Author Topic: Frequency of massive bubbles increasing - thoughts on if and why?  (Read 2245 times)

Dynamic

  • Sr. Member
  • ****
  • Posts: 360
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #10 on: April 26, 2018, 12:53:17 PM »
+1 to vinod1 from me too.

I concur about the promise of many members of the younger generation and the overeagerness to call bubbles. I think the latter can lead to a lot of people to miss out on time-in-the-market by trying too hard in timing the market often at a cost to their long-term returns.


Pelagic

  • Sr. Member
  • ****
  • Posts: 307
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #11 on: April 26, 2018, 02:33:08 PM »
Doesn't ease of participation play a role as well? If someone can participate in BTC when all their friends are already in by just downloading an app on their phone and transferring some money into their account and clicking buy, all done within 10 minutes, the participation as a percentage of the population is significantly higher than even in the late 90s.

Think of the various barriers to entry that have existed when it comes to participating in various financial markets, I can imagine in the 1920s a lot of people wanted to get in to the stock market but filling out the requisite paperwork and transferring the money to your broker whittled down the number of people who actually followed through significantly. The 90s internet bubble saw the birth of online trading accounts, making participation easier than it had ever been.

And now participation has almost no barriers to entry when done through a phone app, and perhaps more importantly can be done immediately when social pressure is at its highest - not, oh I'll go home and talk to the wife about it and setup an account. So yeah, social "FOMO" plays a role but the fact that there are almost no barriers to entry into participating in the latest bubble, coupled with FOMO, means everyone can easily participate. Add in things like social sharing and gamification and participation becomes not only easy but a competition amongst friends.


JimBowerman

  • Newbie
  • *
  • Posts: 16
    • Split Rock Capital Management, LLC
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #12 on: May 01, 2018, 06:56:21 AM »
By this definition, I would put only the dot com stocks in the late 1990's and the housing market in 2005 as a bubble. Others cases do not meet the criteria.

Vinod

IMO there's a lot of monday morning quarterbacking going on with the 2005 housing bubble.  Most simply sight the case shiller chart, but by that logic, why did Australian Housing prices keep going up?  Its really tough to identify a bubble in real time.

https://www.ampcapital.com/AMPCapitalGlobal/media/contents/Blog/olivers-insights/2014/April/03.04/chart-3.png

stahleyp

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 2557
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #13 on: May 01, 2018, 07:09:01 AM »
I won't use the b word, but I do think the sharp swings over the past couple of decades owes a lot to Fed intervention and ballooning debt.

The Fed bailed out the economy during the savings and loans "crisis" and then again during LTCM then again after the dot com boom (which led to the housing/credit crisis). If you let one of these events play out, I doubt the others would have been that bad.

Now, we're basically in a position that the middle class is squeezed and the economy stays afloat due to government spending and low interest rates.

Not to turn this into the political thread, but the Fed should let capitalism work (even when people get hurt).
Paul

Liberty

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 10247
  • twitter.com/libertyRPF
    • twitter.com/libertyRPF
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #14 on: May 01, 2018, 07:17:31 AM »
Not all excess or overvaluation is a bubble. The market doesn't spend much time right on the "historical average" line, it's usually swinging back and forth between over and under. So while I think we've had some bubbles in the recent past, I think some of what many people are qualifying as bubble doesn't fit that description, at least for my understanding.
« Last Edit: May 01, 2018, 09:15:43 AM by Liberty »
"Most haystacks don't even have a needle." |  I'm on Twitter  | Interesting podcast on sleep science

netnet

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 928
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #15 on: May 01, 2018, 02:41:36 PM »
With bubbles there are always methodological questions, as in exactly what defines a bubble, and how do you define it concurrently!  One leading historian makes the case that even the tulip bulb 'craze' was not a bubble! 

https://theconversation.com/tulip-mania-the-classic-story-of-a-dutch-financial-bubble-is-mostly-wrong-91413

I spilled my coffee reading that one.

Cigarbutt

  • Hero Member
  • *****
  • Posts: 1133
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #16 on: May 01, 2018, 07:05:12 PM »
If anything, I see a bubble in people calling out "bubbles". Otherwise, it is same old, same old. Nothing much has changed.

