Bubble spotting

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
User avatar
Topic Author
DartThrower
Posts: 953
Joined: Wed Mar 11, 2009 4:10 pm
Location: Philadelphia

Bubble spotting

Post by DartThrower »

Hi.

I am trying to wrap my mind around the Bogle philosophy of investing, and coming to this site has been a great help. Last year around this time however I remember our friend Jack going on TV and warning of approaching severe problems in the economy, and by extension I took that to mean the stock market. And back in 1999 when I was still an active investor it seemed like tech stock valuations were nothing short of insane. While I wasn't trying to outguess the market I did sense that massive risk was everywhere and acted accordingly.

Is it at all possible to identify something so extreme as the tech bubble or the housing bubble before they crash? Respected investors such as Warren Buffet and economists such as Robert Schiller seem to think so. Staying the course in normal times is a no-brainer, but if the NASDAQ reached 5000 again without any kind of regard to fundamental valuation I'm not sure I could stick with it.

So, are extreme bubbles actionable?
Our patience will achieve more than our force. -Edmund Burke
livesoft
Posts: 75128
Joined: Thu Mar 01, 2007 8:00 pm

Re: Bubble spotting

Post by livesoft »

DartThrower wrote:So, are extreme bubbles actionable?
Absolutely. One can actionably rebalance to get back to your asset allocation. An extreme bubble will put your AA more out of whack and as long as you look, you should be able to detect that you need to rebalance.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc »

Unfortunately smart people were warning of the tech bubble crash as early at 1996 or so. So right, but off in time by 4 years or so.

One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium. And pulling back a little only helps a little (ie you still get run over).

Very tough to look forward and do the right thing.

Always easy in hind sight though. :)
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 10:47 pm

And...

Post by Trev H »

.
One of the best ways to be assured protection from BUBBLY events is not to be TILTED to much (or LUMPED to MUCH) in any one direction.

For example 100% TSM for US Equities.

Very Dangerous.

50/50 TSM or LCB and SCB or SCV would be a much better (ANTI-BUBBLE) mix.

Below is a good example from the Large Growth BUBBLE comparing TSM to 50/50 TSM/SCV.


1995-1999 Large Trounces Small
2000-2005 Small Trounces Large

Code: Select all

==========================================
.........100%........50% TSM.......100%...
.........TSM.........50% SCV.......SCV....
==========================================
Init...10,000.00....10,000.00....10,000.00
1995...13,579.00....13,079.50....12,580.00
1996...16,425.16....15,849.74....15,272.12
1997...21,515.31....20,825.76....20,128.65
1998...26,519.78....22,570.96....18,820.29
1999...32,834.14....25,636.10....19,450.77
2000...29,363.57....27,085.82....23,706.60
2001...26,142.38....27,455.54....26,954.40
2002...20,662.94....22,628.86....23,126.88
2003...27,140.77....30,383.77....31,727.77
2004...30,538.80....35,863.48....39,199.65
2005...32,365.02....38,024.25....41,579.07
==========================================
StDev......19.49........15.44........16.08
TSM Bubbled and Pop'd.

50/50 Did not.

Code: Select all

===================================================
.............100%.....50% LCB....50% LCB....50% MCV
.............TSM......50% SCB....50% SCV....50% MCG
===================================================
1970-1974...19.91......18.34......17.73......21.00
1975-1979...16.82......15.82......16.32......21.38
1980-1984...14.62......16.48......12.86......14.91
1985-1989...12.04......12.74......12.32......11.74
1990-1994...15.54......18.49......18.62......19.44
1995-1999....6.20.......7.78......10.23......10.33
2000-2004...21.78......22.47......18.97......21.05
2005-2008...21.41......21.56......21.31......24.89
===================================================
Above is something else I noticed about that period. Showing StDev for 5 year periods.

When a Equity Asset Class like LCB (or TSM) which has a normal StDev in the 15-20 range, shows a 4-5 year string of volatility in the Treasury Bond Range (like TSM did 1995-1999) that is probably a very good WARNING sign.

Bubble about to POP.

===
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: And...

Post by avalpert »

Trev H wrote:.
One of the best ways to be assured protection from BUBBLY events is not to be TILTED to much (or LUMPED to MUCH) in any one direction.

For example 100% TSM for US Equities.

