New Domestic Small Cap Value Option PXSV

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

New Domestic Small Cap Value Option PXSV

Post by grap0013 » Wed Feb 22, 2012 8:03 pm

I haven't seen a thread on this one yet. Check out this link about the new RAFI Fundamental US based indexes: http://www.researchaffiliates.com/rafi/ ... _Paper.pdf

It looks like PXSV is smaller and more valuey than Russell SCV. Probably the same story against S&P small cap value. It has about 750 stocks. ER 0.39. I'm guessing a size load of 0.7 and a value load of 0.9? Seems like a much "safer" option than RZV. I was trying to find yearly returns for the index but I've had no luck. Should be interesting once we get some more time on it. I exchanged my PRFZ in ROTH to this new one yesterday.

Anybody changing to this one besides me? Are you waiting to see if there's tracking error?
There are no guarantees, only probabilities.

User avatar
Sammy_M
Posts: 1866
Joined: Sun Nov 25, 2007 8:30 am

Re: New Domestic Small Cap Value Option PXSV

Post by Sammy_M » Wed Feb 22, 2012 8:11 pm

This does look like a great alternative to RZV. It's more diversified than RZV, and I'd imagine the RAFI methodology will achieve less negative alpha. Total assets and average volume are still a little light right now (similar to RZV), but I'll probably use it at some point (in tax advantaged).

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Wed Feb 22, 2012 10:04 pm

I ran a Fama French Factor Loading Analysis on PXSV aawhile back.
I got the following results.
ER: 0.39
Turnover: 0.90
Alpha: -0.004 (monthly)
Beta = 0.0897
SMB = 0.708
HML = 0.182
Momemtum (UMD) = -0.044

Projected Expected Real Return (assuming alpha = ER - Turnover) = 6.2%

My feeling was the turnover was too high, the Value Loading too low, and the low alpha was a bit of a red flag too.

VIOV/IJS has SMB = 0.608, HML = 0.605
VBR has SMB = 0.424, HML = 0.673

Both have better ER & Turnover too. I think those are optimal over PXSV.

grok87
Posts: 7595
Joined: Tue Feb 27, 2007 9:00 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grok87 » Wed Feb 22, 2012 10:20 pm

pauliec84 wrote:I ran a Fama French Factor Loading Analysis on PXSV aawhile back.
I got the following results.
ER: 0.39
Turnover: 0.90
Alpha: -0.004 (monthly)
Beta = 0.0897
SMB = 0.708
HML = 0.182
Momemtum (UMD) = -0.044

Projected Expected Real Return (assuming alpha = ER - Turnover) = 6.2%

My feeling was the turnover was too high, the Value Loading too low, and the low alpha was a bit of a red flag too.

VIOV/IJS has SMB = 0.608, HML = 0.605
VBR has SMB = 0.424, HML = 0.673

Both have better ER & Turnover too. I think those are optimal over PXSV.

thanks
"...people always live for ever when there is any annuity to be paid them"- Jane Austen

vesalius
Posts: 778
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: New Domestic Small Cap Value Option PXSV

Post by vesalius » Wed Feb 22, 2012 10:33 pm

Yes made the switch about a month ago from SFSNX. In June of 2011 Powershares switched from the Dynamic Small Cap Value Intellidex Index, which this fund previously followed, to the RAFI Fundamental Small Value Index. Not sure how or even if this recent wholesale switch might have affected the FF analysis

Park
Posts: 467
Joined: Sat Nov 06, 2010 4:56 pm

Re: New Domestic Small Cap Value Option PXSV

Post by Park » Wed Feb 22, 2012 10:54 pm

I only tilt to SCV in my tax advantaged space. Unfortunately, that is a bit less than 1/7 of my entire portfolio. Because of the comparatively small size, I have RZV, due to its strong tilt. I realize that when it comes to momentum, RZV is not strong, but it's an overall compromise.

Should I switch to PXSV?

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: New Domestic Small Cap Value Option PXSV

Post by yobria » Wed Feb 22, 2012 11:10 pm

Thanks for the link. For us small growth investors, PXSG looks promising, if a little expensive.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Wed Feb 22, 2012 11:38 pm

pauliec84 wrote:I ran a Fama French Factor Loading Analysis on PXSV aawhile back.
I got the following results.
ER: 0.39
Turnover: 0.90
Alpha: -0.004 (monthly)
Beta = 0.0897
SMB = 0.708
HML = 0.182
Momemtum (UMD) = -0.044

Projected Expected Real Return (assuming alpha = ER - Turnover) = 6.2%

My feeling was the turnover was too high, the Value Loading too low, and the low alpha was a bit of a red flag too.

VIOV/IJS has SMB = 0.608, HML = 0.605
VBR has SMB = 0.424, HML = 0.673

Both have better ER & Turnover too. I think those are optimal over PXSV.


