SCV premium is a history/math story, its not scientific

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

SCV premium is a history/math story, its not scientific

Post by LH » Mon Apr 14, 2014 12:13 am

SCV tilt to get the SCV premium, will simply work, until it does not. There really is no evidence what it will do going forward.

The market, is a complex chaotic system that humans do not understand, and cannot predict, its a sum total of human actions, human behavioralism, herd behavoir, etc.

Its a rule, just like the rule stocks yield more than bonds was a rule, with real nice history behind it. 50 plus years, until it ceased to work in 1959. Quite the shock at the time, stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks. sounds good, but completely wrong for past 50 years.

SCV, its the latest greatest thing......

Its historical based.

Is it science?????

Well, ask yourself the simple question, since its history based, what exactly the sample size is, and is it significant?

no.

Its not significant.

Its just some math, and has held true over a relatively small timeperiod, because really, all we have is a relatively small timeperiod to work with.

If a financial life is say 40 years. what do we have 100 years of good data, so that would be N=2.5

Maybe put it out to 200 years of data, if you want to think the data is any good starting from the early 1800s....... It get dicey, real real quick.

So its a rule, but one with not much behind it a priori. N=2.5 to N=5 at best.

Then, it gets better, because the market, when patterns become known, markets tend to adapt. so you would have to seriously consider all the prior history before SCV premium was known, to be a significantly different than the current history in progress where the market knows about SCV. Then the N goes down significantly, to basically N<1. The state in which the market knows about the SCV premium does not even have the pathetically small N of a whopping 2.5-5.0

SCV is a risk story, and humans like stories, reinforced with reading the tea leaves, the entrails, non statistically significant historical math, whatever. Now sure, the math is better in a sense than the entrails, but people, its called not significant for a reason.

Now math in the form of physics, sure, it can predict things long term, flight paths of spaceships launched tens of years ago. They know where they will be, with some degree of certainty.

Econ math, we cant predict the market 1 day ahead.......... Much less whether some behavior pattern or risk/reward pattern will persist long term, especially once the market knows about it.

Then, if the story really is about "risk" where risk is NOT just volatility, but real risk, where you lose your money permanently, and that is the type of risk I posit one would be paid for basically.

Well, the risk may show up for SCV in our lifetimes, permanently.

You only expectantly, get paid for taking risk, you run the risk that you take the hit (unless you are saying risk is only volatility, which is a weak argument I do not think many would make).

Either way you look at it, behavioral, or the risk story, or both. Its trivial to come up with the explanation why SCV fails going forward. Humans know about the premium, EF hutton says to buy SCV, the premium disappears. Not too hard. Now, where is the math, Hari Seldon style, to actually look at that. There is none. Now, how much data do we have history wise, to examine what the SCV premium will do when humans know of it, effectively nil.

There really is not much but alternative stories, which both sound good.

Econ is great, its fun stuff, I like it, likely enough to have gone and done it for a career, but you know "risk free" no such thing. "guaranteed" no such thing. and oh..... that statistical significance thing, where you see if your sample size is big enough to really have any significance, oh heck with that too? That does not make sense. you cannot claim mathematical history based degree of certainty without any meaningful statistical signficance of your sample.

you certainly can tell a story though, that includes some math and some recent history.

Its like having a jar of ten thousand marbles randomly assorted, and pulling out 3, and acting you know what the jar population looks like. You really do not. But hey, tell a nice story about it, sure, sounds great, and humans will assume they know whats in the jar based on the three marbles routinely.

I don't think there is any expectant harm with SCV tilt to some extent, but if you put all your eggs in the SCV basket, say I dunno 30 TIP 70 SCV, or whatever, you are more likely to lose out I posit, versus someone who is traditionally diversified.

User avatar
JoMoney
Posts: 5249
Joined: Tue Jul 23, 2013 5:31 am

Re: SCV premium is a history/math story, its not scientific

Post by JoMoney » Mon Apr 14, 2014 6:45 am

LH wrote:...you certainly can tell a story though, that includes some math and some recent history...
Benjamin Graham wrote:...Mathematics is ordinarily considered as producing precise and dependable results; but in the stock market the more elaborate and abstruse the mathematics, the more uncertain and speculative are the conclusions we draw there from ... Whenever calculus is brought in, or higher algebra, you could take it as a warning signal that the operator was trying to substitute theory for experience, and usually also to give to speculation the deceptive guise of investment.
http://johncbogle.com/wordpress/wp-cont ... -21-13.pdf
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
baw703916
Posts: 6676
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: SCV premium is a history/math story, its not scientific

Post by baw703916 » Mon Apr 14, 2014 7:10 am

Nothing in economics or finance is scientific, because you can't perform an adequately controlled experiment.

There's no proof that stocks will outperform bonds (stocks can go to zero by governmental action), nor that bonds are safer than stocks (hyperinflation anyone?), nor that gold is a good hedge (how did it do for U.S. investors from the 1930s through the 1970s?) In short, we can't scientifically prove anything at all. No, not even that the market portfolio is on the efficient frontier--that only holds under the unrealistic assumption that domestic stocks are the only asset class. Add international stocks and bonds, and you can't prove it.

Ok, so now that we have put that exercise in nihilism behind us, we still have to invest our money somehow.
Most of my posts assume no behavioral errors.

User avatar
matjen
Posts: 1883
Joined: Sun Nov 20, 2011 11:30 pm

Re: SCV premium is a history/math story, its not scientific

Post by matjen » Mon Apr 14, 2014 7:12 am

LH wrote:I don't think there is any expectant harm with SCV tilt to some extent, but if you put all your eggs in the SCV basket, say I dunno 30 TIP 70 SCV, or whatever, you are more likely to lose out I posit, versus someone who is traditionally diversified.
Phew! I was getting worried there for a minute. Good thing virtually everyone on this board and everyone I am aware of that suggests it has a tilt along the lines of 4 to 1 or perhaps 3 to 1 TSM to SCV. Even the Larry Portfolio is the complete opposite of the ratio quoted above. Roughly 30 SCV and 70 FI.
A man is rich in proportion to the number of things he can afford to let alone.

