"Smart Beta", Value and other Factors/Premia

[CorpRaider]

I have attached a paper from Research Affiliates discussing some analysis of combining various alternative index weighting methodologies/factor exposures.  I read these kinds of papers fairly often.  I thought this one was pretty interesting.  It's not new, so may have been circulated here before.  If so, my apologies.

 

 

 

Building_A_Better_Beta.pdf

[CorpRaider]

I figured I would modify this thread so we could share interesting academic, quasi-academic papers, articles, blog posts and so forth concerning value, other potential premia, the various potential factors identified by researchers and strategies to try and take advantage thereof.  The strategies section seemed most appropriate.  I know most of us practice active, fundamental stock picking, but I personally am interested in some of these issues; for use as recommendations for friends and family if nothing else. 

 

Here's another one;  Jason HSU of RAFI on traditional value indexes versus their fundamental indexes.  Obviously he has an agenda, but nevertheless:

 

Became somewhat interested in this question (fundamental versus traditional value index) based on Greenblatt's recommendation of IWD on his most recent wealthtrack appearance, given his conclusions in the Big Secret which led me to expect him to recommend a RAFI or EW index.

JII_Summer_2014_RA.pdf

[innerscorecard]

Makes sense that there's a complete lack of interest in factor investing here. Makes sense. People have spent a lot of time learning to analyze companies. They don't want to throw it all away in favor of factor tilts.

 

It's also conversely funny to me how obsessed Bogleheads are with factor tilts. They're convinced that picking any individual securities is foolhardy gambling, yet express little skepticism that widely known factors are a way to increase returns.

[CorpRaider]

Hey, there are about 20 people who seem to be downloading.  I've got another one I've been meaning to upload from Blackrock on a strategy combining quality and value that is implemented in their enhanced index funds (I'm not sure what device I have it saved on).

 

One thing I've finally been able to get comfortable with is that I no longer believe in the small cap factor.  I've read some of the research (I think some of it was from RAFI), including a neat post summarizing the same from Wes Gray, but additional confirmation came from finding quotes from Buffett (maybe Munger) and now Greenblatt agreeing on this as well.  I have no idea where I found the Buffett one, but JG discussed it in the Schwager HF Wizards book. 

 

I sort of just try and stay abreast of it and track some of the strategies as benchmarks.

[The Falcon]

CorpRaider, appreciate the efforts. Its an interesting subject, thanks for the taking to time to upload the docs.

[CorpRaider]

Thanks.  As punishment for your kind response, here's a paper from S&P just evaluating various alternative weighting methodologies.  haha.  Nothing new really, just another source sort of verifying what RAFI and others (including J Greenblatt) have concluded based on current evidence.

 

research-evaluating-alternative-beta-strategies.pdf

[CorpRaider]

...and here's the paper from the BlackRock researchers on a quality-value strategy.  Seems more quality focused.  I believe this is the strategy currently being implemented by the BlackRock Enhanced factor etfs (IELG, IESM, IEIS, IEIL).  I have also attached the much discussed paper from Novy-Marx examining various measures of quality (came across it via references reading some stuff from AQR and RAFI).

 

Qual_+_Value_BlackRock.pdf

Norvy-Marx_Quality_Accruals.pdf

[Pretium]

So what worries me about these "smart" beta strategies is that there is a significant amount of data mining going on. Basically you take all this historical data and try to find factors that explain outperformance. Eventually you are going to find something in the data, but it doesn't mean it will persist into the future.

 

Also, see Bill Sharpe's response to smart beta (http://www.advisorperspectives.com/newsletters14/Bill_Sharpe-Smart_beta_makes_me_sick.php)

 

Essentially, smart beta relies on dumb beta. Somebody has to be on the other side of the trade. Now that the historical evidence is published and found ubiquitously on the web, it's very likely to be arbitraged away.

[CorpRaider]

Yeah, you might be right.  It is a legitimate concern.  I personally doubt that value will be arbitraged away for the reasons you posted in another thread (see someone did notice your post, haha).  Glamour and recent performance seem to attract we humans like moths to a flame.  Yet value will surely go through periods of material underperformance. 

 

Undoubtedly some purported factors are the result of data mining.  Perhaps the first of these to be revealed will be the small cap premium "discovered" by the most respected of all the EMH proponents.  I would note that the RAFI 1000 etf has been "live" since 12/2005 and has outperformed the R1000 by around 1.3% and of course the equal weight S&P etf has been around and outperforming for over a decade in live fire action.  I also think S&P and/or MSCI have been calculating the low volatility indexes in real time for a while.

[Pretium]

Every time Joel Greenblatt does an interview about his book, "The Little Book that Beats the Market", they ask him why the value premium will persist now that he's revealed the secret. And every time he responds "Because the strategy doesn't work all the time, and institutional investors don't have the patience. But over the long term you get 1-2% over the index".

 

I've definitely found this lack of patience to be true with the firms I work with.