Your number crunching is appreciated! :D Did you run the numbers on the index or the fund itself? What time frame? PXSV has only been around for about 6 months. Prior to 06/16/11 it was a much different fund. That HML does not jive with the link I posted and that turnover is likely from the old fund? I think we may have to wait for the new data before we can draw conclusions? Can you run the numbers from June 2011 thru Jan. 2012?
Last edited by grap0013 on Wed Feb 22, 2012 11:47 pm, edited 3 times in total.
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Wed Feb 22, 2012 11:44 pm

Park wrote:I only tilt to SCV in my tax advantaged space. Unfortunately, that is a bit less than 1/7 of my entire portfolio. Because of the comparatively small size, I have RZV, due to its strong tilt. I realize that when it comes to momentum, RZV is not strong, but it's an overall compromise.

Should I switch to PXSV?


FWIW, I'm drinking the fundamental Kool-Aid and I ran some backtests a while back from about 1992-2007 and PRFZ got the annualized return as RZV with much less volatility. Since PXSV is smaller and more valuey than PRFZ I expect it to perform even better. Only time will tell.

For some reason fundamental indexes appear to get some of their returns from something not explained by factor loadings. I am not sure what, but they have done a good job of avoiding negative momentum that hinders some other value funds.
There are no guarantees, only probabilities.

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Wed Feb 22, 2012 11:56 pm

Your number crunching is appreciated! Did you run the numbers on the index or the fund itself? What time frame? PXSV has only been around for about 6 months. Prior to 06/16/11 it was a much different fund. That HML does not jive with the link I posted and that turnover is likely from the old fund? I think we may have to wait for the new data before we can draw conclusions? Can you run the numbers from June 2011 thru Jan. 2012?


I crunched these numbers while ago. I think it was from the fund not the index. It may not be the best. If you know where the index returns are available I am happy (and curious) to run those through. Always on the look out for new funds, now only if there was an international small cap value fund!

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Thu Feb 23, 2012 12:05 am

pauliec84 wrote:
Your number crunching is appreciated! Did you run the numbers on the index or the fund itself? What time frame? PXSV has only been around for about 6 months. Prior to 06/16/11 it was a much different fund. That HML does not jive with the link I posted and that turnover is likely from the old fund? I think we may have to wait for the new data before we can draw conclusions? Can you run the numbers from June 2011 thru Jan. 2012?


I crunched these numbers while ago. I think it was from the fund not the index. It may not be the best. If you know where the index returns are available I am happy (and curious) to run those through. Always on the look out for new funds, now only if there was an international small cap value fund!


I've just done a quick few google searches and have come up empty handed.
There are no guarantees, only probabilities.

User avatar
empb
Posts: 780
Joined: Tue Aug 11, 2009 1:22 pm
Location: E1W

Re: New Domestic Small Cap Value Option PXSV

Post by empb » Thu Feb 23, 2012 7:24 am

pauliec84 wrote:I ran a Fama French Factor Loading Analysis on PXSV aawhile back.
Beta = 0.0897

Am I missing something?

User avatar
RaleighStClaire
Posts: 731
Joined: Mon Jun 18, 2007 2:08 pm
Location: Tilting

Re: New Domestic Small Cap Value Option PXSV

Post by RaleighStClaire » Thu Feb 23, 2012 9:01 am

Having done zero investigation I'm just going to ask, what's the big difference between PXSV and SFSNX? Doesn't SFSNX follow the same RAFI SV index or is it a mid-small index that SFSNX follows?

Also, those factor loads are definitely incorrect. All of the RAFI funds get around a .4 or more on the HML loading.

New SV options are also interesting, though. For now I'll stick with IJS until someone can mimic DFSVX better.
Where's that red one gonna go?

User avatar
Sammy_M
Posts: 1866
Joined: Sun Nov 25, 2007 8:30 am

Re: New Domestic Small Cap Value Option PXSV

Post by Sammy_M » Thu Feb 23, 2012 9:46 am

RaleighStClaire wrote:Having done zero investigation I'm just going to ask, what's the big difference between PXSV and SFSNX? Doesn't SFSNX follow the same RAFI SV index or is it a mid-small index that SFSNX follows?

Also, those factor loads are definitely incorrect. All of the RAFI funds get around a .4 or more on the HML loading.

New SV options are also interesting, though. For now I'll stick with IJS until someone can mimic DFSVX better.

As I understand it,

SFSNX follows the FTSE RAFI US Mid-Small 1500 Index, which doesn't exclude small-mid growth, the RAFI methodology just weights them at a lower level

PXSV now follows the RAFI Fundamental Small Value Index, which slices the above to only include style specific holdings, and then weights them according to RAFI methodology. The morningstar box shows it has having deep small and value.

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Thu Feb 23, 2012 11:09 am

I ran a Fama French Factor Loading Analysis on PXSV aawhile back.
Beta = 0.0897

Am I missing something?