User avatar
nisiprius
Advisory Board
Posts: 36025
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: SCV premium is a history/math story, its not scientific

Post by nisiprius » Mon Apr 14, 2014 7:37 am

LH, I think you're right, but I also think that most of it applies to most investing strategies. Thus, I don't quite go along with Bogle's statement, but almost:
There are an infinite number of strategies worse than this one: Commit, over a period of a few years, half of your assets to a stock index fund and half to a bond index fund. Ignore interim fluctuations in their net asset values. Hold your positions for as long as you live, subject only to infrequent and marginal adjustments as your circumstances change. When there are multiple solutions to a problem, choose the simplest one.
I take mild issue with it in two regards.

First, "There are an "infinite number of strategies worse than this one" is vacuously true. I would reword it slightly: there are an infinite number of strategies that cannot be proved to be better or worse than this one. Even in hindsight, the other strategies will either have done better or worse, but there will be no way to tell whether the strategy itself was better or worse or whether the outcome was the result of chance. Larry Swedroe says we should not confuse strategy and outcome, but IMHO it is all but impossible to evaluate strategies in any fundamental way. So I would reword his statement to say "you can't tell whether the other strategies are better or worse, so you might as well choose the simplest, especially if the others will cost more."

Second, I think there are really about three different, distinguishable strategies: 25% stocks, 50% stocks, 75% stocks. It does take three sizes, but three sizes fit all. The difference between 25% stocks and 50% stocks is big enough that you might be able to say that for a certain person, one is "wrong" and the other is "right."

But, assuming some reasonably big, diversified, representative sampling of stocks, whether it is one index or another, whether it includes international or not, whether it's tilted or cap-weighted or conservatively actively managed... you are never going to know which is better.

You have to know very broadly what your risk tolerance is, and get your stock allocation roughly right. You don't want to be 0% stock at age 30, you don't want to be 100% stock at age 70. All the rest is taste, opinion, and unprovable.

You always have a "history" story but financial data is so badly behaved that there's never enough history. And all formal statistics assume that you have an accurate view of how the system works, and that's not true in investing. So you cannot say "because the historic statistics of the stock market were thus-and-such we can be reasonably sure that the true underlying mean is thus-and-such."

I'm not sure what you mean by a "math" story.

As for "science," the scientific part is all "if we had some ham, we could have ham and eggs, if we had some eggs." That is to say, we get very sophisticated analysis of would be the statistically best portfolio going forward if we had accurate foreknowledge of the statistical behavior of all the asset classes going forward. It is a pseudo-sophisticated approach. "We can't predict the actual price movements or dividend income in time going forward, but we do know the statistical averages and standard deviations and skew and kurtosis and correlations and so forth." But we don't. Correlations are no more predictable, no more a property of the asset class, than plain old "return" is.

As for exotic subsets within the realm of securities, you can't prove anything and it doesn't matter. As for "non-traditional investing" outside the realm of securities, even less so. I think it's all the search for something new to sell--and, the search for things that are new enough that the people promoting them have asymmetrical information about them.
Last edited by nisiprius on Mon Apr 14, 2014 7:42 am, edited 2 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: SCV premium is a history/math story, its not scientific

Post by dbr » Mon Apr 14, 2014 7:41 am

The comments above are outstanding and should be enshrined somewhere.

larryswedroe
Posts: 15635
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: SCV premium is a history/math story, its not scientific

Post by larryswedroe » Mon Apr 14, 2014 7:48 am

well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size
B) We have the same premiums in developed markets all around the globe, including EM. And they are pretty similar in size as well

While investing isn't a hard science like physics we can apply scientific methods to test theories and use statistical analysis to tell us the significance of the data.

The same type data and statistics that supports the ERP supports the size and value premiums, and MOM premium as well and while no one questions the ERP many do the others, despite all the out of sample tests and statistical significance

Bottom line you can ignore the data or not. In either case your taking risks that the premiums will not show up. Take your pick, but ignoring the premiums means you must take more equity risk to achieve the same expected return, and that means more tail risk. The interesting thing is that risk averse investors, and almost all investors are risk aversion, should prefer portfolios with low beta and high tilt to a market like portfolio, assuming they can take the tracking error.

Best wishes
Larry

dbr
Posts: 27207
Joined: Sun Mar 04, 2007 9:50 am

Re: SCV premium is a history/math story, its not scientific

Post by dbr » Mon Apr 14, 2014 8:01 am

Strangely enough, I agree with Larry above as well as with what Nisi wrote. But the point is that as believable as the analysis is, in my judgement it does not dictate that an investor "should" seek the advantages of small and value premiums. It does dictate that an investor can feel able to seek those advantages with understanding of what to expect. I think that ranking those additional factors in the same order of importance to investors as the equity risk premium is a red herring.

I think that Fama himself has said pretty much the same thing.

thx1138
Posts: 747
Joined: Fri Jul 12, 2013 2:14 pm

Re: SCV premium is a history/math story, its not scientific

Post by thx1138 » Mon Apr 14, 2014 8:11 am


larryswedroe
Posts: 15635
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: SCV premium is a history/math story, its not scientific

Post by larryswedroe » Mon Apr 14, 2014 8:41 am

dbr
I think that is exactly the point I made in my post, you should have a logical reason to take any risk, be it ERP or size or value or EM or term or default, or whatever
Larry

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 8:53 am

stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks. sounds good, but completely wrong for past 50 years.
You lost me here.

No one knowledgeable believes the second clause of the first sentence. Indeed as stated you have an oxymoron. A statement that is clearly false.

And what evidence do you have for the second sentence? As far as I can tell it is false as well.

You might be right about SCV, time will tell. But you seem to start with two falsehoods so I stopped reading.

