Larry Swedroe: Problems With The Factor Zoo

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Random Walker
Posts: 2123
Joined: Fri Feb 23, 2007 8:21 pm

Larry Swedroe: Problems With The Factor Zoo

Post by Random Walker » Fri May 19, 2017 9:36 am

http://www.etf.com/sections/index-inves ... factor-zoo

Larry explains how a world of 600 potential factors can be narrowed down to 7 that are significant. As I believe is the important lesson from his most recent book, he describes 5 criteria the investor can use to decide for himself if a new potential factor is worthy of inclusion in the portfolio: persistent, pervasive, robust, intuitive, investable.

Dave

Random Walker
Posts: 2123
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by Random Walker » Fri May 19, 2017 9:44 am

And to reply to my own post :-), I'll say that when considering a new addition to a portfolio, one should look at expected return, volatility, correlations, costs. The most basic, important, and cheapest allocation decision is the basic stock/bond split. Additional more esoteric factors have decreasing marginal benefit and increased cost. But they nonetheless do have benefit! There is a true cost to a less efficient portfolio, and improved portfolio efficiency can make the increased costs of more esoteric sources of return worthwhile.

Dave

QuietProsperity
Posts: 343
Joined: Mon May 11, 2015 2:53 pm
Contact:

Re: Larry Swedroe: Problems With The Factor Zoo

Post by QuietProsperity » Fri May 19, 2017 11:18 am

What I found interesting in that paper, "Replicating Anomalies" by Hou, Xue & Zhang, is that a number of the Quality/Profitability factors did not pass their testing.

The following were all found to be insignificant:

Return on Equity (6 and 12-month re-balancing periods)
Return on Assets (6 and 12-month re-balancing periods)
Return on Net Operating Assets
Profit Margin
Operating Profits to Equity
Fundamental Score (F Score)

There were others, but those are the ones I read about the most out in the wild.

It was not a good showing for the quality/profitability factors.

From a personal stand-point, the paper gave me additional confidence in my choice to use small, value and momentum to accompany my overall market exposure, but it is just another reminder of just how much a crap-shoot this all is.

Another good paper on the subject of the significance of factors is "...and the Cross-Section of Expected Returns" by Campbell, Liu & Zhu

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Fri May 19, 2017 5:20 pm

Intuitive: There are logical risk-based or behavioral-based explanations for its premium and why it should continue to exist.

The good news is that, among all the factors in the zoo, we show that you need to focus only on the eight that meet our criteria: beta, size, value, momentum, profitability, quality, term and carry.
Anyone seen an intuitive argument for quality or profitability that you found convincing?

livesoft
Posts: 57281
Joined: Thu Mar 01, 2007 8:00 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by livesoft » Fri May 19, 2017 5:35 pm

Random Walker wrote:And to reply to my own post :-), I'll say that when considering a new addition to a portfolio,….
I'll just say one only has to look for the Larry Swedroe Seal of Approval.
This signature message sponsored by sscritic: Learn to fish.

User avatar
knpstr
Posts: 1827
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: Larry Swedroe: Problems With The Factor Zoo

Post by knpstr » Fri May 19, 2017 6:35 pm

livesoft wrote:
Random Walker wrote:And to reply to my own post :-), I'll say that when considering a new addition to a portfolio,….
I'll just say one only has to look for the Larry Swedroe Seal of Approval.
+1
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

garlandwhizzer
Posts: 1670
Joined: Fri Aug 06, 2010 3:42 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by garlandwhizzer » Fri May 19, 2017 8:15 pm

I would like to thank Random Walker for posting Larry's articles consistently. I always find them to be educational and I suspect most Forum members do as well. I miss Larry's postings but fully understand why as a busy professional he decided to spend his time and effort elsewhere. At least thanks to Random Walker we still get Larry's new articles which is a lot better than no input at all.