Typo. Should be 0.897

User avatar
Rick Ferri
Posts: 8203
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Rick Ferri » Thu Feb 23, 2012 11:19 am

I looked at PXSV some time ago and posted the information this site. While it is one to watch, there are concerns:

1) The amount of money in the fund is less than $50 million. Trading volume is light.
2) The prospectus has the fee at 0.80% with a fee wave until August 2012 down to 0.39%
3) PowerShares has a history of changing indexes and closing funds if they don't attract enough money.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

ftobin
Posts: 1000
Joined: Fri Mar 20, 2009 3:28 pm

Re: New Domestic Small Cap Value Option PXSV

Post by ftobin » Thu Feb 23, 2012 1:48 pm

Rick Ferri wrote:1) The amount of money in the fund is less than $50 million. Trading volume is light.
2) The prospectus has the fee at 0.80% with a fee wave until August 2012 down to 0.39%
3) PowerShares has a history of changing indexes and closing funds if they don't attract enough money.

I'll note that the ETF at least has a reasonable spread (0.6%, fair for a long-term holding), with a good, with $50-100K of capital at inside. What really matters with ETFs is if market makers have programmed their models to keep tight arb limits on the ETF, not how much volume transacts (though volume, of course, attracts more, tighter arb competition).

caklim00
Posts: 1707
Joined: Mon May 26, 2008 10:09 am

Re: New Domestic Small Cap Value Option PXSV

Post by caklim00 » Thu Feb 23, 2012 2:11 pm

RZV is stiller much smaller (mkt cap is half PXSV) and more valuey (PB is under 1) and has a lower dividend yield than PXSV (so in theory more tax efficient). That said, PXSV is very interesting. Negative momentum is one of my bigger fears about holding RZV. But, I'm not jumping off the RZV train for this and don't think I'll even use this in place of VBR in tax deferred accounts.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Thu Feb 23, 2012 2:20 pm

Rick Ferri wrote:I looked at PXSV some time ago and posted the information this site. While it is one to watch, there are concerns:

1) The amount of money in the fund is less than $50 million. Trading volume is light.
2) The prospectus has the fee at 0.80% with a fee wave until August 2012 down to 0.39%
3) PowerShares has a history of changing indexes and closing funds if they don't attract enough money.

Rick Ferri


It looks pretty safe in tax deferred accounts. I will hold off on purchasing in taxable much like other posters. When I bought some in Roth IRA this past Tuesday I ate about a 0.1% spread on the sell side of PRFZ and about 0.25% spread on the buy side of PXSV. I make 1 trade per year in that account. I think it's worth it.

Rick,

Do you have or could you please obtain, year over year returns for the fundamental pure small cap value index? It would be fun running some numbers on it. Thanks in advance if you can do it!
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Thu Feb 23, 2012 2:26 pm

caklim00 wrote:RZV is stiller much smaller (mkt cap is half PXSV) and more valuey (PB is under 1) and has a lower dividend yield than PXSV (so in theory more tax efficient). That said, PXSV is very interesting. Negative momentum is one of my bigger fears about holding RZV. But, I'm not jumping off the RZV train for this and don't think I'll even use this in place of VBR in tax deferred accounts.


Yes negative momentum is definitely a concern with RZV. Plus I don't like the small number of holdings. I'm all for taking some risk, but I would be uncomfortable hoping for a large outperform of 10-15 stocks that could drive most of the excess returns of small cap value stocks in RZV. You might overshoot, but you have a good chance of undershooting as well. If you're gonna ride that roller coster you might as well increase your chances of getting compensated for puking by holding more stocks.
There are no guarantees, only probabilities.

User avatar
Rick Ferri
Posts: 8203
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Rick Ferri » Thu Feb 23, 2012 2:41 pm

grap0013 wrote:Rick, do you have or could you please obtain, year over year returns for the fundamental pure small cap value index? It would be fun running some numbers on it. Thanks in advance if you can do it!


I do, somewhere. Got them from Research Affiliates. I'll have to see if I can dig them up.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Thu Feb 23, 2012 2:43 pm

If you can get your hands on them (need month over month), I would be happy to run some fama french regressions on them to figure out the factor loads.

User avatar
Rick Ferri
Posts: 8203
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Rick Ferri » Thu Feb 23, 2012 2:53 pm

Here is the RAFI Pure Small Value index data from Reseach Affiliates through 9.30.2011

RAFI Pure Small Value Index data

I have already run 3-factor regressions on the data. The index shows very good small and value loadings, and it is well diversified with over 700 names. This is the best I have seen so far from a US small value index. It's the ETF issues noted above that are holding me back.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Thu Feb 23, 2012 3:04 pm

Thanks for Link

The RAFI SV
MKTRF: 0.994
SMB: 0.862
HML 0.669
UMD (Momentum): -0.179

After accounting for the ER and Turnover I have an expected real retrun of 8.2%.