ADDED: Now that I have read it, I think the replies are on target. All the OP has done is supply his own story and spin and offered up an unsubstantiated opinion. Could be right or wrong, but pretty empty of substance.
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.

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

Re: SCV premium is a history/math story, its not scientific

Post by avalpert » Mon Apr 14, 2014 9:23 am

Rodc wrote:
stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks. sounds good, but completely wrong for past 50 years.
You lost me here.

No one knowledgeable believes the second clause of the first sentence. Indeed as stated you have an oxymoron. A statement that is clearly false.

And what evidence do you have for the second sentence? As far as I can tell it is false as well.

You might be right about SCV, time will tell. But you seem to start with two falsehoods so I stopped reading.

ADDED: Now that I have read it, I think the replies are on target. All the OP has done is supply his own story and spin and offered up an unsubstantiated opinion. Could be right or wrong, but pretty empty of substance.
He's used this in a few threads now - seems to be on some kick where you ignore total returns and focus on the change in stock yield over time as if it is the only compensation an investor would take for the risk. It's a fairly odd, fallacious argument.

pkcrafter
Posts: 12815
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: SCV premium is a history/math story, its not scientific

Post by pkcrafter » Mon Apr 14, 2014 9:39 am

Rodc, are you saying this is not true?
stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
grayfox
Posts: 4885
Joined: Sat Sep 15, 2007 4:30 am

Re: SCV premium is a history/math story, its not scientific

Post by grayfox » Mon Apr 14, 2014 9:40 am

I'm willing to believe that there is small company risk.
Large companies have the ways and means to crush their smaller competitors.
E.g. Standard Oil drove many small oil companies out of business.
The "Big Three" auto makers put all Hudsons and Packards out of business.

Most industries seem to consolidate into only a few big players. The smaller companies either get taken over or driven out.

The risk of being small is also true nature:
big fish eat little fish; the runt of the litter often does not survive; the big gorilla gets all the bananas.
In war, if you have 500,000 soldiers and I have 1,000,000, all things being equal, I will usually destroy you.
Sports: football and basketball players bulk up to get bigger, bigger, BIGGER
World: big countries pushes around smaller countries.
At the beach, the big jock kick sand on the 90-pound weakling.

Small anything is usually at a disadvantage. (except microeleectronics!)

So if it's risky to be small, why would you bet on small?

The only reason I can think is if the potential reward is greater.
Would you bet on risky smaller stocks if the expected return is the same as the larger stocks?
No, it has to offer the potential for higher returns, a small stock risk premium.

Here is data I got from Vanguard:

Code: Select all

SYMBOL		VTSMX	VLCAX	VSMAX
STYLE		US-TSM	US-LC	US-SC
NUM STOCKS		3629	657	1435
MEDIAN MKT CAP		42.9	61.6	3.2
 ($billions)

P/E Ratio		19.8	18.8	28.3
P/B Ratio		2.6	2.7	2.5
ROE		17.1%	18.2%	10.0%
EARN GROWTH		12.0%	12.1%	11.1%

ROE/PtB		6.6%	6.7%	4.0%
I calculated the ratio ROE/PtB as one estimate the expected return.
Large stocks ROE/PtB=6.7%. Small stocks=4.0%
Doesn't look like there is a small-stock premium at this time.
The price or valuation (P/E, P/B) of small-cap stocks is too high. If they are greater risk than large stocks, they should be cheaper.
Last edited by grayfox on Mon Apr 14, 2014 9:54 am, edited 1 time in total.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 9:54 am

pkcrafter wrote:Rodc, are you saying this is not true?
stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks.

Paul
Yes.

Due to the fact that risk may well show up. Stocks are more risky so they are (generally) priced so that there is a reasonable statistical expectation of a pay off, but there is no guarantee. So the "must" is much too strong.
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.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 9:59 am

If they are greater risk than large stocks, they should be cheaper.
I do not know the answer to this question: if the e in p/e is expected to grow faster for small companies vs for large (more stagnant?) companies then the use of past e might not be the same as the e the market is using to price small companies, but the past e might be the e used by the market for large companies.

So, I'm sure it is out there, but does any know off hand the average historical p/e for small vs large stocks? Does it show that p/e for small is generally rather smaller than p/e for large?
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.

pkcrafter
Posts: 12815
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: SCV premium is a history/math story, its not scientific

Post by pkcrafter » Mon Apr 14, 2014 10:04 am

Rodc wrote:
So the "must" is much too strong.
Ah, yes, I "must" agree.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 10:10 am

pkcrafter wrote:Rodc wrote:
So the "must" is much too strong.
Ah, yes, I "must" agree.

Paul
Hi Paul,

Thought you might. :)
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
grayfox
Posts: 4885
Joined: Sat Sep 15, 2007 4:30 am

Re: SCV premium is a history/math story, its not scientific

Post by grayfox » Mon Apr 14, 2014 10:11 am

This chart from nisiprius shows that the small-stock premium comes and goes.

Image
http://www.bogleheads.org/forum/viewtop ... 9#p1939229

We are probably in a period when small-cap premium has gone away.
I would not tilt to small-cap until the expected return goes well above large-cap.

User avatar
JoMoney
Posts: 5249
Joined: Tue Jul 23, 2013 5:31 am

Re: SCV premium is a history/math story, its not scientific

Post by JoMoney » Mon Apr 14, 2014 4:14 pm

grayfox wrote:...We are probably in a period when small-cap premium has gone away. I would not tilt to small-cap until the expected return goes well above large-cap...
I agree with you. I have no question about whether or not small-caps are riskier than large-caps as a group. If one looks at individual companies the risks might be assessed differently, but I don't think many of us are looking to evaluate the fundamentals and future outlook of individual companies. In general, as a percentage, more of the larger companies today will be around a decade from now than the small companies. Of the small companies that do survive, their potential for faster growth is higher than large companies. But as a group do the small super-stars compensate enough for the larger percentage of failures? I would say sometimes they do, sometimes they don't. Same goes for junk bonds vs investment grade bonds. Some will have been shown to be a bargain, and many more will have justified the name as "junk". From my view, looking at long periods of time seems to reveal a lot of RTM and a weak story that their is a persistent premium.
But from time to time there does seem to be an exploitable premium in one subset of the market or another, but an exploitable premium should not exist under the premises of a market efficiently pricing "risk premiums" unless that premium is the result of accepting more risk. So when a period comes around where small-caps are outperforming we're feed a "risk story" and given methods to calculate a risk premium that seems to disregard old fashion fundamentals like earnings and dividends. And as long as it works, we have a story that fits into our "efficient market" theory and explains one of the many market anomalies in a way that sounds logical... But if it later shows to have failed, it wouldn't be the first time the academics have declared the old model "dead" and gone back to the drawing board to develop a new theory that describes the market but in some quantifiable way.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
nisiprius
Advisory Board
Posts: 36025
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: SCV premium is a history/math story, its not scientific