Very Dangerous.

50/50 TSM or LCB and SCB or SCV would be a much better (ANTI-BUBBLE) mix.

Below is a good example from the Large Growth BUBBLE comparing TSM to 50/50 TSM/SCV.


1995-1999 Large Trounces Small
2000-2005 Small Trounces Large

Code: Select all

==========================================
.........100%........50% TSM.......100%...
.........TSM.........50% SCV.......SCV....
==========================================
Init...10,000.00....10,000.00....10,000.00
1995...13,579.00....13,079.50....12,580.00
1996...16,425.16....15,849.74....15,272.12
1997...21,515.31....20,825.76....20,128.65
1998...26,519.78....22,570.96....18,820.29
1999...32,834.14....25,636.10....19,450.77
2000...29,363.57....27,085.82....23,706.60
2001...26,142.38....27,455.54....26,954.40
2002...20,662.94....22,628.86....23,126.88
2003...27,140.77....30,383.77....31,727.77
2004...30,538.80....35,863.48....39,199.65
2005...32,365.02....38,024.25....41,579.07
==========================================
StDev......19.49........15.44........16.08
TSM Bubbled and Pop'd.

50/50 Did not.

Code: Select all

===================================================
.............100%.....50% LCB....50% LCB....50% MCV
.............TSM......50% SCB....50% SCV....50% MCG
===================================================
1970-1974...19.91......18.34......17.73......21.00
1975-1979...16.82......15.82......16.32......21.38
1980-1984...14.62......16.48......12.86......14.91
1985-1989...12.04......12.74......12.32......11.74
1990-1994...15.54......18.49......18.62......19.44
1995-1999....6.20.......7.78......10.23......10.33
2000-2004...21.78......22.47......18.97......21.05
2005-2008...21.41......21.56......21.31......24.89
===================================================
Above is something else I noticed about that period. Showing StDev for 5 year periods.

When a Equity Asset Class like LCB (or TSM) which has a normal StDev in the 15-20 range, shows a 4-5 year string of volatility in the Treasury Bond Range (like TSM did 1995-1999) that is probably a very good WARNING sign.

Bubble about to POP.

===
In other words diversify across asset classes to limit the impact of a bubble in any one of them...
User avatar
bottlecap
Posts: 6605
Joined: Tue Mar 06, 2007 11:21 pm
Location: Tennessee

Post by bottlecap »

Rodc wrote:Unfortunately smart people were warning of the tech bubble crash as early at 1996 or so. So right, but off in time by 4 years or so.
This, along with livesoft's observation, is key. If you were really, truly smart and foresaw the "irrationality" of it all long before the "pop", you were a loser, failing to capture the illusory gains of a market out of whack. If you stuck with an AA and rebalanced at a predetermined "imbalance points," you made money off of the market without having to be prescient or even right.

In a way this is an obvious statement, but what's not so obvious is that an investor can be completely right about an analysis of the "value" of the market, and still lose out because no one else realizes the overvaluation for a significant period of time during which gains could be made.

If even if extreme bubbles are identifiable and, as such actionable, an investor can't time one to achieve maximum return with any consistency, if just because it is difficult to determine when an "irrational" market will become rational. Thus, having a predetermined AA eliminates the concern.

I feel better about my AA already....
jar2574
Posts: 256
Joined: Mon Mar 19, 2007 2:19 pm

Post by jar2574 »

Rodc wrote: One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium.
I don't think this is necessarily true. In fact, often when an investor's risk is high it is because he is accepting a lower risk premium than he should be accepting.
Last edited by jar2574 on Fri May 08, 2009 5:14 pm, edited 1 time in total.
User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Post by bob90245 »

"I've learned that market timing can ruin you."
-- Elaine Garzarelli

"If I have noticed anything over these 60 years on Wall Street, it is that people do not succeed in forecasting what's going to happen to the stock market."
-- Benjamin Graham

Plenty more experts quoted here:

http://socialize.morningstar.com/NewSoc ... ageIndex=5
grumel
Posts: 1629
Joined: Fri Mar 30, 2007 1:38 am
Location: Germany

Post by grumel »

Unfortunately smart people were warning of the tech bubble crash as early at 1996 or so. So right, but off in time by 4 years or so.
Were they really smart? Performance from 1996 till today for many hype stocks like Yahoo, Netscape or Dell is still looking good.
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 10:47 pm

Post by Trev H »

.
Avalpert said..

