Post by HomerJ » Thu Feb 21, 2019 1:51 pm

Chris42163 wrote:
Thu Feb 21, 2019 1:46 pm
Lastly, your objection would dismiss every metric as valueless, regardless of the truth. I don't think that's a wise approach, even if most metrics are bad. Even if they are ALL bad, they may still be valuable for the thoughts and intuitions they provoke.
Personally, I think it's a wiser approach to ignore all metrics, because most are bad. Since it's very difficult to tell which ones are the good ones, there's little value to making large investment changes (and often!) using metrics, when the odds are high that they will be wrong.

I hate it when people say "Well, CAPE is a bad tool, but it's the best of the bad tools..."

Well, if all you have is a screwdriver and a saw, you can say the screwdriver is the BEST tool you have to drive in nails, but it's still pretty terrible for that purpose.

One doesn't need to look at stock market metrics to save, invest, and get wealthy over time. You made large changes in the span of months, even when the metrics you are following didn't change that much. You may be just looking for external justification for your moves.
Last edited by HomerJ on Thu Feb 21, 2019 2:00 pm, edited 2 times in total.
The J stands for Jay

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 1:59 pm

HomerJ wrote:
Thu Feb 21, 2019 1:25 pm
Chris42163 wrote:
Thu Feb 21, 2019 12:45 pm
I need to achieve some measure of security against this craziness. I'm thinking something like 50-60% towards money markets and/or bond funds.
Welcome to the Bogleheads forum.

I suggest you read some articles on the wiki, so you understand our basic investing philosophy around here.

Market-timing and trying to predict the future based on "signals" is normally frowned upon.

But, of course, we often have discussions about it. Just trying to let you know, as a new member, what kind of forum this is.

It would be like joining a "Ford" forum, and your first few posts are all about the new Chevy trucks that are coming out next year.
Haha! Thank you! I'm not trying to ruffle feathers, but I don't want to live in my own bubble, either. I can't play devil's advocate well when I don't have a compelling counterargument. I want to have the discussion, and this thread seems like the right one. I'll do my best to be respectful and will leave if the subject matter is too divisive. I'd just ask for respectful dialogue, though I know that's a tall order on the internet. lol

marcopolo
Posts: 2399
Joined: Sat Dec 03, 2016 10:22 am

Re: You Can Market Time At Valuation Extremes

Post by marcopolo » Thu Feb 21, 2019 2:23 pm

Chris42163 wrote:
Thu Feb 21, 2019 1:25 pm
@marcopolo

On Christmas eve, Cape was lower than today, and CAPE/8 was under 25. I haven't recalculated and for that reason, CAPE/10 is what I'm looking at & referencing again, but I believe I addressed your question in the first paragraph where I stated that CAPE/8 should be roughly 3 points lower than the CAPE/10.

You're coming across a little aggressively -- sarcastic: "brilliant insights" -- and I'm here posting because I'm looking for wisdom and constructive discussion. If you'd like to make it personal, please find another forum, or I'll just ignore you.
I was being a bit sarcastic, not aggressive.

It appears you want to do some market timing and are looking for some validation that appears to quantitative.

CAPE10 itself did not really change all that much (32 --> 27 --> 30) in the time frame you are discussing. Certainly not enough to justify wholesale shift out, back in, and out again of equities. These were fairly routine gyrations. You made them seem like much bigger shifts by changing the ruler by which you are measuring. In early December you got out based on an unmodified CAPE being at 32. Then you got back in based on some modified metric you devised being at 22. Now that you want to get back in, you are going back to unmodified CAPE being at 30. So, it makes it look like the metric changed a lot (32 - 22 - 30). Still not huge, but maybe more reasonable to justify big moves. There is the danger that the moves are being driven more by emotion, and the changing metrics are being used to justify the moves.

You believe you have found a way to time the market. Maybe you are correct, and it has worked well for you last time, but the track record of people being able to do that successfully over the long run has not been that great.

Good luck to you.
Once in a while you get shown the light, in the strangest of places if you look at it right.

Boglegrappler
Posts: 1198
Joined: Wed Aug 01, 2012 9:24 am

Re: You Can Market Time At Valuation Extremes

Post by Boglegrappler » Thu Feb 21, 2019 2:47 pm

Among the many reasons that market timing is so likely to be wrong is that the market is a leading indicator of things.

I had a boss who noted that, within any given sector that was experiencing a cyclical peak or trough, there was a syndrome among the analysts---- that made sense. At market troughs, stocks would begin to rise before the fundamentals improved. Analysts would note this and pronounce the move to be a false rally. Sales weren't improving and things were in fact getting worse. It usually took six to nine months for the reality to catch up to the stock market, making it very difficult to forecast this.