Post by nisiprius » Mon Apr 14, 2014 4:21 pm

Oh, and of course we always need to remember that...

"Investing is not a science."--John C. Bogle, Clash of the Cultures, p. 1.

Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: SCV premium is a history/math story, its not scientific

Post by steve_14 » Mon Apr 14, 2014 4:34 pm

grayfox wrote:I'm willing to believe that there is small company risk.
Large companies have the ways and means to crush their smaller competitors.
The real question is: how much more return does a typical $1B small cap have to offer investors than a $100B large cap? We can infer the answer for equity by looking at interest rates small vs large cap companies offer on their debt. How much different are they? I haven't seen any studies on this.

User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: SCV premium is a history/math story, its not scientific

Post by Clearly_Irrational » Mon Apr 14, 2014 4:37 pm

nisiprius wrote:Oh, and of course we always need to remember that...

"Investing is not a science."--John C. Bogle, Clash of the Cultures, p. 1.
The "hardness" of science is on a spectrum. http://xkcd.com/435/ Despite that, I think it can be useful in investing as long as you understand the limitations.
nisiprius wrote:First, "There are an "infinite number of strategies worse than this one" is vacuously true. I would reword it slightly: there are an infinite number of strategies that cannot be proved to be better or worse than this one. Even in hindsight, the other strategies will either have done better or worse, but there will be no way to tell whether the strategy itself was better or worse or whether the outcome was the result of chance. Larry Swedroe says we should not confuse strategy and outcome, but IMHO it is all but impossible to evaluate strategies in any fundamental way. So I would reword his statement to say "you can't tell whether the other strategies are better or worse, so you might as well choose the simplest, especially if the others will cost more."
I don't agree. One of the few things we can do is show which strategies are bad, however absence of proof of badness is not necessarily evidence of goodness.

steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: SCV premium is a history/math story, its not scientific

Post by steve_14 » Mon Apr 14, 2014 4:40 pm

larryswedroe wrote:well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size
Not in real fund data. DFA started their small cap fund in 1982, only to experience violent underperformance during the 80s and 90s vs large caps. And we have Vanguard growht/value pairs going back 30 years showing no value premium.

larryswedroe
Posts: 15635
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: SCV premium is a history/math story, its not scientific

Post by larryswedroe » Mon Apr 14, 2014 4:57 pm

"Larry Swedroe says we should not confuse strategy and outcome, but IMHO it is all but impossible to evaluate strategies in any fundamental way"
Consider the following

The best example is someone who is married and has young kids and decides to buy life insurance to protect against premature death. 20 years later has not died and decides his strategy was bad having wasted all those premiums. Or conversely he decides to not insure because the odds are highly in favor of the insurance company winning that bet over the next 20 years until his kids have all graduated and moved on, and he's had time to accumulate enough assets to take care of his family---and he doesn't die and thinks his strategy was good. In both cases that is making mistake of confusing strategy with outcomes.


The same type thinking applies to stocks and I'm sure we can all think of many strategies that are bad regardless of the outcome, like loading up on your company stock--some end up with Googles and others with Enrons, in either case the strategy of combining your labor and financial capital and overweighting your employer's stock is bad idea regardless of the outcome

One might say that investing in high expense actively managed funds is another example, as would be investing in hedge funds or substituting dividend paying stocks for safe bonds, and many, many other examples of this mistake.
In many cases strategies can be determined to be right or wrong BEFORE we know the outcome.


Larry

User avatar
grap0013
Posts: 1881
Joined: Thu Mar 18, 2010 1:24 pm

Re: SCV premium is a history/math story, its not scientific

Post by grap0013 » Mon Apr 14, 2014 5:57 pm

LH wrote: you certainly can tell a story though, that includes some math and some recent history.
I see you have stated the word "math" 7 times. I've got some math for you! :twisted:

Growth of 10K since my initial purchase:

US
VTI (large caps): +41.2%
PXSV +52.5%

International developed:
VEA (large caps): +36.4%
SFILX: +46.3%

Emerging markets:
VWO (large caps): +10.6%
DGS: +25.7%
There are no guarantees, only probabilities.

steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: SCV premium is a history/math story, its not scientific

Post by steve_14 » Mon Apr 14, 2014 6:14 pm

nisiprius wrote:You always have a "history" story but financial data is so badly behaved that there's never enough history. And all formal statistics assume that you have an accurate view of how the system works, and that's not true in investing. So you cannot say "because the historic statistics of the stock market were thus-and-such we can be reasonably sure that the true underlying mean is thus-and-such."
Right, so the real debate becomes: not which strategy is right or wrong, but which level of certainty is right or wrong. Some say the value premium is certain to show up most of the time. Some say they have no idea, so you'd want to diversify broadly, for example.

cowboysFan
Posts: 290
Joined: Sat Jan 25, 2014 1:46 pm

Re: SCV premium is a history/math story, its not scientific

Post by cowboysFan » Mon Apr 14, 2014 6:19 pm

larryswedroe wrote:well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size
B) We have the same premiums in developed markets all around the globe, including EM. And they are pretty similar in size as well

While investing isn't a hard science like physics we can apply scientific methods to test theories and use statistical analysis to tell us the significance of the data.