Garland Whizzer

slowmoney
Posts: 164
Joined: Tue Dec 28, 2010 4:41 pm
Location: Louisiana Purchase

Re: Larry Swedroe: Problems With The Factor Zoo

Post by slowmoney » Fri May 19, 2017 8:28 pm

Garland Whizzer wrote:
I would like to thank Random Walker for posting Larry's articles consistently.
I agree. :D


Larry Swedroe finished the article with this:
“First, as we have discussed, all factors—including the ones we have recommended—have experienced long periods of underperformance. So, before investing, be sure that you believe strongly in the rationale behind the factor and the reasons why you trust it will persist in the long run. Without this strong belief, it is unlikely that you will be able to maintain discipline during the inevitable long periods of underperformance. And discipline is one of the keys to being a successful investor. Finally, because there is no way to know which factors will deliver premiums in the future, we recommend that you build a portfolio broadly diversified across them. Remember, it has been said that diversification is the only free lunch in investing. Thus, we suggest you eat as much of it as you can!”
Information is more valuable sold than used. - Fischer Black (1938-1995)

QuietProsperity
Posts: 343
Joined: Mon May 11, 2015 2:53 pm
Contact:

Re: Larry Swedroe: Problems With The Factor Zoo

Post by QuietProsperity » Fri May 19, 2017 10:17 pm

lazyday wrote:Anyone seen an intuitive argument for quality or profitability that you found convincing?
As a screen, I like it. As a stand alone factor, I have not.

To me, factors like size, value and momentum have stories I can believe in. The quality/profitability story, at least as a standalone factor, has always been hard for me to believe in.

To me, the most important part about the paper was that they excluded micro-cap stocks from their analysis. I think that is what caused a lot of the anomalies to disappear...which is a good thing.

Most of us will never have meaningful exposure (if any) to those tiny tiny micro stocks...so if certain factors only seem to be significant in that space, that does us, the investors, no good.

Just my thoughts.

Random Walker
Posts: 2123
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by Random Walker » Fri May 19, 2017 10:46 pm

I have trouble with quality/profitability as well. Perhaps cause I don't understand it well. I once heard a tenuous Risk explanation that more profitable companies are most subject to new competition.

Dave

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: Larry Swedroe: Problems With The Factor Zoo

Post by NiceUnparticularMan » Sat May 20, 2017 6:57 am

I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Sat May 20, 2017 3:00 pm

QuietProsperity wrote:As a screen, I like it.
NiceUnparticularMan wrote:I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.
Is there a fund or index you have in mind?

gtwhitegold
Posts: 200
Joined: Fri Sep 21, 2012 1:55 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by gtwhitegold » Sat May 20, 2017 6:53 pm

AQR Multi-Style, DFA, and the iShares Edge Multi-factor funds all screen for quality. Also, any funds that us price to equity or price to cash flow screens will have a quality tilt.
lazyday wrote:
QuietProsperity wrote:As a screen, I like it.
NiceUnparticularMan wrote:I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.
Is there a fund or index you have in mind?

User avatar
knpstr
Posts: 1827
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: Larry Swedroe: Problems With The Factor Zoo

Post by knpstr » Sat May 20, 2017 7:09 pm

slowmoney wrote:Garland Whizzer wrote:
I would like to thank Random Walker for posting Larry's articles consistently.
I agree. :D
Why not just bookmark his etf.com page?
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

User avatar
knpstr
Posts: 1827
Joined: Thu Nov 20, 2014 8:57 pm
Location: Michigan

Re: Larry Swedroe: Problems With The Factor Zoo

Post by knpstr » Sat May 20, 2017 7:12 pm

lazyday wrote:
QuietProsperity wrote:As a screen, I like it.
NiceUnparticularMan wrote:I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.
Is there a fund or index you have in mind?
Why not just screen for stocks with "quality" or whatever other "factors", rather than funds?
Very little is needed to make a happy life; it is all within yourself, in your way of thinking. -Marcus Aurelius

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: Larry Swedroe: Problems With The Factor Zoo

Post by NiceUnparticularMan » Sat May 20, 2017 9:46 pm

lazyday wrote:
QuietProsperity wrote:As a screen, I like it.
NiceUnparticularMan wrote:I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.
Is there a fund or index you have in mind?
So the three funds I actually use are DFFVX (DFA Targeted Value), IJS (iShares S&P 600 Value), and VBR (Vanguard Small Cap Value). All of them have generated positive alpha in three-factor regressions since inception, which disappears in five-factor regressions.