Likewise, at market peaks, the market heads down before the fundamentals nose over. I had a friend who was an accountant for an oilfield services company. In late 1980, he asked me what was "wrong" with the company's stock. He noted that the stock was down significantly over the past 3-4 monks, and yet every day was a record day for operations. Well, in December of 1980, spot prices peaked and headed south, and continued that way for the next five or six years. Analysts typically aren't going to call a downturn in the market when sales and earnings are going strong. But somehow, the market itself seems to be able to anticipate these things. It's ironic that its so hard to do it as an individual investor.

User avatar
gmaynardkrebs
Posts: 1660
Joined: Sun Feb 10, 2008 11:48 am

Re: You Can Market Time At Valuation Extremes

Post by gmaynardkrebs » Thu Feb 21, 2019 2:59 pm

I think I might have said this earlier in the thread, but remember, Robert Shiller does not consider CAPE to be valid or useful tool for market timing. He's been super-clear about this. But ironically, if a lot of people think that CAPE can predict short term swings, even though they are mistaken, that might make it a useful timing tool after all -- at least until too many people start using it for that purpose. I still think that market timing is irrational, but maybe it's just a little bit irrational, especially if you like the thrill of gambling. But it's not a widows and orphans strategy, of that I am sure. And then, there's taxes and trading costs....

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 4:34 pm

HomerJ wrote:
Thu Feb 21, 2019 1:51 pm
One doesn't need to look at stock market metrics to save, invest, and get wealthy over time. You made large changes in the span of months, even when the metrics you are following didn't change that much. You may be just looking for external justification for your moves.
Point taken on the rest. I agree with you to an extent on this point, too. I'm not happy that the timing was so short. When I pulled out, I intended to wait until valuations dropped to 25-ish. It's not a magic number. It's not a marker for a market bottom. It's a guess at what may be only slightly irrational, as opposed to flat out bonkers! lol. I was prepared to sit out (in bonds/money-markets) for years. I believe staying away from the market at extremes may be appropriate. By definition, that would be timing. I believe staying invested at high, but not extreme, valuations is also appropriate because high valuations alone don't seem to be clear enough to act on.

Some range for the CAPE may just be noise. Any particular number from me is just a guess without a great deal more research. I'm granting your points... there are arithmetic holes in this conceptual argument. I accept that without relinquishing the assertion that it wise not to be invested at extreme valuations. I would posit that the CAPE at its extremes (say above 30 (?) and below its long-term mean nowadays (?)) is not noise, even in the '96 example as discussed. I think that's why it merited a Nobel Prize.

This thread is about timing at extreme valuations. By 'timing,' I mean timing over the long run. If you're not prepared to wait years (possibly a decade) for extreme valuations to return to moderate valuations in the 'noise' range, then you should not be using long-term indicators such as the CAPE.

Extreme long-term valuations to long-term timing is the exception that I'm willing to entertain, even though I haven't ever timed the market before. If you're adjusting allocations based on market valuations, as some apparent non-market-timers are discussing, you're doing the same thing, though I admit my 100% adjustment is too extreme.

Thank you again for your feedback. I really value the discussion even if I'm, like most, likely to stick with my original perceptions. The discussions nevertheless serve to 'plant the seed' for my own analysis, intuitions, and future adjustments in perception.

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 4:55 pm

marcopolo wrote:
Thu Feb 21, 2019 2:23 pm

CAPE10 itself did not really change all that much (32 --> 27 --> 30) in the time frame you are discussing. Certainly not enough to justify wholesale shift out, back in, and out again of equities.
Fair criticism. I don't think I should conduct wholesale shifts, either. I will resist doing so again. I think the adjustments should be commensurate with my level of confidence. That is to say, I was not 100% confident, and therefore should not adjust 100% of my assets out. I'm nearly certain that over a long-enough investment horizon, the markets will go up. So, I should be invested to some extent over that time frame.
These were fairly routine gyrations. You made them seem like much bigger shifts by changing the ruler by which you are measuring. In early December you got out based on an unmodified CAPE being at 32. Then you got back in based on some modified metric you devised being at 22. Now that you want to get back in, you are going back to unmodified CAPE being at 30.
My hangups here are that I generally agree with you. I came out at an extreme CAPE, and even the adjusted value in the 27+ range is very high, though perhaps not extreme. After CAPE/10 came down to merely high instead of extreme, I was more motivated to investigate just how 'off' it may be.
Realized how much it's inflated by my own calculations. And yes... Got lucky in the timing of it.
So, it makes it look like the metric changed a lot (32 - 22 - 30). Still not huge, but maybe more reasonable to justify big moves. There is the danger that the moves are being driven more by emotion, and the changing metrics are being used to justify the moves.

You believe you have found a way to time the market. Maybe you are correct, and it has worked well for you last time, but the track record of people being able to do that successfully over the long run has not been that great.