The same type data and statistics that supports the ERP supports the size and value premiums, and MOM premium as well and while no one questions the ERP many do the others, despite all the out of sample tests and statistical significance

Bottom line you can ignore the data or not. In either case your taking risks that the premiums will not show up. Take your pick, but ignoring the premiums means you must take more equity risk to achieve the same expected return, and that means more tail risk. The interesting thing is that risk averse investors, and almost all investors are risk aversion, should prefer portfolios with low beta and high tilt to a market like portfolio, assuming they can take the tracking error.

Best wishes
Larry
It seems that if markets are efficient, then any benefits there might be to a strategy of tilting to SCV will eventually go away as the strategy becomes more popular. If it really is possible to obtain a higher return for a given amount of risk through a combination of SCV/bonds compared to a combination of the the total market/bonds, then it seems that the institutional investors would soon figure this out. After they do, they will sell large caps driving down their price and buying SCV, driving up its price, until SCV has a worse expected risk-adjusted return compared to owning the total market.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 7:10 pm

larryswedroe wrote:"Larry Swedroe says we should not confuse strategy and outcome, but IMHO it is all but impossible to evaluate strategies in any fundamental way"
Consider the following

The best example is someone who is married and has young kids and decides to buy life insurance to protect against premature death. 20 years later has not died and decides his strategy was bad having wasted all those premiums. Or conversely he decides to not insure because the odds are highly in favor of the insurance company winning that bet over the next 20 years until his kids have all graduated and moved on, and he's had time to accumulate enough assets to take care of his family---and he doesn't die and thinks his strategy was good. In both cases that is making mistake of confusing strategy with outcomes.


The same type thinking applies to stocks and I'm sure we can all think of many strategies that are bad regardless of the outcome, like loading up on your company stock--some end up with Googles and others with Enrons, in either case the strategy of combining your labor and financial capital and overweighting your employer's stock is bad idea regardless of the outcome

One might say that investing in high expense actively managed funds is another example, as would be investing in hedge funds or substituting dividend paying stocks for safe bonds, and many, many other examples of this mistake.
In many cases strategies can be determined to be right or wrong BEFORE we know the outcome.


Larry
There is some wisdom in not confusing outcome with strategy. We all know people who made sensible choices that turned out badly and people who made stupid choices and things came up smelling roses. In the latter case just because things came up roses does not mean the choice was wise: I tell my kids I did stupid things as a teem and I got to where I am today not because of those things but in-spite of those things...

But there also comes a time when the evidence is just too overwhelming and we have to admit we made a mistake.

The challenge is differentiating between those cases.
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.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Re: SCV premium is a history/math story, its not scientific

Post by Rodc » Mon Apr 14, 2014 7:17 pm

cowboysFan wrote:
larryswedroe wrote:well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size
B) We have the same premiums in developed markets all around the globe, including EM. And they are pretty similar in size as well

While investing isn't a hard science like physics we can apply scientific methods to test theories and use statistical analysis to tell us the significance of the data.

The same type data and statistics that supports the ERP supports the size and value premiums, and MOM premium as well and while no one questions the ERP many do the others, despite all the out of sample tests and statistical significance

Bottom line you can ignore the data or not. In either case your taking risks that the premiums will not show up. Take your pick, but ignoring the premiums means you must take more equity risk to achieve the same expected return, and that means more tail risk. The interesting thing is that risk averse investors, and almost all investors are risk aversion, should prefer portfolios with low beta and high tilt to a market like portfolio, assuming they can take the tracking error.

Best wishes
Larry
It seems that if markets are efficient, then any benefits there might be to a strategy of tilting to SCV will eventually go away as the strategy becomes more popular. If it really is possible to obtain a higher return for a given amount of risk through a combination of SCV/bonds compared to a combination of the the total market/bonds, then it seems that the institutional investors would soon figure this out. After they do, they will sell large caps driving down their price and buying SCV, driving up its price, until SCV has a worse expected risk-adjusted return compared to owning the total market.
A couple of things. If it really is a risk story then it may come and go, but will likely be there often enough. If it is not a risk story then it may well go away forever.

If it really is a risk story it may not show well in simple risk metrics like standard deviation. But the risk may still be there.

If it really is a risk story you have to decide if it is a risk you are ok taking on.

If it is not a risk story and it goes away, I don't see any reason why the risk return would not match the risk return of the total market. That is there is no reason why it must be worse - why would people not stop bidding up the price when the risk return matched the total market - that is what classical economics would predict. In real life of course you would expect some noise causing some drift high and low, averaging about right.
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
nisiprius
Advisory Board
Posts: 36025
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: SCV premium is a history/math story, its not scientific

Post by nisiprius » Mon Apr 14, 2014 8:14 pm

Clearly_Irrational wrote:
nisiprius wrote:...First, "There are an "infinite number of strategies worse than this one" is vacuously true. I would reword it slightly: there are an infinite number of strategies that cannot be proved to be better or worse than this one. Even in hindsight, the other strategies will either have done better or worse, but there will be no way to tell whether the strategy itself was better or worse or whether the outcome was the result of chance. Larry Swedroe says we should not confuse strategy and outcome, but IMHO it is all but impossible to evaluate strategies in any fundamental way...
I don't agree. One of the few things we can do is show which strategies are bad, however absence of proof of badness is not necessarily evidence of goodness.
You're right.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
IlikeJackB
Posts: 173
Joined: Fri Dec 31, 2010 9:59 pm

Re: SCV premium is a history/math story, its not scientific

Post by IlikeJackB » Mon Apr 14, 2014 9:56 pm

Rodc wrote:
stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks. sounds good, but completely wrong for past 50 years.
You lost me here.

No one knowledgeable believes the second clause of the first sentence. Indeed as stated you have an oxymoron. A statement that is clearly false.

And what evidence do you have for the second sentence? As far as I can tell it is false as well.