DFFVX appears most effective, but it isn't that cheap if you have to pay an additional free for access. I'm stuck with the fee in a certain 401K, so I use it in that account since the fee is a sunk cost.

IJS is a nice choice in taxable accounts, so I use it there. VBR is very low cost in tax-advantaged accounts, so I use it in those.

By the way, I am a proponent of tilting using SP500 and one or more of these funds rather than TSM in place of SP500. SP500 also de facto adds a quality screen (it screens out "junk" small growth stocks too), but you do have to use more SV to then balance out the fact SP500 is negative small.

QuietProsperity
Posts: 343
Joined: Mon May 11, 2015 2:53 pm
Contact:

Re: Larry Swedroe: Problems With The Factor Zoo

Post by QuietProsperity » Sat May 20, 2017 9:49 pm

knpstr wrote: Why not just screen for stocks with "quality" or whatever other "factors", rather than funds?
For myself, three reasons:

1. With how inexpensive factor funds are these days, the time/cost trade off of holding a few funds versus building a portfolio with small-cap value/quality stocks, large-cap momentum stocks in both the US and international markets, is not worth it to me.
2. From an emotional standpoint, I learned a long time ago that I do not do well with individual stocks...too quick to pull the trigger on buys/sells.
3. The savings from ditching "fund expenses" would quickly be lost from bid/ask spreads in the small-cap space for both the U.S. and International markets imo.

So sure, I could "save" ~20bps a year in fund expenses and instead build some elaborate stock screener, but for the above reasons, not worth it to me.

NiceUnparticularMan
Posts: 1740
Joined: Sat Mar 11, 2017 7:51 am

Re: Larry Swedroe: Problems With The Factor Zoo

Post by NiceUnparticularMan » Sat May 20, 2017 9:50 pm

knpstr wrote:
lazyday wrote:
QuietProsperity wrote:As a screen, I like it.
NiceUnparticularMan wrote:I'm also more interested in "quality" factors as screening rather than independent sources of return. And the nice thing about that perspective is you can get quality screening, which shows up as "alpha" in three-factor regressions that disappears in five-factor regressions, REALLY cheap.
Is there a fund or index you have in mind?
Why not just screen for stocks with "quality" or whatever other "factors", rather than funds?
I'm not sure I understand the question. Assembling your own diversified portfolio of individual stocks is a bit of a pain in the neck. If you use any of the funds I mentioned, plus SP500 if you don't want to tilt that far, you can get a diversified portfolio with a quality screen for cheap, and very easily.

Theoretical
Posts: 1097
Joined: Tue Aug 19, 2014 10:09 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by Theoretical » Sat May 20, 2017 10:38 pm

I tend to view quality/profitability as one of two ways.

The best positive explanation (for it being a risk factor) I've heard is that it relates to competitive advantages that can be targeted for reduction or elimination by rival companies. For example, having THE top technology and the execution to the marketplace for it or an unusually good management culture. This "target on the back" premium is to me a bit weak.

What I think the stronger argument is is that the short side does better than it should for long periods, given its junkiness. So it shows up as a risk in bull markets that the quality stocks will underperform. The outperformance is that the junk stocks end up being somewhat to grossly overvalued in times of distress while the quality stocks don't suffer as much.

Neither one is enough for me to explicitly target the factor. I'm good with screens like the S&P, DFA or the PIMCO RAE approaches. I'm not so interested in explicitly targeting "Quality stocks"

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Sun May 21, 2017 5:19 am

Thanks for the replies on funds that screen.

I've heard of DFA and Bridgeway having various screens and was curious about others.