Good luck to you.
Thank you!

Not with conviction, but yes. Here's the thing: If the market reflects the period from 1929-ish to 1997-ish, 60-some-odd years, I may never be in the position to time again in this manner. Am I therefore a market timer? Yes, absolutely. Am I likely to do this many times in the future? No.

gmaynardkrebs has asserted similar arguments. If nothing else, this sort of approach SHOULD, but won't necessarily, avoid bigger drops and lesser gains. I consider this a conservative approach.

For the statistical analysts in the room, if market returns are symmetrically distributed and quasi-random and the long-term expectation is 0% nominal return, then 50% of the time, you'll miss a positive return over the time frame you're analyzing.

User avatar
HomerJ
Posts: 13338
Joined: Fri Jun 06, 2008 12:50 pm

Re: You Can Market Time At Valuation Extremes

Post by HomerJ » Thu Feb 21, 2019 5:26 pm

Chris42163 wrote:
Thu Feb 21, 2019 4:34 pm
This thread is about timing at extreme valuations. By 'timing,' I mean timing over the long run. If you're not prepared to wait years (possibly a decade) for extreme valuations to return to moderate valuations in the 'noise' range, then you should not be using long-term indicators such as the CAPE.
Just one note. The 10% historical long-term return of the stock market?

That INCLUDES the crashes. Read that again. In the past, you didn't have to try to avoid the crashes to make a lot of money and become wealthy.

Sure, if you can avoid the bad times, you can make more. But you can also make mistakes and make less. And most people who time do make less.

Market-timing is hard. Again, 1996, valuations were extreme. Near absolute highs over the past 100 years. And it turned out to be a great time to buy stocks. Stocks were CHEAP in 1996 (compared to the next 23 years).

Market timing is not as easy as you think.

And it's not necessary.

But good luck to you. Maybe you are one of the rare people who can time the market successfully.
The J stands for Jay

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 8:18 pm

Who knows? I certainly wouldn't make adjustments I thought were likely to hurt my returns.

Did you see my discussion about your 1996 example above? 2 points. 1) It looks as though one stood to lose next to nothing invested in bonds or money markets until the CAPE returned to acceptable levels. 2) The CAPE and TMC/GDP signals diverged greatly. The Buffett-indicator, signaled fair value.

By the way, the decision to pull out was gut-wrenching. Reinvesting was not. I don't think timing is easy. I'm not at all confident I should. In fact, I'm pretty insecure about being divested. However, I knew I'd regret absorbing the possible drop by staying in more than I'd regret the missed gains by staying out. Ultimately, that's what drove me to finally act on my analysis. It turned out very well. It may not next time. I accept that.

But I have a question for you. What valuations would make you uncomfortable? Can you honestly say that you wouldn't cash out no matter the situation?

User avatar
HomerJ
Posts: 13338
Joined: Fri Jun 06, 2008 12:50 pm

Re: You Can Market Time At Valuation Extremes

Post by HomerJ » Thu Feb 21, 2019 8:42 pm

Chris42163 wrote:
Thu Feb 21, 2019 8:18 pm
But I have a question for you. What valuations would make you uncomfortable? Can you honestly say that you wouldn't cash out no matter the situation?
My asset allocation is exactly the same regardless of valuations.

Because the risk is never zero. I pick an asset allocation based on the idea that the market could crash 50% tomorrow and take 5 years to recover. Because it might. I don't care what valuations are. If I'm already prepared for a crash, it doesn't matter if a crash is possibly more likely. I'm already ready for it.

I'm 5 years from retirement, and I'm 50/50 stocks/bonds. It doesn't matter what valuations are... If valuations were low, I wouldn't load up on stocks, because I'm going to need that money soon, and the risk of a crash is never zero.

Valuations being high don't matter to me either. They've been high for 27 years (since 1992), and returns have been great. I don't believe in valuations at all. The theory makes sense to me, but in practice, valuations had been terrible at predicting the market.

By the way, the S&P 500 is nearly back where it started back when you got out. I've lost nothing by doing nothing.

You gained, but if you had timed it wrong, you could have lost.

Riding the crash down and back up again is no big deal. It's how the stock market works..

Again, the 10% historical return INCLUDES the crashes. You still got 10% in the past riding the crash down and riding it back up. Jumping in and out is not necessary.

Money I invested in 2000, at the highest valuations ever, has still returned like 6.2% a year over the past 19 years. And of course, money I invested in 1997,1998,1999,2001,2002,2003 has all done better.

The absolute worst time to invest in recent history was 2000. And we've still gotten 6.2% a year on that money over the long-term. That's our recent WORST-CASE annual long-term return.