Not Bad but...

That is the same as VBR and less than VIOV/IJS. So you are taking more risk on in the factor loadings for the same expected return. Not worth it in my opinion.

The RAFI SV does have a positive alpha of 0.002 (monthly) over the time period. However I never expect alphas to persist, especially in backward looking indexes, as
a) There is no drag of the transaction costs & ER
b) The index was probably constructed with getting the maxmium alpha in mind.

User avatar
Rick Ferri
Posts: 8203
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Rick Ferri » Thu Feb 23, 2012 3:15 pm

I agree, 8% real return expected, not bad, BUT....

In addition to what you wrote, there is the ETF liquidity issue (for large trades, which I do) and the potential jump in fees after the waiver expires in August 2012.

Definitely one to keep in mind, though.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

User avatar
RaleighStClaire
Posts: 731
Joined: Mon Jun 18, 2007 2:08 pm
Location: Tilting

Re: New Domestic Small Cap Value Option PXSV

Post by RaleighStClaire » Thu Feb 23, 2012 10:12 pm

Someone needs to tell Schwab that this index is available so that they can just make an open-ended fund version of this index for .35 and we all can jump in. I know I would. This index seems a lot more appealing than the current mid-small 1500 index they use for SFSNX.
Where's that red one gonna go?

User avatar
Sammy_M
Posts: 1866
Joined: Sun Nov 25, 2007 8:30 am

Re: New Domestic Small Cap Value Option PXSV

Post by Sammy_M » Fri Feb 24, 2012 8:35 am

Rick Ferri wrote:Here is the RAFI Pure Small Value index data from Reseach Affiliates through 9.30.2011

RAFI Pure Small Value Index data

Thank you Rick. And thanks pauliec84 for sharing the regressions. I'm on target with my desired small/value exposure, but when I need to pick up more I will be using this within Roth. That way if the fee jumps, I can jump. And if they shutdown, no worries, I'll move to something else. I'll plan to stick with IJS in taxable.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Fri Feb 24, 2012 9:29 am

Sammy_M wrote:
Rick Ferri wrote:Here is the RAFI Pure Small Value index data from Reseach Affiliates through 9.30.2011

RAFI Pure Small Value Index data

Thank you Rick. And thanks pauliec84 for sharing the regressions. I'm on target with my desired small/value exposure, but when I need to pick up more I will be using this within Roth. That way if the fee jumps, I can jump. And if they shutdown, no worries, I'll move to something else. I'll plan to stick with IJS in taxable.


+1
Likewise. Thanks Rick and pauliec84!

Pauliec84,

Would you mind posting the year by year returns for this new index? I tried last night on excel to convert the monthly data into annualized returns and I'm honestly very bad at it. I'm getting error messages. You are clearly very good with numbers and your expertise would be appreciated! :D Thanks!!
There are no guarantees, only probabilities.

pauliec84
Posts: 411
Joined: Wed Feb 09, 2011 11:29 am

Re: New Domestic Small Cap Value Option PXSV

Post by pauliec84 » Fri Feb 24, 2012 9:33 am

grap0013, it is a real pain in the neck to convert monthly to early in excel, no easy way to do it to my knowledge.

User avatar
MossySF
Posts: 2208
Joined: Thu Apr 19, 2007 9:51 pm
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by MossySF » Fri Feb 24, 2012 10:33 am

I don't know what the monthly data looks like but here's excel techniques based on 2 possible data sources.

If the data is ending month price, easiest method is to ignore the months in between and only look at begin price to year price.

Code: Select all

     A       B     C
1    12/2010
2    01/2011 43.67
3    02/2011 42.88
...
12   11/2011 44.71
13   12/2011 44.56 =(B13/B1)-1


If the data is monthly percentage change, we have to multiply every month together.

Code: Select all

     A      B             C
1           1
2    +0.2%  =B1*(1+A2)
3    -0.1%  =B2*(1+A3)
...
12   +0.8%  =B11*(1+A12)
13   +1.2%  =B12*(1+A13)  =B13-1

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Fri Feb 24, 2012 3:41 pm

pauliec84 wrote:grap0013, it is a real pain in the neck to convert monthly to early in excel, no easy way to do it to my knowledge.


Thanks anyways. I'll cruch 'em out and post them when I get a chance. I like to plug the data into Simba's excel backtest. I really do think there is a little more juice to the fundamental strategy that cannot be explained by FF factor loads. Maybe FF underestimate the value load of Fundamental indexes because they solely use P/B? Fundamental indexes are also cheap across P/E, P/CF, and P/D. Maybe there is some intrinsic rebalancing bonus between the various sectors within the index because it's always buying the most out of favor stocks?
There are no guarantees, only probabilities.