 

On the flip side, the S&P 500 Index has worked. It beats the majority of active funds. Warren Buffett recommends it because you own a cross section of America's businesses, which are almost assured to march forward over time. And you pay very low costs. Why fix what ain't broken?

 

Of course we all know that 1-2% over 20 years is absolutely huge in terms of final growth of capital. It's just a question of distinguishing between marketing and a sound strategy.

[innerscorecard]

Every time Joel Greenblatt does an interview about his book, "The Little Book that Beats the Market", they ask him why the value premium will persist now that he's revealed the secret. And every time he responds "Because the strategy doesn't work all the time, and institutional investors don't have the patience. But over the long term you get 1-2% over the index".

 

I've definitely found this lack of patience to be true with the firms I work with.

 

On the flip side, the S&P 500 Index has worked. It beats the majority of active funds. Warren Buffett recommends it because you own a cross section of America's businesses, which are almost assured to march forward over time. And you pay very low costs. Why fix what ain't broken?

 

Of course we all know that 1-2% over 20 years is absolutely huge in terms of final growth of capital. It's just a question of distinguishing between marketing and a sound strategy.

 

There's nothing magical about the S&P 500 index in particular. For example, a broad-market index is even more representative of the economy and likely just as cheap to represent in index fund form.

[CorpRaider]

Of course, as you know, the market cap weighted index is broken according to Arnott, Greenblatt and others (or at least it has a big problems).  It will systematically overweight the glamour stocks, according to them.

[Pretium]

Of course, as you know, the market cap weighted index is broken according to Arnott, Greenblatt and others (or at least it has a big problems).  It will systematically overweight the glamour stocks, according to them.

 

I will point out that both Greenblatt and Arnott are selling products that benefit if people believe that market cap weighting is flawed. What they argue makes sense, but so does what the academic world says (see attached).

 

Response_to_Rob_Arnott.pdf

[CorpRaider]

Yeah, I enjoyed reading both of those. Thanks for posting.  Arnott's response to the prior response was really interesting.  Those circa 2009 questions about the relative underperformance of the RAFI 1000 haven't been asked in a while.  haha. 

 

Cliff Asness has (famously) made the argument that the RAFI is merely a value tilt, but is "still awesome."  Perhaps more interesting to me is that he also makes the argument that research indicating that weighting by anything else other than market cap or their inverses, as researched by Arnott and others, also represents a value tilt. 

 

After reading Arnott's response referenced in the linked article, where Arnott makes the argument that based on his perspective, the market cap weighted index is really a tilt to growth and/or momentum, so that most things other than the cap weighted index are going to be labeled a value tilt, seemingly just the flip side of a coin.  That perspective does, however, make more sense when you take this evidence about any other weighting method outperforming (dividends, sales, earning, fundamental index, low volatility, equal weighting, etc...).  Then, maybe Arnott is onto something and it looks more like you are removing the drag of growth/glamour tilt rather than one specific method that represents a value tilt or saying that all these alternative weighting methods are value tilts.

 

I guess the big risk is that you could get a value tilt cheaper, but the Schwab RAFI indexes are relatively cheap and the Blackrock one (VLUE) is about as cheap as most anything out there (.15% ER), even the cap-weighted value subset indexes.

[CorpRaider]

http://aswathdamodaran.blogspot.com/2015/04/the-small-cap-premium-fact-fiction-and.html

 

Another piece questioning the existence of the small cap premium.  This time from Prof. Damodaran of NYU. 

[CorpRaider]

"Value weighted" 500 index ETF from Diamond Hill Asset Management went live today (link below to site).  It seems like one to watch, especially if the fee waiver sticks (at .10% basis point ER for duration of fee waiver).  Closest thing to big secret for small investor (Greenblatt) that I've seen in an ETF (RPV and the pure style suite maybe being the next closest). 

 

Anyone familiar with Diamond Hill?  I don't think I've heard of them.

 

http://www.diamond-hill.com/strategies/long-only-equities/etf/valuation-weighted-500-etf/overview/

[EliG]

Closest thing to big secret for small investor (Greenblatt) that I've seen in an ETF (RPV and the pure style suite maybe being the next closest).

 

Why do you think it's closer to Greenblatt than RPV?

 

RPV is also weighted by value but it's more concentrated. 120 stocks in RPV. 500 stocks in Diamond Hill.

 

I note that Diamond Hill uses a forward-looking valuation process. RPV is 100% backward-looking (it uses a composite of TTM p/e, p/b, p/s). I like the general idea of forward-looking valuation, but the devil is in the details.

[CorpRaider]

Well mainly because jgb classifies it as a classic value fund on the valueweighted index site, i suppose thats maybe due to the concentration.  I dont like that  the diamond hill etf uses forward looking analyst estimates, as most of the literature ive read indicates estimates are often wrong maybe for the same reasons as cap weighting and suffers from recency bias and the other behavioral flaws endemic in the markets.  But it is worth watching.