You might be right about SCV, time will tell. But you seem to start with two falsehoods so I stopped reading.

ADDED: Now that I have read it, I think the replies are on target. All the OP has done is supply his own story and spin and offered up an unsubstantiated opinion. Could be right or wrong, but pretty empty of substance.
LH was referring to the time (1958/59) when the dividend yield from stocks first dropped below those of bonds. The thinking at the time was that "stocks must pay out more than bonds to get people to take the risks of stocks". And, for the next 50 years that thinking was wrong, because people still took the risk of stocks even though the dividend yield was less than bonds.

Drop down a few paragraphs in this link to see what Peter Bernstein had to say about it:
http://www.marketwatch.com/story/for-fi ... than-bonds
"Do what you will, the capital is at hazard." Justice Samuel Putnam, Harvard College vs Amory, 1830. The "Prudent Man Rule."

larryswedroe
Posts: 15635
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: SCV premium is a history/math story, its not scientific

Post by larryswedroe » Mon Apr 14, 2014 10:25 pm

cowboysfan
It seems that if markets are efficient, then any benefits there might be to a strategy of tilting to SCV will eventually go away as the strategy becomes more popular. If it really is possible to obtain a higher return for a given amount of risk through a combination of SCV/bonds compared to a combination of the the total market/bonds, then it seems that the institutional investors would soon figure this out. After they do, they will sell large caps driving down their price and buying SCV, driving up its price, until SCV has a worse expected risk-adjusted return compared to owning the total market
If you are correct than the existence and persistence of anomalies would not be possible, yet they exist and persist and have been known in many cases for decades. This is true in stocks and bonds as well. There are many reasons why such phenomenon exist and persist even after they are well known.

The bottom line is that the EMH is a very good and important model that helps us understand how markets work, but it's also true that market efficiency fails all three versions of the EMH, weak, semi-strong and strong, and anomalies do exist and persist even well after they are discovered for all types of reasons.

Now that still leaves us with the fact that for most investors the best strategy is to act as if the markets are perfectly efficient, and the EMH does pass the one test that matters most to investors---can active managers persistently outperform.


Best wishes
Larry

cowboysFan
Posts: 290
Joined: Sat Jan 25, 2014 1:46 pm

Re: SCV premium is a history/math story, its not scientific

Post by cowboysFan » Mon Apr 14, 2014 11:37 pm

larryswedroe wrote:well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size.

Best wishes
Larry
I'm not sure the data shows this after 92. When I plot the Vanguard funds over the past 10 years, small caps do win out over large caps, but small cap growth does better than SCV.

User avatar
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Re: SCV premium is a history/math story, its not scientific

Post by LH » Tue Apr 15, 2014 4:59 am

IlikeJackB wrote:
Rodc wrote:
stocks are more risky, they must pay out more than bonds to get people to take the risk of stocks. sounds good, but completely wrong for past 50 years.
You lost me here.

No one knowledgeable believes the second clause of the first sentence. Indeed as stated you have an oxymoron. A statement that is clearly false.

And what evidence do you have for the second sentence? As far as I can tell it is false as well.

You might be right about SCV, time will tell. But you seem to start with two falsehoods so I stopped reading.

ADDED: Now that I have read it, I think the replies are on target. All the OP has done is supply his own story and spin and offered up an unsubstantiated opinion. Could be right or wrong, but pretty empty of substance.
LH was referring to the time (1958/59) when the dividend yield from stocks first dropped below those of bonds. The thinking at the time was that "stocks must pay out more than bonds to get people to take the risks of stocks". And, for the next 50 years that thinking was wrong, because people still took the risk of stocks even though the dividend yield was less than bonds.

Drop down a few paragraphs in this link to see what Peter Bernstein had to say about it:
http://www.marketwatch.com/story/for-fi ... than-bonds
Yep.

If you can divorce yourself from current time, and put yourself circa 1959, think how they thought, look at what they were exposed to, its a real nice parallel. Pretty nice history, consistent, coupled with a risk story. Made great sense.

Why people back then, could even cite recent events (defined here as roughly less than 40-100 years, remember SCV is on tens of years scale minimum expectationally) to support the fact that stock yields must be greater than bond yields, or people will not take the risk! Then of course.... wrong for 50 years.

Had I wrote what I wrote in OP about the inversion circa 1959, I would have been met with a nice risk story, and non statistically significant recent history and math, and some disbelief.

Its recency bias. We just look at a recent pattern. Put some math behind it, then ignore that the math is not mathmatically signficant, then, and this is pure human, extrapolate the percieved pattern. Its humans doing what humans are wired to do. Its completely normal.

We could be having the same conversation about the stock/bond yield ratio circa 1959, as we are having about this possible SCV pattern, it would not be substantively different.

Its really unknowable in our financial lifetimes.

Maybe in 99/100 in 40 year financial lifetimes, SCV DOES win out, but maybe ours is the one it will fail, maybe we experience the 1/100, and we would say, ah, it doesnt work! We would then tell the trivially obvious story that the market knew about it, and then obliterated it. But then 99 40 year time periods after that in a row, it does work, ah, then the orginal FF story was true (after we are dead and gone, and the high SCV tilters lost out, even though they were actually "right").

Maybe its 50/50 SCV premium exists/doesnt exist (to use the term exist loosely/poorly). Who can really say? The market is wonderously complex self aware chaotic system..... Its even better than weather, as the clouds cant "decide" to go in herd behavoir sometimes, etc. (though that gets existential).

We simply do not know if SCV will pan out or not.

We are in the first financial stage where the market knows about it, the experiment has just begun.

Hey, I like the story, and the math too. I SCV tilt. Its only human, I am part of the herd here. If I was in 1959, I would likely be betting on the stock/bond yield ratio normalizing soon. It is what it is.