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

Re: New Domestic Small Cap Value Option PXSV

Post by Clearly_Irrational » Fri Feb 24, 2012 6:37 pm

I didn't like the fees and turnover on this fund compared to VBR. RZV actually has higher loadings than PXSV if that's your main objective. I didn't like RZV because it didn't look like the extra loadings were being compensated (in comparison). In the end I stuck with VBR.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Tue Apr 17, 2012 1:45 pm

Here's some updates on PXSV:

http://www.invescopowershares.com/produ ... icker=PXSV

Thru 3/31/12:
YTD NAV: 13.41
Index: 13.53

So the tracking error of the ETF to the index looks minimal. Fundamental U.S. funds have done a good job of tracking their indexes overall. Total assets have now broken the 50 million mark.

I used the monthly returns, a few hours, 2 beers, and calculated out annual returns. I subtracted out 0.5% for the fundamental index funds for fees. Here are the results using my customized Simba backtest document:

1979-2011

VBR (Vanguard fund)
Annualized return: 12.9%
Standard deviation: 18.5

PRFZ (my 2nd favorite domestic SCV fund)
Annualized return: 14.2%
Standard deviation: 21.5

PXSV
Annualized return: 17.3%
Standard deviation: 23.2

A few other thoughts. Risk and return are evident. Even if you run the data from 2000 to 2011 you get about the same spreads in returns, even other time frames for that matter. PXSV:5 year treasuries of 80:20 had the same SD of 100% VBR but with about 3% higher annualized return. Finally the pure small cap value fundamental index is really contratraded as evidenced during its 2000-2002 performance. It got REALLY small and valuey at the time and its AR was 12.6% during those 3 years.

Now I just want to see if Powershares will increase the ER after August 31st. I'd pay up to 0.5%, but no more than that.
There are no guarantees, only probabilities.

User avatar
RaleighStClaire
Posts: 731
Joined: Mon Jun 18, 2007 2:08 pm
Location: Tilting

Re: New Domestic Small Cap Value Option PXSV

Post by RaleighStClaire » Thu Apr 19, 2012 7:38 am

grap, thanks for doing the legwork here.

From the looks of it, PXSV the first legitimate competition to DFSVX that I've come across. Really hard to argue with that data. More risk, but more compensated return. I like it.

I really hope Schwab picks this index up for their line of fundamental index funds. C'mon Schwab!
Where's that red one gonna go?

Jerry_lee
Posts: 915
Joined: Tue Mar 06, 2012 1:55 pm

Re: New Domestic Small Cap Value Option PXSV

Post by Jerry_lee » Thu Apr 19, 2012 12:06 pm

I skimmed through this thread and a number of red flags pop up:

1. this constant tweaking of an asset class portfolio will almost certainly result in sub-par results. There are plenty of fine retail SV index options on the market (IJS, VBR, and IWN being 3 of them), the search for the magic ETF should be immediately dismissed. If, 5-7 years from now (which should be the MINIMUM amount of live data to gauge portfolio attributes and implementation), something looks superior to the three core options above, then the discussion can begin. But definitely not before then.

2. running regressions on backtested data is no different than rearranging deck chairs on the Titanic. There will never be any shortage of products with sexy backtested results. They are as close to meaningless as anything I can think of. How long do you think the analysts and # crunchers sit around reconfiguring the data before they actually release the "results". If you had a chance to see all the many derivations of backtested returns that weren't published, you'd be far more cautious

3. what matters most is your overall asset allocation. If you are using a fund like Vanguard SV, with less of a size tilt than others, just hold a bit more of it. For example, where I might have my US allocation split 30% S&P 500, 30% Russell 1000 Value, and 40% Russell 2000 Value, if I were using Vanguard I might go 30% S&P 500, 25% MSCI 750 Value, 45% MSCI 1750 Value. Maybe +5% is the right overweight, maybe +10%. Whatever you decide, stay with it and stop trying to perfect an already excellent mix.

4. duplicating DFAs approach to targeting small and/or value premiums, or managing small value funds, isn't purely about holding the smallest stocks or most value oriented companies. DFAs process is more important than their portfolio characteristics, something that product focused investors and advisors never seem to understand. Examples include:
a) trading patiently and using cash flows to rebalance.
b) not plowing into negative momentum or fighting positive momentum
c) screening out stocks that may meet arbitrary valuation criteria but haven't historically delivered the return premia (utilities for LV and REITs for SV)
d) avoiding excessive data-mining on valuations sorts, instead opting for common sense. Apparently the latest marketing spin in the indexing world is to sort companies on multiple factors (p/e, p/cf, p/s, and p/b) believing that more "sorting" always yields to superior results. What we know is, more sorts = more turnover. And more turnover = higher costs. Whether a 4 factor sort will outpace a 1 factor sort over the next 20 years on an after tax, net of fee basis is highly questionable. Especially when the 1 factor is book value, the most stable fundamental metric of all, yielding the lowest possible turnover. One thing's for sure, for investors and advisors with an insatiable appetite for tweaking portfolios, multi factor sorts are the latest crack cocaine.
e) limiting access to clients of RIAs who are supposed to (but don't always) preach discipline and adherence to their plans to help screen out the hot money that increases costs and lowers returns and increases tax consequences. I can't imagine how much higher the turnover on DFSVX or DTMVX would be if they were available to retail investors who would be sure to jump in after a 5 year "run", only to jump out after some new ETF with better backtested results emerges.