We first need to have a working definition of a bubble. Rob Arnott's definition would be a good starting point,

We define a bubble as a circumstance in which asset prices 1) offer little chance of any positive risk premium relative to bonds or cash, using any reasonable projection of expected cash flows, and 2) are sustained because investors believe they can sell the asset to someone else for a higher price tomorrow, with little regard for the underlying fundamentals.


By this definition, I would put only the dot com stocks in the late 1990's and the housing market in 2005 as a bubble. Others cases do not meet the criteria.

I do not think we are having more bubbles nowadays than we had in the past.

It is just that people scare themselves silly, looking for bubbles everywhere.

High prices? Yes. Bubbles? No.

Vinod

Vinod,
This post is not to pick on you. Just trying to understand.
BTW, spending most of my time these days trying to find individual issues selling below intrinsic value. (aren't we all?)
Still, interesting to occasionally have a "feel" for the environment.
Don't know if bubbles are becoming more or less frequent.
The challenge is that bubbles, almost by definition, are recognized only in hindsight.
It's possible that many "bubbles" were never recognized because imbalances did not reach a certain threshold or because a certain confluence of factors was simply absent, allowing the imbalances to correct over time.
So, let's stay away from the bubble terminology and try to assess, à la Grantham, how far the market is swinging from the norm.

On this Board, on December 1st, 2009 (yes there was a discussion about a bubble forming then...), you mentioned:

"Agree on the sanity check. I think we disagree due to differences in what is meant by a bubble. To me that means a near unambigous and demonstratable overvaluation. Take S&P 500 in 2000, at its peak of 1500 its dividend yield of 0.9% and normalized PE's in the 40's the expected real return on stocks in the long term is less than the return available on TIPS which at that time were having a real yield of 4.5%. This to me is a clear and demonstratable case of a bubble.

Take the present situation, using any of the most commonly used methods of estimating fair value for S&P 500, leads to a fair value range in the 850-1000. With nominal treasury yields in the 3-4% range and TIPS real yields around 2%, the expected returns on stocks over the very long term is much higher than bonds. It looks a little expensive but no where near the bubble range."

The S&P was around 1300 then but I agree with your assessment of "fair value" for the S&P at that point.
If you try to think like Mr. Buffett (at least from what he wrote in 1999, see below), what happened to the S&P, since your quoted post, is very interesting.
http://archive.fortune.com/magazines/fortune/fortune_archive/1999/11/22/269071/index.htm

If you follow Mr. Buffett's logic about how growth of the S&P should be connected to the GDP growth over the long term, starting with your assessment of "fair value" then, the S&P should trade between 1100 and 1300 today.  ???

Helpful variables:
-S&P profit margin over GDP:     2010: +/- 10%     now: +/- 10%
-10 year government yield:        end 2010: 3,3%    now: +/- 3,0%
-the S&P index closed at +/- 2650 today

What "explains" (math point of view) the difference between projected and realized is that, for the S&P 500, PE went from +/- 16 to +/- 25.

Of course the best may yet to come but one should perhaps discounts the possibility that PE stays at the same level (or even decreases?).

I agree that one should look from the bottom up, but when I look at an opportunity that sells for twice fair value, I tend to wonder (and, bubble or no bubble, unambiguously go for the next idea).

So, this discussion does not change what I do day to day but I wonder if general expectations are in line with historical parameters.
Or is it different this time?


« Last Edit: May 01, 2018, 07:08:37 PM by Cigarbutt »

vinod1

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 1252
    • My Blog
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #17 on: May 02, 2018, 05:02:03 AM »
Cigarbutt,

Snce 2009, I changed my mind on two points relevant to S&P 500 valuation:

1. Mean reversion of profit margins. I must have spent several hundreds of hours thinking and researching this issue over the last 17 years. I would not go into detail, but I no longer expect mean reversion of profit margins on a sustained basis to the levels of the period before 1980s. Not saying profit margins dont fluctuate or go down. Just the mean has shifted to a much higher level than the past. It might shift lower again sometime in future but not anytime soon.