Jumping in and out is not necessary to get rich.
The J stands for Jay

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 9:34 pm

I appreciate your thoughts and responses. I don't disagree with your approach. In fact, it's exactly what I recommend to my friends and family. Pick an allocation, rebalance occasionally, and stick with it through the ups and downs. Perhaps, I'll learn that there's nothing special about even extreme valuations and end up back with your approach again.

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Thu Feb 21, 2019 10:06 pm

marcopolo wrote:
Thu Feb 21, 2019 2:23 pm
In early December you got out based on an unmodified CAPE being at 32.
Also, for clarity, I got out on October 17th. Market was on 2809. I was mistaken to say I was in on xmas eve. I was back in on 20DEC. The bottom of the market was xmas eve, about 5% lower than my reinvestment point.

User avatar
gmaynardkrebs
Posts: 1660
Joined: Sun Feb 10, 2008 11:48 am

Re: You Can Market Time At Valuation Extremes

Post by gmaynardkrebs » Thu Feb 21, 2019 10:43 pm

Chris42163 wrote:
Thu Feb 21, 2019 9:34 pm
.. Perhaps, I'll learn that there's nothing special about even extreme valuations .
I hope not.

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Sun Mar 17, 2019 8:37 pm

I've just moved to a 60/40 allocation. I have no short term prediction, but I feel like the downside potential in the next 5-10 years exceeds the upside. If the market fell 20% from current valuations (not prices), then I'd move back to an aggressive position. Conversely, if valuations jumped 20% from current, I'd be wholly out of stocks. In between that range, I'll adjust my allocation accordingly.

I'm going to time the market at extremes, but I'm going to try and do so with less extreme moves than last year. I'd been 100% stocks to that point. I think an 80-90% stock allocation is sensible given my age and ambitions, but at current valuations, I believe 60/40 or 70/30 is more appropriate for me.

NibbanaBanana
Posts: 247
Joined: Sun Jan 22, 2017 10:34 pm

Re: You Can Market Time At Valuation Extremes

Post by NibbanaBanana » Sun Mar 17, 2019 9:28 pm

Look at this chart. http://pages.stern.nyu.edu/~adamodar/Ne ... spearn.htm

Look at the increasing dividends. They almost always go up. A few drops but pretty quickly recover. (Earnings a little more volatile but still increase over time.) Don't you want in on that action?

User avatar
gmaynardkrebs
Posts: 1660
Joined: Sun Feb 10, 2008 11:48 am

Re: You Can Market Time At Valuation Extremes

Post by gmaynardkrebs » Sun Mar 17, 2019 9:45 pm

Chris42163 wrote:
Sun Mar 17, 2019 8:37 pm
I've just moved to a 60/40 allocation. I have no short term prediction, but I feel like the downside potential in the next 5-10 years exceeds the upside. If the market fell 20% from current valuations (not prices), then I'd move back to an aggressive position. Conversely, if valuations jumped 20% from current, I'd be wholly out of stocks. In between that range, I'll adjust my allocation accordingly.

I'm going to time the market at extremes, but I'm going to try and do so with less extreme moves than last year. I'd been 100% stocks to that point. I think an 80-90% stock allocation is sensible given my age and ambitions, but at current valuations, I believe 60/40 or 70/30 is more appropriate for me.
I think 60/40 is plenty of equity exposure right now and through most of your working life. If the markets go up, you share in it, and if there's a major downturn, you'll be buying more shares at a lower price. I was never above that level, and it doesn't matter to me that some people made more money by taking more risk.

Chris42163
Posts: 84
Joined: Fri Feb 15, 2019 8:03 pm

Re: You Can Market Time At Valuation Extremes

Post by Chris42163 » Sun Mar 17, 2019 10:23 pm

NibbanaBanana wrote:
Sun Mar 17, 2019 9:28 pm
Look at this chart. http://pages.stern.nyu.edu/~adamodar/Ne ... spearn.htm

Look at the increasing dividends. They almost always go up. A few drops but pretty quickly recover. (Earnings a little more volatile but still increase over time.) Don't you want in on that action?
At 60/40, I am in on that action. But no, reducing my exposure seems the prudent course of action considering my goals and current situation. I am ahead of schedule to achieve financial independence, and I can afford not to be greedy when the markets are overvalued.
gmaynardkrebs wrote:I think 60/40 is plenty of equity exposure right now and through most of your working life. If the markets go up, you share in it, and if there's a major downturn, you'll be buying more shares at a lower price. I was never above that level, and it doesn't matter to me that some people made more money by taking more risk.
I made plenty at 100% exposure over the last 12 years. The market has been wonderful to me. Now is the time to stop being greedy and take a little off of the table. I hope it results in buying opportunities in the future, but even if it doesn't, I'm in good shape to achieve financial goals.

Post Reply