LH

User avatar
JoMoney
Posts: 5249
Joined: Tue Jul 23, 2013 5:31 am

Re: SCV premium is a history/math story, its not scientific

Post by JoMoney » Tue Apr 15, 2014 6:28 am

cowboysFan wrote:...When I plot the Vanguard funds over the past 10 years, small caps do win out over large caps, but small cap growth does better than SCV.
The past is the past, but I don't think small-caps of any flavor are a panacea... I'm going to post this chart once again:

Image
...along with this one...
Image
and note that Morningstar has Forward looking P/E on the Russell 2000 around 19.35
and forward P/E on Russell Top 200 around 16.00
19.35/16 = 1.2+ ... and if you looked at the spread between small and large on trailing P/E instead of forward estimates it's astonishingly high even considering differences in actual earnings growth rates...

IF we were measuring from 1983 forward, and the scenario like the past 30 years repeated, we could be talking about a 60 year period where it appeared that large-caps ruled the roost, with future generations (hypothetically) talking about Large-Caps and "Quality" premiums "except for that silly small-cap craze in the early 2000's"...
It doesn't seem far-fetched to me considering the small-cap boom that ended in the early 80's began with the fall of the "Nifty-Fifty" large-growth-quality craze of the 70's, and the recent small-cap boom began with the ending of the large-cap boom of the late 90's dot-com bubble...
..and I recently stumbled across an old quote indicating similar boom-bust trends dating back earlier:
The "new-era" doctrine—that "good" stocks (or "blue chips") were sound investments regardless of how high the price paid for them—was at the bottom only a means of rationalizing under the title of "investment" the well-nigh universal capitulation to the gambling fever.
—Benjamin Graham and David Dodd, 1934
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

larryswedroe
Posts: 15635
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: SCV premium is a history/math story, its not scientific

Post by larryswedroe » Tue Apr 15, 2014 7:48 am

cowboys
would you draw the same conclusion by looking at the returns of stocks vs LT Treasuries for the much longer 40 year period 69-08 when they had same returns, no ERP?
Over even fairly long periods risks can show up and that's what happened over that particular 10 year period.
Try the 5 year period and the 15 year period and the longer period. The longer the data the more confidence we can have in it's reliability (which is never 100%)
Larry

User avatar
Ketawa
Posts: 1921
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: SCV premium is a history/math story, its not scientific

Post by Ketawa » Tue Apr 15, 2014 8:13 am

cowboysFan wrote:
larryswedroe wrote:well I guess that it's purely a coincidence that
A) After FF original study showing size and value premiums for 63-92, that we now have out of sample tests both before 63 and after 92 that show not only the same premiums but similar size.

Best wishes
Larry
I'm not sure the data shows this after 92. When I plot the Vanguard funds over the past 10 years, small caps do win out over large caps, but small cap growth does better than SCV.
If you only looked at the last 10 years, you aren't looking at "after 92". In the last 10 years we had a recession where value stocks were hit especially hard. And I don't think any value investors expect that it will always beat growth over 10 year periods, just like stocks don't always beat bonds.

If you go back to inception of the Vanguard funds, you should know that there was an index switch in 2003 which caused Vanguard SCG to outperform either index that it followed. Market timing worked!

freebeer
Posts: 1979
Joined: Wed May 02, 2007 8:30 am
Location: Seattle area USA

Re: SCV premium is a history/math story, its not scientific

Post by freebeer » Tue Apr 15, 2014 8:55 am

nisiprius wrote:Oh, and of course we always need to remember that...
"Investing is not a science."--John C. Bogle...
Tell that to the quants that are making billions for proprietary trading firms via high frequency trading!

Well, as LH noted each of us is part of the investing herd and what one does as part of a herd is not scientific, it is also emotional. That's true for all herds. So in that sense Mr. Bogle was totally right.

But science can to some extent predict herd behaviors - the fact that lemmings aren't fully rational in jumping off cliffs doesn't mean that science is inapplicable to studying and in some cases predicting lemming behavior in the aggregate. Same holds for the behavior, in the aggregate, of humans investing in financial markets.

Quincy
Posts: 41
Joined: Tue May 28, 2013 3:28 pm

Re: SCV premium is a history/math story, its not scientific

Post by Quincy » Tue Apr 15, 2014 9:17 am

Comparing the ERP to the SCV premium is not comparing apples to apples. The SCV premium has many different and contentious explanations, some of which you have to "believe". The ERP is undeniably a case of higher risk and if there are any truths at all in finance risk=expected return should be considered one of them. The ERP is dictated by law, bondholders get paid first. The SCV premium is a "risk story" or maybe a "behavior story" or maybe just an anomaly or maybe it doesn't really exist at all. Nobody knows for sure going forward. Until the laws change equities are undeniabley riskier than bonds. I believe constructing portfolios going forward expecting the same premium from SCV that you get from the ERP is a mistake, but that is just my amateur opinion and we all know what those are like. I would like to add that our SCV heroes Berstein, Swedroe, and Ferri are indeed all paid financial advisors. I highly value their opinions and teachings. I want to thank all of them for educating me about the financial markets. However, how much is their opinion on the small and value story tainted by the fact that they get paid a lot of money to construct complicated slice and dice portfolios for clients that provide higher returns for less risk? If you were a wealthy client and walked into their firms and were told to put all of your money in a 3 fund portfolio at Vanguard because beating the market has proven to be a losers game for the vast majority of investors, would you feel like you were getting your moneys worth? In the end our index heroes may have fallen victim to the "I can beat the market" trap that every financial advisor espouses. Sorry guys, but you are the ones who taught me that.