[netnet]

Sharpe is a really smart guy, but he argument if followed to its logical conclusion is that everyone would be invested in indexes, which is clearly not true.  He acknowledges that in fact that "worried" him but people still look for over and priced securities, thus he destroys his own argument against the smart beta funds.

[frommi]

Found a very interesting paper regarding momentum premium and concentration. Monthly return of 2.9% after trading costs with 8 stock portfolios from 1991 to 2010. Returns look a lot higher than what i have seen with value investing, and when i look at some of the recent return threads it looks like some people here are doing something like this and just label it value investing. Maybe a good idea.

SSRN-id2420743.pdf

[WideMoat]

Found a very interesting paper regarding momentum premium and concentration. Monthly return of 2.9% after trading costs with 8 stock portfolios from 1991 to 2010. Returns look a lot higher than what i have seen with value investing, and when i look at some of the recent return threads it looks like some people here are doing something like this and just label it value investing. Maybe a good idea.

 

The tools that I have immediately available don't have six month price performance data.  If you take YTD price performance for large cap US stocks, today the strategy would put you in Anacor Pharma (up 337% YTD), Skechers (155%), Netflix (110%), Dexcom (82%), Intrexon (81%), Amazon (74%), Incyte (72%), and Jetblue (70%).  The strategy would have you revisit these stocks in 1-6 months' time.

 

If you take 13 week price performance for US stocks above 1.16B market cap, the strategy would put you in Anacor (up 95%), Intra Cellular Therapies (86%), Lexicon Pharma (85%), Exelixis (70%), Sucampo Pharma (66%), LendingTree (61%), DBV Technologies (56%), Abiomed (54%). 

 

I own none.

[frommi]

Found a very interesting paper regarding momentum premium and concentration. Monthly return of 2.9% after trading costs with 8 stock portfolios from 1991 to 2010. Returns look a lot higher than what i have seen with value investing, and when i look at some of the recent return threads it looks like some people here are doing something like this and just label it value investing. Maybe a good idea.

 

The tools that I have immediately available don't have six month price performance data.  If you take YTD price performance for large cap US stocks, today the strategy would put you in Anacor Pharma (up 337% YTD), Skechers (155%), Netflix (110%), Dexcom (82%), Intrexon (81%), Amazon (74%), Incyte (72%), and Jetblue (70%).  The strategy would have you revisit these stocks in 1-6 months' time.

 

If you take 13 week price performance for US stocks above 1.16B market cap, the strategy would put you in Anacor (up 95%), Intra Cellular Therapies (86%), Lexicon Pharma (85%), Exelixis (70%), Sucampo Pharma (66%), LendingTree (61%), DBV Technologies (56%), Abiomed (54%). 

 

I own none.

 

I would probably never do it with an unfiltered list of stocks, but when you have a watchlist of high quality names trading below fair value it may be a good sort criteria.

On my list it would put me into IBKR,CSU.TO,MKL,LKQ,MCO,TDG,PCLN,MA. Whats interesting is that there is a one month reversal window, so ideally you want the stocks that are up the most in the last 6 months and are down in the last month.

[theasiareport]

Found a very interesting paper regarding momentum premium and concentration. Monthly return of 2.9% after trading costs with 8 stock portfolios from 1991 to 2010. Returns look a lot higher than what i have seen with value investing, and when i look at some of the recent return threads it looks like some people here are doing something like this and just label it value investing. Maybe a good idea.

 

The tools that I have immediately available don't have six month price performance data.  If you take YTD price performance for large cap US stocks, today the strategy would put you in Anacor Pharma (up 337% YTD), Skechers (155%), Netflix (110%), Dexcom (82%), Intrexon (81%), Amazon (74%), Incyte (72%), and Jetblue (70%).  The strategy would have you revisit these stocks in 1-6 months' time.

 

If you take 13 week price performance for US stocks above 1.16B market cap, the strategy would put you in Anacor (up 95%), Intra Cellular Therapies (86%), Lexicon Pharma (85%), Exelixis (70%), Sucampo Pharma (66%), LendingTree (61%), DBV Technologies (56%), Abiomed (54%). 

 

I own none.

 

Won't you get the reverse momentum effect when prices mean revert?

[frommi]

Attached a study about seasonality, value and size premium. According to that the p/b and smallcap outperformance is mainly attributable to the january effect.

SSRN-id784186.pdf

[CorpRaider]

Hey, have you guys seen some of the recent discussion (I actually think the research is old) about skewness in the market and how most stocks underperformed T-bills and the overall market return premium is attributable to a small number of huge baggers? 

 

Here's a placeholder blurb from Bloomberg about it:

 

https://www.bloomberg.com/news/articles/2017-04-09/lopsided-stocks-and-the-math-explaining-active-manager-futility

 

I will find a link to some posts/papers, but man that makes it hard to envision any strategy beating the cap weighted total market over the long term.  Kind of depressing.