==
In other words diversify across asset classes to limit the impact of a bubble in any one of them...
==

Exactly...

Or at least go to opposite ends of the ranges.

TSM (In Larry's words) is something like 75% Large Growth (think that is based on DFA's size/valuation methods).

So no better place to go to offset that than to SV.

===
grok87
Posts: 9280
Joined: Tue Feb 27, 2007 9:00 pm

Post by grok87 »

Rodc wrote:Unfortunately smart people were warning of the tech bubble crash as early at 1996 or so. So right, but off in time by 4 years or so.

One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium. And pulling back a little only helps a little (ie you still get run over).

Very tough to look forward and do the right thing.

Always easy in hind sight though. :)
+1
RIP Mr. Bogle.
grok87
Posts: 9280
Joined: Tue Feb 27, 2007 9:00 pm

Post by grok87 »

Trev H wrote:.
Avalpert said..

==
In other words diversify across asset classes to limit the impact of a bubble in any one of them...
==

Exactly...

Or at least go to opposite ends of the ranges.

TSM (In Larry's words) is something like 75% Large Growth (think that is based on DFA's size/valuation methods).

So no better place to go to offset that than to SV.

===
I guess DFA's approach must be pretty different from Morningstar. Morningstar shows VTSMX (total stock market) as

26....22....24
6.......6......7
3.......3......3

Morningstar shows Vanguard SMall Cap Value as:

0..........0...............0
9..........15.............4
38.........28.............7

cheers,
RIP Mr. Bogle.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc »

jar2574 wrote:
Rodc wrote: One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium.
I don't think this is necessarily true. In fact, often when an investor's risk is high it is because he is accepting a lower risk premium than he should be accepting.
That sounds like hindsight bias.

Remember, don't confuse the results you get with the strategy.

That is, the statistical expectation is not always the same as what shows up.

If you flip a coin 10 times the expectation, that statistically most likely outcome is 5 heads and 5 tails, and that is the smart way to bet. And that is true even if you end up with 7 heads and 3 tails.

When risk is perceived to be high the expected return is also high. But if the risk materializes, you lose. If you only play when risk is perceived to be low you have to accept an expected small risk premium, though you may get lucky.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.
User avatar
cato
Posts: 230
Joined: Thu Dec 20, 2007 7:39 pm
Location: Portola Valley, CA

Re: Bubble spotting

Post by cato »

DartThrower wrote: Is it at all possible to identify something so extreme as the tech bubble or the housing bubble before they crash?
....
So, are extreme bubbles actionable?

The answer to the first question is emphatically Yes. There are a number of ratios that tend to revert back to their means over time. When these ratios get a few SDs away from their means, you can be pretty confident there's a bubble.

The answer to the 2nd question is "maybe." When a bubble actually pops is a matter of guesswork. One of the most obvious idiotic bubbles -- the Tulip Bubble -- lasted for almost 4 years. So, yes, it would have made sense to dig up the bulbs in your wife's garden in 1635. It would not have made sense to sell bulbs short. Of course, this is in the case of a really absurd bubble. It ain't usually that simple.


Now you'll get a lot of lectures about efficient markets here and all, but the bottom-line is if you see P/E, P/S, P/B and all the other ratios near their highest level in decades (or in history), it makes sense to take something off the table. I rely on some *proprietary* sentiment indicators too.

People can laugh about that, but should restrain themselves unless they sold an entire Tech portfolio in Feb. 2000, shorted the QQQ and put everything into MLPs like KMP and TPP. (I retired after that at 40.)


-cato


p.s. I sold my house in 2005 and am now renting.
Citigroup delenda est.
grok87
Posts: 9280
Joined: Tue Feb 27, 2007 9:00 pm

Re: Bubble spotting

Post by grok87 »

cato wrote:
DartThrower wrote: Is it at all possible to identify something so extreme as the tech bubble or the housing bubble before they crash?
....
So, are extreme bubbles actionable?

The answer to the first question is emphatically Yes. There are a number of ratios that tend to revert back to their means over time. When these ratios get a few SDs away from their means, you can be pretty confident there's a bubble.