S&P's methodology for choosing stocks in the 500/400/600 seems to be something of a quality screen. But I already forgot about that when I asked!
gtwhitegold wrote:AQR Multi-Style, DFA, and the iShares Edge Multi-factor funds all screen for quality. Also, any funds that us price to equity or price to cash flow screens will have a quality tilt.
If I remember correctly, the Edge funds have a quality weighting, but not a screen?

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Sun May 21, 2017 5:21 am

knpstr wrote:Why not just screen for stocks with "quality" or whatever other "factors", rather than funds?
Around 2010 or so, I did something like that. There were no Quality funds I cared for back then.

In this thread I'm not asking for funds I'll actually buy, but out of curiosity.

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Sun May 21, 2017 5:24 am

Random Walker wrote:I once heard a tenuous Risk explanation that more profitable companies are most subject to new competition.
Theoretical wrote:The best positive explanation (for it being a risk factor) I've heard is that it relates to competitive advantages that can be targeted for reduction or elimination by rival companies. For example, having THE top technology and the execution to the marketplace for it or an unusually good management culture. This "target on the back" premium is to me a bit weak.

What I think the stronger argument is is that the short side does better than it should for long periods, given its junkiness. So it shows up as a risk in bull markets that the quality stocks will underperform. The outperformance is that the junk stocks end up being somewhat to grossly overvalued in times of distress while the quality stocks don't suffer as much.
From your full posts that I snipped, it sounds like you're not convinced either! My thoughts on the snips:

If competition comes and your profitability drops to average, that hurts, but now you're average, not dead. It should be priced into the stock, but not make for better than market returns.

Also, underperforming in a bull isn't so bad. I feel that a stock which does well in bad times and poorly in good times should require less return than the market, not more.

Theoretical
Posts: 1097
Joined: Tue Aug 19, 2014 10:09 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by Theoretical » Sun May 21, 2017 12:29 pm

Quite right. I think if anything there is a premium to quality, but it's a premium in the insurance sense, not the positive sense.

For what it's worth, the quality-screened Bridgeway microcap fund got hammered on the way down in 2008, and the recovery was most centered on junky microcaps.

With bonds, I think it's a bit different story because it's a fixed obligation that starts at one quality and is now at a different quality, and where artificial market constraints independent of price adjust/affect the market pricing. Bond A that was issued at A and is now a BB trading at 70 is a vastly different (and better) security than a bond issued at par at BB. But institutional investors, pensions, and the like can't invest in either, leading to the fallen angel effect Vanguard's High yield fund and ANGL exploit.

lazyday
Posts: 3015
Joined: Wed Mar 14, 2007 10:27 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by lazyday » Sun May 21, 2017 2:17 pm

Theoretical wrote:I think if anything there is a premium to quality, but it's a premium in the insurance sense, not the positive sense.
That's what would make sense to me in the long run. I'm a bit confused why Larry considers there to be an intuitive reason for a premium for profitability and quality.
Larry wrote:Intuitive: There are logical risk-based or behavioral-based explanations for its premium and why it should continue to exist.
Today, if we take a simplistic view, Quality still seems to be priced for us to collect a premium instead of paying one.
For what it's worth, the quality-screened Bridgeway microcap fund got hammered on the way down in 2008, and the recovery was most centered on junky microcaps.
I read the BRSIX shareholder letters at that time, and have the impression that the screens were expanded around the time of the financial crisis. Which could explain hurting worse on the way up than helping on the way down. But I might not remember correctly.

gtwhitegold
Posts: 200
Joined: Fri Sep 21, 2012 1:55 pm

Re: Larry Swedroe: Problems With The Factor Zoo

Post by gtwhitegold » Mon May 22, 2017 9:48 am

The thing with quality is that alone it performs better than the market over long periods of time, but it's even better combined with value. Putting the two together gets rid of junk value stocks and overvalued quality stocks leaving you with equities somewhat like a Buffet portfolio, "quality at a reasonable price."

Post Reply