2. Expected returns on stock market. I previously expected real returns of 6 to 7% as fair value for the stock market. With inflation of 2-3%, nominal returns of around 9%. In line with what stocks had historically provided. I now think fair value for stock market would be lower returns that this for a several reasons:

a) Even though 9% were the realized returns historically, it is very expensive in the past to get those returns. And few could actually get this return from a diversified portfolio. Think how expensive it would have been in 1900 or 1920 for a person to build a diversified portfolio of stocks to get this return. Think of the difficulty of getting relevant information about companies financials. Think of broker costs, transaction costs, etc. Fraud costs involved with physical shares. Risk of losing those physical shares, etc. A lot of the 9% expected or realized return on stock market would have been consumed by this. So if a person wants to hold 50 stocks, I would think that these costs could easily eat up 2% or more of the return. Now you can get a very diversified portfolio for less than 0.1% cost. So the expected return that the investor could actually realize has gone up by nearly 2% just from these lower costs.

b) Economic risks have gone down as well. Look at the frequency and magnitude of the recessions in the past compared to now (ya even with the 2008 Great Recession). This should naturally reduce the stock risk premium as well.

c) Inflation risk has gone down as well.

When you take all these factors (a, b & c) into account and consider that cash/bond returns are also going to be low (low real rate + low inflation + low inflation risk premium), then a higher multiple can be justified for S&P 500.

So not only are normalized earnings much higher in 2017 than I expected in 2009 (due to point #1) but the multiple that can be justified is also much higher than I expected in 2009 (due to point #2).

So I was actually wrong in my assessment of S&P 500 value in 2009. I made the mistake of reducing my allocation due to this a couple of years later. But was saved from this mistake due to opportunity in financials in 2011/12 period.

I actually reduced allocation in 2006/7/8 due to worries about profit margins and mean reversion. When markets fell in 2008/9 I felt vindicated. I was wrong of course.  But got bailed out by the financial crisis in 2008/9. The reason I changed my mind about profit margins/mean reversion is that when reducing allocation in 2006/7/8, I made a note to myself that if profit margins again go up in 5 years, then I should take a hard look at this issue again.

Vinod
The fundamental algorithm of life: repeat what works. –Charlie Munger

Cigarbutt

  • Hero Member
  • *****
  • Posts: 1133
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #18 on: May 02, 2018, 06:08:31 AM »
Thanks for the reply.
I don't use the same prisms and your perspective is helpful in terms of mind preparation.

Just to add a quote from a 2015 report I read from Morningstar reviewing the evolution of "costs" and "fees" from the previous 10 years:

"...industry fee revenue is at an all-time high, reaching $88 billion, up from $50 billion 10 years ago. During that period, industry assets under management have increased 143% while the asset weighted expense ratio has declined 27%, and industry fee revenue has grown by approximately 78%. Thus, a much larger share of the benefits of the increase in assets under management has stayed with the industry rather than being returned to fund shareholders."

To conclude, another quote (modified) from the investment industry:

"Past performance is not indicative of future results", unless the conditions that led to the past performance remain unchanged.



vinod1

  • Lifetime Member
  • Hero Member
  • *****
  • Posts: 1252
    • My Blog
Re: Frequency of massive bubbles increasing - thoughts on if and why?
« Reply #19 on: May 02, 2018, 06:34:55 AM »
Cigarbutt,

If I came across as bullish, I am not. I am more agnostic about market valuations. Do not have very strong views either way.

We have less than 150 years of data for the stock market. If you define a long term hold of 30 years as one representative sample. We have about 5 samples in stock market of long term returns (non-overlapping). Most of us would not put too much faith in a sample size of 5. That is where we are with stock market data.

Imagine 5 coin tosses, results in 3 heads. "Historically" we infer that heads come up 60%, then use that damn thing to make predictions. The stock market equivalent are the historical 6.5% annual real returns and the 6% or so net profit margins.

In another couple of thousand years, we might have a better idea. But until then, it is best not to have strongly held views on macro stock market valuations.

Vinod
The fundamental algorithm of life: repeat what works. –Charlie Munger