Does any of this remotely sound like the constant merry-go-round of ETF products?

Remember, the benefits of asset class investing and embracing the 3F model is, over long periods of time, if held consistently, you will get enhanced diversification benefits and higher expected returns than a Total Market mix. Constantly tweaking your allocation and searching for something better will almost always result in less.

When I hear of advisors who are hired to lend structure, discipline, and consistency to the investment process, but are instead constantly tweaking portfolios with commodities, momentum strategies, junk bonds, or unproven ETFs, I cringe. If you are an individual DIY investor, you would be wise to learn from their mistakes.
The most disciplined investor in the world.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: New Domestic Small Cap Value Option PXSV

Post by yobria » Thu Apr 19, 2012 12:12 pm

FYI this is at least the third fund I've seen in the past six months chasing that small pool of extreme small/value stocks. Bubble watchers take note.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Thu Apr 19, 2012 1:25 pm

RaleighStClaire wrote:grap, thanks for doing the legwork here.

From the looks of it, PXSV the first legitimate competition to DFSVX that I've come across. Really hard to argue with that data. More risk, but more compensated return. I like it.

I really hope Schwab picks this index up for their line of fundamental index funds. C'mon Schwab!


You're welcome! I've noticed it's been a lot harder to find fundamental index funds on Schwab's website. Maybe they are not advertising them a lot because they are not big money makers? Or losing interest in them? I hope they pick it up as well!
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Thu Apr 19, 2012 1:32 pm

Jerry_lee wrote:I skimmed through this thread and a number of red flags pop up:

1. this constant tweaking of an asset class portfolio will almost certainly result in sub-par results. There are plenty of fine retail SV index options on the market (IJS, VBR, and IWN being 3 of them), the search for the magic ETF should be immediately dismissed. If, 5-7 years from now (which should be the MINIMUM amount of live data to gauge portfolio attributes and implementation), something looks superior to the three core options above, then the discussion can begin. But definitely not before then.

2. running regressions on backtested data is no different than rearranging deck chairs on the Titanic. There will never be any shortage of products with sexy backtested results. They are as close to meaningless as anything I can think of. How long do you think the analysts and # crunchers sit around reconfiguring the data before they actually release the "results". If you had a chance to see all the many derivations of backtested returns that weren't published, you'd be far more cautious

3. what matters most is your overall asset allocation. If you are using a fund like Vanguard SV, with less of a size tilt than others, just hold a bit more of it. For example, where I might have my US allocation split 30% S&P 500, 30% Russell 1000 Value, and 40% Russell 2000 Value, if I were using Vanguard I might go 30% S&P 500, 25% MSCI 750 Value, 45% MSCI 1750 Value. Maybe +5% is the right overweight, maybe +10%. Whatever you decide, stay with it and stop trying to perfect an already excellent mix.

4. duplicating DFAs approach to targeting small and/or value premiums, or managing small value funds, isn't purely about holding the smallest stocks or most value oriented companies. DFAs process is more important than their portfolio characteristics, something that product focused investors and advisors never seem to understand. Examples include:
a) trading patiently and using cash flows to rebalance.
b) not plowing into negative momentum or fighting positive momentum
c) screening out stocks that may meet arbitrary valuation criteria but haven't historically delivered the return premia (utilities for LV and REITs for SV)
d) avoiding excessive data-mining on valuations sorts, instead opting for common sense. Apparently the latest marketing spin in the indexing world is to sort companies on multiple factors (p/e, p/cf, p/s, and p/b) believing that more "sorting" always yields to superior results. What we know is, more sorts = more turnover. And more turnover = higher costs. Whether a 4 factor sort will outpace a 1 factor sort over the next 20 years on an after tax, net of fee basis is highly questionable. Especially when the 1 factor is book value, the most stable fundamental metric of all, yielding the lowest possible turnover. One thing's for sure, for investors and advisors with an insatiable appetite for tweaking portfolios, multi factor sorts are the latest crack cocaine.
e) limiting access to clients of RIAs who are supposed to (but don't always) preach discipline and adherence to their plans to help screen out the hot money that increases costs and lowers returns and increases tax consequences. I can't imagine how much higher the turnover on DFSVX or DTMVX would be if they were available to retail investors who would be sure to jump in after a 5 year "run", only to jump out after some new ETF with better backtested results emerges.

Does any of this remotely sound like the constant merry-go-round of ETF products?