The answer to the 2nd question is "maybe." When a bubble actually pops is a matter of guesswork. One of the most obvious idiotic bubbles -- the Tulip Bubble -- lasted for almost 4 years. So, yes, it would have made sense to dig up the bulbs in your wife's garden in 1635. It would not have made sense to sell bulbs short. Of course, this is in the case of a really absurd bubble. It ain't usually that simple.


Now you'll get a lot of lectures about efficient markets here and all, but the bottom-line is if you see P/E, P/S, P/B and all the other ratios near their highest level in decades (or in history), it makes sense to take something off the table. I rely on some *proprietary* sentiment indicators too.

People can laugh about that, but should restrain themselves unless they sold an entire Tech portfolio in Feb. 2000, shorted the QQQ and put everything into MLPs like KMP and TPP. (I retired after that at 40.)


-cato


p.s. I sold my house in 2005 and am now renting.
cato,
First let me say I love your tagline "Citigroup delenda est"!
And is sounds like you've made some smart moves- more power to you!
But the reality is, there are 99 other catos out there (cato1, cato2, cato3,...cato99) who thought their market timing skills were just as good and failed miserably. Those catos are not posting because, well let's face it they have nothing good to post about. So as boglehead viewers of this forum we get a very biased view.

The same thing is true of me of course. Catch me at a cocktail party and I might regale you with some stories from my pre-Boglehead days when I doubled my money in GM (bought at about 18, sold at 36). What I would never tell you of course is when I bought Corning at 24, doubled down at 12, and sold at 6!

cheers,
RIP Mr. Bogle.
User avatar
cato
Posts: 230
Joined: Thu Dec 20, 2007 7:39 pm
Location: Portola Valley, CA

Re: Bubble spotting

Post by cato »

grok87 wrote: First let me say I love your tagline "Citigroup delenda est"!
And is sounds like you've made some smart moves- more power to you!
But the reality is, there are 99 other catos out there (cato1, cato2, cato3,...cato99) who thought their market timing skills were just as good and failed miserably. Those catos are not posting because, well let's face it they have nothing good to post about. So as boglehead viewers of this forum we get a very biased view.

cheers,

Thanks Grok,

I do get your point and agree that regular market-timing is a fool's game. It's just that every now and then, things get so absurd, that life couldn't continue if market assumptions were to hold true.

For example, I know there are a quite a few Bogleheads that dropped their usual rhetoric and snapped up TIPs like crazy when the real yields for off-years were around 4%. Bunch of market-timers!! ;-)
Citigroup delenda est.
will23
Posts: 112
Joined: Mon Mar 16, 2009 10:32 pm

Post by will23 »

jar2574 wrote:
Rodc wrote: One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium.
I don't think this is necessarily true. In fact, often when an investor's risk is high it is because he is accepting a lower risk premium than he should be accepting.
Pretty much the definition of a bubble is high risk and irrationally low risk premium. The hard part isn't spotting a bubble but it's figuring out what will be impacted when it bursts.
grayfox
Posts: 5569
Joined: Sat Sep 15, 2007 4:30 am

Re: Bubble spotting

Post by grayfox »

DartThrower wrote:So, are extreme bubbles actionable?
I think there are two perspectives you can take on bubbles. One is from the individual's point of view. The other is from the macro point of view.

For the individual, I think it is possible to reduce or eliminate exposure to assets that are in bubbles. One way as pointed out is diversifying across more asset classes. Split 50/50 or how about 25% in each of four. (Harry B.?) Another way is to do some fundamental analysis on valuation. Learn how to estimate expected return. If something is way overvalued, buy something else.

From the macro point of view, if investors were better informed I think it is possible to not have bubbles form in the first place. If every investor knew how to estimate the return of assets like the equity risk premium for stocks, then the bubbles would not form as easily. Back in 2000 the ERP was negative but most people weren't aware of it. If you were given a choice between 4% real return guaranteed or 3% real with great uncertainty, which would you choose? That was the choice in 2000, but nobody understood it except for maybe Larry and others like him.

I don't remember seeing any headlines that said Equity Risk Premium is Negative. All I remember is BULL MARKET: Stocks Reach New High.
SP-diceman
Posts: 3968
Joined: Sun Oct 05, 2008 9:17 am

Post by SP-diceman »

So, are extreme bubbles actionable?
I would say yes.
(to a certain extent)

However wouldn't the real question be: Can you do it?