User avatar
matjen
Posts: 1883
Joined: Sun Nov 20, 2011 11:30 pm

Re: SCV premium is a history/math story, its not scientific

Post by matjen » Tue Apr 15, 2014 10:01 am

Quincy wrote: However, how much is their opinion on the small and value story tainted by the fact that they get paid a lot of money to construct complicated slice and dice portfolios for clients that provide higher returns for less risk? If you were a wealthy client and walked into their firms and were told to put all of your money in a 3 fund portfolio at Vanguard because beating the market has proven to be a losers game for the vast majority of investors, would you feel like you were getting your moneys worth? In the end our index heroes may have fallen victim to the "I can beat the market" trap that every financial advisor espouses. Sorry guys, but you are the ones who taught me that.
Understand where you are going but doesn't explain Robert T and his immense contributions to this forum. I count him as a "hero." http://www.bogleheads.org/forum/viewtop ... f=1&t=7353

There are many others who are pretty technical who have looked at the data/evidence and decided to tilt SCV, etc.
A man is rich in proportion to the number of things he can afford to let alone.

steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: SCV premium is a history/math story, its not scientific

Post by steve_14 » Tue Apr 15, 2014 11:53 am

Quincy wrote:I would like to add that our SCV heroes Berstein, Swedroe, and Ferri are indeed all paid financial advisors. I highly value their opinions and teachings. I want to thank all of them for educating me about the financial markets. However, how much is their opinion on the small and value story tainted by the fact that they get paid a lot of money to construct complicated slice and dice portfolios for clients that provide higher returns for less risk?
Good points, keep in mind that two of those three folks have also, publicly and in the media, advocated total market portfolios as well. I don't think anyone except Larry advertises slice and dice as a guaranteed home run.

So the list of "experts" advocating such a strategy is tiny. The stream of newbie investors wanting to believe on this forum, however, is steady. And the list of average Americans that would buy into it based on a nice back-tested chart or two? Endless. DFA was brilliant to start a small growth fund a couple of years ago. You want to make sure your funds cover all the backtestable bases 10 years from now.

To generalize, the test of your investment expertise is, when confronted with the claim of a free lunch, how easily you buy into it.

swaption
Posts: 1184
Joined: Tue Jul 29, 2008 11:48 am

Re: SCV premium is a history/math story, its not scientific

Post by swaption » Tue Apr 15, 2014 12:15 pm

Rodc wrote:A couple of things. If it really is a risk story then it may come and go, but will likely be there often enough. If it is not a risk story then it may well go away forever.

If it really is a risk story it may not show well in simple risk metrics like standard deviation. But the risk may still be there.

If it really is a risk story you have to decide if it is a risk you are ok taking on.

If it is not a risk story and it goes away, I don't see any reason why the risk return would not match the risk return of the total market. That is there is no reason why it must be worse - why would people not stop bidding up the price when the risk return matched the total market - that is what classical economics would predict. In real life of course you would expect some noise causing some drift high and low, averaging about right.
I guess if it is not a risk story, then do we have enough data to assume it is some sort of story? The logical candidate being something behavioral, which really doesn't get a lot of respect in terms of potentiial for any sort of systematic phenomena. I'm at least open to the possibility of something that can come and go, but not necessarily be related to an increased risk.

User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: SCV premium is a history/math story, its not scientific

Post by Clearly_Irrational » Tue Apr 15, 2014 12:24 pm

swaption wrote:I guess if it is not a risk story, then do we have enough data to assume it is some sort of story? The logical candidate being something behavioral, which really doesn't get a lot of respect in terms of potentiial for any sort of systematic phenomena. I'm at least open to the possibility of something that can come and go, but not necessarily be related to an increased risk.
My personal opinion is that small is a risk story and value is a behavioral story like momentum. Since both of those (value & momentum) are based on human nature I don't think that will change until we get 2nd generation strong AI (AI's designed by other AI's), and at that point pretty much all bets are off anyways. I suppose the addition of friendly capitalist aliens or the genetic re-engineering of humans with different mental characteristics would also be possibilities for change. In the meantime I'd say they're pretty likely to continue.

steve_14
Posts: 1507
Joined: Wed Jun 20, 2012 12:05 am

Re: SCV premium is a history/math story, its not scientific

Post by steve_14 » Tue Apr 15, 2014 12:29 pm

swaption wrote:I guess if it is not a risk story, then do we have enough data to assume it is some sort of story?
It might be useful to think in terms of countries. Australia has outperformed Italy by a large margin. What's the story there (there must be one), and should be overweight Australian stocks?

User avatar
tadamsmar
Posts: 7681
Joined: Mon May 07, 2007 12:33 pm

Re: SCV premium is a history/math story, its not scientific

Post by tadamsmar » Tue Apr 15, 2014 12:42 pm

larryswedroe wrote: Bottom line you can ignore the data or not. In either case your taking risks that the premiums will not show up. Take your pick, but ignoring the premiums means you must take more equity risk to achieve the same expected return, and that means more tail risk. The interesting thing is that risk averse investors, and almost all investors are risk aversion, should prefer portfolios with low beta and high tilt to a market like portfolio, assuming they can take the tracking error.
That's true, ignoring these premiums mean you have to take on more risk to get the same expected return. This is a fact whether the premiums exist or not.

Inspired by this amazing logic, I have decided to invent the tadamsmar premium. The tadamsmar premium starts now and all stocks will yield this premium in the future. They will grow 10 times faster than their historical rate.

I won't bother to provide any evidence for the tadamsmar premium, it does not matter if it exists or not.

Take your pick, but ignoring the tadamsmar premium means you must take more equity risk to achieve the same expected return, and that means more tail risk.

Yes, Virginia, mere faith in a premium changes your expectations.

(Of course, there is historical evidence for the value and size premium, but projecting it into the future is a bit theory-laden. But then there is evidence that it worked in the future after is was revealed. But I for one am not convinced that it will persist, since it's persistence depends on it's ability to continue to be unconvincing no matter how much evidence piles up.)
Last edited by tadamsmar on Tue Apr 15, 2014 12:59 pm, edited 5 times in total.

Quincy
Posts: 41
Joined: Tue May 28, 2013 3:28 pm

Re: SCV premium is a history/math story, its not scientific

Post by Quincy » Tue Apr 15, 2014 12:42 pm

Warren Buffett, famed investor: "There seems to be some perverse human characteristic that likes to make easy things difficult." Most of us just cannot accept the majesty of simplicity.

Post Reply