Remember, the benefits of asset class investing and embracing the 3F model is, over long periods of time, if held consistently, you will get enhanced diversification benefits and higher expected returns than a Total Market mix. Constantly tweaking your allocation and searching for something better will almost always result in less.

When I hear of advisors who are hired to lend structure, discipline, and consistency to the investment process, but are instead constantly tweaking portfolios with commodities, momentum strategies, junk bonds, or unproven ETFs, I cringe. If you are an individual DIY investor, you would be wise to learn from their mistakes.


Skimming is your first problem! :wink:
I've already said everything I want to about Fundamental Indexes, but I can't find the thread via google. Please read, "The Fundamental Index" by Arnott and come back on here and tell me if you feel the same way. I am not constantly tweaking. I invest in the ETFs/Mutual funds that I think are best for capturing small cap value premiums and when better ones come out, I adjust. I exchanged about 10% of my portfolio from PRFZ to PXSV. I don't think I did any major damage or performance chasing.
There are no guarantees, only probabilities.

User avatar
RaleighStClaire
Posts: 731
Joined: Mon Jun 18, 2007 2:08 pm
Location: Tilting

Re: New Domestic Small Cap Value Option PXSV

Post by RaleighStClaire » Wed Apr 25, 2012 9:07 am

grap: not a good sign :(
Where's that red one gonna go?

User avatar
RaleighStClaire
Posts: 731
Joined: Mon Jun 18, 2007 2:08 pm
Location: Tilting

Re: New Domestic Small Cap Value Option PXSV

Post by RaleighStClaire » Mon Aug 06, 2012 1:53 pm

Since I just ran the numbers I felt it necessary to post them in case anyone was interested.

Since the strategy change in June of 2011 here are the factor weights for PXSV, DFSVX and IJS for comparison

Code: Select all

             a        Mkt      SMB         HML
PXSV        0.49     1.05      0.88       0.78
DFSVX       0.07     1.08      0.96       0.28
IJS         0.44     0.98      0.90       0.05


So, not sure how relevant 13 months of data is -- my guess is not very. But there you have it. I still watch PXSV on a semi-regular basis since I like the RAFI stuff and this index looks great on paper. We'll see how it does long term.
Where's that red one gonna go?

User avatar
Sammy_M
Posts: 1866
Joined: Sun Nov 25, 2007 8:30 am

Re: New Domestic Small Cap Value Option PXSV

Post by Sammy_M » Mon Aug 06, 2012 3:28 pm

Thanks for sharing.

User avatar
grap0013
Posts: 1879
Joined: Thu Mar 18, 2010 1:24 pm

Re: New Domestic Small Cap Value Option PXSV

Post by grap0013 » Wed Aug 08, 2012 2:58 pm

RaleighStClaire wrote:Since I just ran the numbers I felt it necessary to post them in case anyone was interested.

Since the strategy change in June of 2011 here are the factor weights for PXSV, DFSVX and IJS for comparison

Code: Select all

             a        Mkt      SMB         HML
PXSV        0.49     1.05      0.88       0.78
DFSVX       0.07     1.08      0.96       0.28
IJS         0.44     0.98      0.90       0.05


So, not sure how relevant 13 months of data is -- my guess is not very. But there you have it. I still watch PXSV on a semi-regular basis since I like the RAFI stuff and this index looks great on paper. We'll see how it does long term.


Thanks! Your work is greatly appreciated!

FYI, Powershares extended the fee waiver for the expense ratio until August 2013, so it's still at 0.39 for at least another year. Now we just have to hope the AUM is not too low that the ETF closes. It looks like is borderline profitable right now still hovering around an AUM of 50 million.
There are no guarantees, only probabilities.

User avatar
Rick Ferri
Posts: 8203
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Rick Ferri » Sun Aug 19, 2012 10:06 pm

Remember that HmL in the above analysis is using only one value factor - book to market. There are many other ways to measure value.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: New Domestic Small Cap Value Option PXSV

Post by yobria » Mon Aug 20, 2012 2:03 am

There's something about the names of these Powershares ETFs:

Powershares Active Alpha Q
PowerShares DWA Technical Leaders Portfolio
PowerShares Fundamental Pure Small Value
Etc.

That might make a skeptical investor wary of the company in particular, and such market beating strategies generally.

User avatar
Robert T
Posts: 2456
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: New Domestic Small Cap Value Option PXSV

Post by Robert T » Wed Jul 31, 2013 6:40 am

.
Thanks Rick for the earlier data.

Here's a bit more analysis. The double sort of the RAFI Fundamental SV series seems to pick up some of the profitability premium (t-stats in parentheses) in addition to size and value. Using Novy-Mark PMU series from his website.