One of the issues of shifting assets or market timing is
folks believe if it can be done, everyone who try's can do it.

If we all went out and bought golf clubs, would we all have
the same score?

Would you be willing to start following major moving averages,
yield curves, PE ratios, market industry sectors, inflation?
Then be able to accept the fact that your analysis could be wrong?

That's ultimately what trying to avoid a bubble entails.


Thanks
SP-diceman
jar2574
Posts: 256
Joined: Mon Mar 19, 2007 2:19 pm

Post by jar2574 »

Rodc wrote:
jar2574 wrote:
Rodc wrote: One can certainly pull back if they sense risk is high. Unfortunately when risk is high, so is the risk premium.
I don't think this is necessarily true. In fact, often when an investor's risk is high it is because he is accepting a lower risk premium than he should be accepting.
That sounds like hindsight bias.

. . .


When risk is perceived to be high the expected return is also high. But if the risk materializes, you lose. If you only play when risk is perceived to be low you have to accept an expected small risk premium, though you may get lucky.

Well, I was talking about actual risk when I addressed the statement: "when risk is high, so is the risk premium."

But judging by your follow-up, it looks like you meant to address percieved risk rather than actual risk. Since an investor's risk premium is by definition tied to his perception of risk, your follow-up makes more sense.

But I don't think that the statement: "often when an investor's risk is high it is because he is accepting a lower risk premium than he should be accepting," is an example of hindsight bias. This seems like a true factual statement to me, and it doesn't necessarily imply anything about that particular investor's ability to have changed his behavior.

Now in some cases, I do think some individuals have the ability to spot investments that are over or under-priced. But I wasn't saying that actual risk is always predictable.
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 10:47 pm

Grok..

Post by Trev H »

.
Grok,

Yes, DFA's size/valuations must be quite different than M*.

But even with M* size/valuation details

VTSMX = Extremely tilted in one direction, bubble possibility = extremely high.

26-22-24
06-06-07
03-03-03

Much like 10% LCB, 90% SCV would be very tilted in one direction.

My own personal choice...

50% VLACX (LC Index)
50% VISVX (SV Index)


15-12-13
08-10-05
19-14-03

Does not place nearly as much faith in either extreme but on the small end does favor SV quite a bit over SG.

If you have no faith in academics then you could go 50/50 Large Blend/Small Blend.

The 50/50 mix of LB/SV is a very low correlation mix and bubble proof.

===
Larry
Posts: 81
Joined: Fri Apr 11, 2008 7:54 pm

Post by Larry »

WSJ about 20 years ago quoting a forgotten high profile player "When the whole worlds gone crazy you'd be crazy if you weren't a little crazy too" so yes the tech bubble was forecasted as early as 1996 if you bailed to CD's then you would have missed a 4 year run up and oh jeez what if you gave up on that and went all in just before the crash you would have missed four years of the run up and took the crash. Kind of makes a good AA and stay the course look attractive.
User avatar
tadamsmar
Posts: 9253
Joined: Mon May 07, 2007 12:33 pm

Re: Bubble spotting

Post by tadamsmar »

DartThrower wrote:Hi.

I am trying to wrap my mind around the Bogle philosophy of investing, and coming to this site has been a great help. Last year around this time however I remember our friend Jack going on TV and warning of approaching severe problems in the economy, and by extension I took that to mean the stock market. And back in 1999 when I was still an active investor it seemed like tech stock valuations were nothing short of insane. While I wasn't trying to outguess the market I did sense that massive risk was everywhere and acted accordingly.

Is it at all possible to identify something so extreme as the tech bubble or the housing bubble before they crash? Respected investors such as Warren Buffet and economists such as Robert Schiller seem to think so. Staying the course in normal times is a no-brainer, but if the NASDAQ reached 5000 again without any kind of regard to fundamental valuation I'm not sure I could stick with it.

So, are extreme bubbles actionable?
Ignore that man behind the curtain and stay the course!
User avatar
tadamsmar
Posts: 9253
Joined: Mon May 07, 2007 12:33 pm

Post by tadamsmar »

DartThrower, you are a quick study. I can't believe the number Diehards who can't figure out the Bogle is a walking contradiction spouter. You figured it out real fast.
Post Reply