Code: Select all

4/30/1993–9/30/2011 

                     Alpha(t-stat)  Mkt   Size  Value Profitability R^2

DFA US Small Value   -0.07 (-0.93)  1.03  0.83  0.67    0.06 (1.97) 0.96     
RAFI US 1500          0.01  (0.14)  1.08  0.72  0.55    0.11 (2.55) 0.94
RAFI Fundamental SV   0.05  (0.37)  1.05  0.81  0.87    0.22 (3.61) 0.89


Code: Select all

4/30/1993–9/30/2011

Annualized returns

DFA US Small Value    10.4%
RAFI US 1500          11.4%
RAFI Fundamental SV   13.6%


Code: Select all

2/28/1979 – 9/30/2011

                     Alpha(t-stat)  Mkt   Size  Value Profitability R^2

RAFI US 1500         -0.06 (-0.82)  1.07  0.76  0.45    0.16 (5.33) 0.95
RAFI Fundamental SV  -0.04 (-0.41)  1.03  0.85  0.78    0.24 (5.90) 0.90


Robert
.

garlandwhizzer
Posts: 1671
Joined: Fri Aug 06, 2010 3:42 pm

Re: New Domestic Small Cap Value Option PXSV

Post by garlandwhizzer » Wed Jul 31, 2013 11:32 am

Sorry to bring this up again but I'm skeptical by nature.

1) Everyone in the professional investment world knows about these premiums (small, value, momentum, profitability, etc.,).
2) The vast majority of all investment assets are controlled by professionals with the above knowledge.
3) Why is it that the available alpha of these premiums don't get arbitraged away?

Explanations exist, usually relating to human psychology, but it is not at all clear to me that past investing behavior can't change and that we can predict future excess returns based exactly on current factor loads of value, size, etc.. As a contrarian I suspect that since Mid caps are largely neglected in the quest for alpha discussion (Mel's unloved mid-caps), their valuations probably reflect that neglect and therefore they may well outperform everything going forward.

I don't know, but there is one thing I do feel certain about. TSM should be the lion's share of my US equity portfolio and variations from it should be limited and modest because predicting the future of markets is always uncertain.

Garland Whizzer

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

Re: New Domestic Small Cap Value Option PXSV

Post by Clearly_Irrational » Wed Jul 31, 2013 11:46 am

garlandwhizzer wrote:3) Why is it that the available alpha of these premiums don't get arbitraged away?


If it could be arbitraged away it wouldn't be a risk factor. That's like asking why the equity premium doesn't get arbitraged away because bonds exist.

User avatar
G-Money
Posts: 2852
Joined: Sun Dec 09, 2007 7:12 am

Re: New Domestic Small Cap Value Option PXSV

Post by G-Money » Wed Jul 31, 2013 12:28 pm

Clearly_Irrational wrote:That's like asking why the equity premium doesn't get arbitraged away because bonds exist.

Right. I'm yet to encounter someone who doesn't believe in the equity risk premium. Most people also recognize the fixed income premia of term risk and credit risk. No one suggests that these premia can be arb'd away.

This really just boils down to whether an individual believes other risk premia exist or not.
Don't assume I know what I'm talking about.

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

Re: New Domestic Small Cap Value Option PXSV

Post by Random Walker » Wed Jul 31, 2013 1:19 pm

I've often wondered how others cannot believe there would be risk premia for small and value. Who would a bank rather lend money to, a big successful growing company or a struggling leveraged company in an unpopular industry. Would a bank rather lend to McDonalds or SAMs burger shack? Makes sense to me that the cost of capital is the expected return for an investment.

Dave

garlandwhizzer
Posts: 1671
Joined: Fri Aug 06, 2010 3:42 pm

Re: New Domestic Small Cap Value Option PXSV

Post by garlandwhizzer » Wed Jul 31, 2013 2:18 pm

Clearly_Irrational wrote
Clearly_Irrational » Wed Jul 31, 2013 12:46 pm

garlandwhizzer wrote:
3) Why is it that the available alpha of these premiums don't get arbitraged away?


If it could be arbitraged away it wouldn't be a risk factor. That's like asking why the equity premium doesn't get arbitraged away because bonds exist.


Clearly there is risk/reward tradeoff in all markets, the riskier the asset the greater the expected long term return. Everyone agrees that the increased SCV return comes at the price of more volatility and that for long periods of time it disappears. The bigger question is: can the SCV premium be captured in after-tax-dollars by real time funds with real expense ratios, real trading costs, and real taxes generated? There is no question that academic research based on no cost, no tax consequences indexes demonstrates the SCV premium over significant periods of time. Unfortunately investors get only what their SCV funds deliver in real dollars and that often falls short of the research. For example, since inception in 1998, Vanguard SCV index has under-perfomed their SCG index. SCG is according to academic research the black hole of investment dollars, the exact opposite of SCV. But the Vanguard SCV index has an annual return of 10.14 versus SCG at 11.13 over that 15 year period, which seems like a long time to me. So are those who purchased that SCV fund 15 years ago to harvest their outsized gains supposed to continue holding on for another decade or two, reassured that academic research is on their side?

Garland Whizzer

Post Reply