Vanguard New Factor Funds Portfolio Statistics

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Thu Mar 08, 2018 1:14 pm

I was a bit impatient for portfolio statistic on the new factor funds, so I've started calculating them, at least roughly. You can export the portfolio's from Vanguard, tweak the CSV file, and import the portfolio into M*'s portfolio manager. The big limitation is a limit of 200 holdings. So I have to split the portfolios into segments by market cap, pull the statistics, and then do a weighted average.

I've done VFVA (Vanguard US Value Factor ETF) so far.

Code: Select all

Pos#s	Total Value	P/PE	P/B	ROA	ROE	PEG	Yield	Avg Mkt Cap 	LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	$4,174,389.08 	11.43	1.49	5.25%	14.8%	8.44%	2.10%	$26,127MM 	41	19	2	18	3	1	9	3	1
201-400	$1,596,208.67 	11.67	1.29	3.58%	13.99%	9.11%	2.08%	$2,515MM 	4	1	0	19	6	4	48	12	3
401-600	$1,154,778.84 	13.01	1.43	3.33%	5.95%	9.74%	1.79%	$1,780MM 	0	0	0	9	7	0	44	31	7
601-785	$727,221.83 	13.49	1.57	2.56%	21.55%	10.63%	1.49%	$1,270MM	1	0	0	2	1	1	44	44	7
Total	$7,652,598.42 	11.91	1.45	4.36%	13.94%	8.98%	1.99%	$15,166MM 	23	11	1	15	4	1	26	13	3
P/PE = Price / Projected Earnings
PEG = Projected 5-year EPS Growth %


On a first impression, it seems very value-y. It has a much lower forward P/E than any of Vanguard's other value funds, large, mid, or small. P/B is much lower. And it has a *lot* of small and mid cap. They said it would be an all cap fund, and it is.

I'm going to do the multi-factor fund next. It's probably the only other one I care enough about.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Thu Mar 08, 2018 2:03 pm

VFMF (Vanguard US Multifactor ETF)

Code: Select all

Pos#s	Total Value	P/PE	P/B	ROA	ROE	PEG	Yield	Avg Mkt Cap 	LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	$4,865,365.77 	15.61	2.45	6.68%	31.75%	11.38%	1.51%	$26,988MM 	28	14	17	12	10	7	6	3	3
201-400	$1,833,106.22 	17.10	2.29	5.87%	12.90%	10.61%	1.05%	$2,322MM 	1	0	2	4	5	8	24	34	21
401-558	$926,821.00 	14.98	1.99	4.07%	10.23%	12.16%	1.40%	$1,610MM 	1	0	0	2	1	1	34	34	27
Total 	$7,625,292.99 	15.89	2.36	6.17%	24.60%	11.29%	1.39%	$17,973MM 	18	9	11	9	8	7	14	14	10
P/PE = Price / Projected Earnings
PEG = Projected 5-year EPS Growth %

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Thu Mar 08, 2018 2:05 pm

I bought some of the Multifactor Admiral Shares on the first day. I'm planning to stick with it as a core position, but I'll probably take my existing large value and move it to VFVA. It seems to be more value-y than Vanguard's existing CRSP value funds. I may swap a couple of percent to the position from small value, because it includes more small and mid cap value.

Angst
Posts: 1847
Joined: Sat Jun 09, 2007 11:31 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Angst » Thu Mar 08, 2018 3:10 pm

Thank you for posting this, it's a nice preview! I too am interested in both the value and multi-factor ETFs. VFVA's Avg Mrkt Cap is actually comparable to VOE, Vanguard's CRSP Mid-Cap value fund, although VFVA has more than 3 times as many companies and obviously covers a much wider range of cap-sizes, and is clearly more valuey as you say. Here's its style box based on your data:

Code: Select all

   V   B   G
L  23  11  01
M  15  04  01
S  26  13  03

And here it is for VFMF, the multi-factor fund:

Code: Select all

   V   B   G
L  18  09  11
M  09  08  07
S  14  14  10

User avatar
220volt
Posts: 88
Joined: Tue Jan 30, 2018 10:33 pm
Location: USA

Re: Vanguard New Factor Funds Portfolio Statistics

Post by 220volt » Thu Mar 08, 2018 3:33 pm

Bloomberg said the other day that VFVA (Vanguard US Value Factor ETF) usually has 75% stocks in circulation that are not part of the S&P500. If that's true, this would be a very nice diversifier.
"If I had only followed the advice of financial analysts in 2008, I'd have a million dollars today, provided I started with a hundred million dollars" - Jon Stewart

Theoretical
Posts: 1422
Joined: Tue Aug 19, 2014 10:09 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Theoretical » Thu Mar 08, 2018 9:16 pm

In terms of the value fund, it is more valuey and has similar size characteristics to a 50-50 DFA Large Cap Value and DFA Small Cap Value.

I am extremely impressed.

Dominic
Posts: 203
Joined: Sat Jul 02, 2016 11:36 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Dominic » Thu Mar 08, 2018 9:43 pm

Thank you for doing this! It's nice to get some hard data on these new funds. I've been holding out on buying because I wasn't really sure what the portfolios looked like from a factor standpoint. The multifactor fund looks pretty impressive in terms of its valuation and quality characteristics. I'll probably shift some of my holdings into this fund.

Theoretical
Posts: 1422
Joined: Tue Aug 19, 2014 10:09 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Theoretical » Thu Mar 08, 2018 10:26 pm

How'd you calculate the combined style box, @Angst?

User avatar
whodidntante
Posts: 4065
Joined: Thu Jan 21, 2016 11:11 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by whodidntante » Thu Mar 08, 2018 10:30 pm

Let's keep these threads going until the volume improves.

User avatar
grabiner
Advisory Board
Posts: 22728
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Vanguard New Factor Funds Portfolio Statistics

Post by grabiner » Thu Mar 08, 2018 11:13 pm

jhfenton wrote:
Thu Mar 08, 2018 1:14 pm
I've done VFVA (Vanguard US Value Factor ETF) so far.

Code: Select all

Pos#s	Total Value	P/PE	P/B	ROA	ROE	PEG	Yield	Avg Mkt Cap 	LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	$4,174,389.08 	11.43	1.49	5.25%	14.8%	8.44%	2.10%	$26,127MM 	41	19	2	18	3	1	9	3	1
201-400	$1,596,208.67 	11.67	1.29	3.58%	13.99%	9.11%	2.08%	$2,515MM 	4	1	0	19	6	4	48	12	3
401-600	$1,154,778.84 	13.01	1.43	3.33%	5.95%	9.74%	1.79%	$1,780MM 	0	0	0	9	7	0	44	31	7
601-785	$727,221.83 	13.49	1.57	2.56%	21.55%	10.63%	1.49%	$1,270MM	1	0	0	2	1	1	44	44	7
Total	$7,652,598.42 	11.91	1.45	4.36%	13.94%	8.98%	1.99%	$15,166MM 	23	11	1	15	4	1	26	13	3
P/PE = Price / Projected Earnings
PEG = Projected 5-year EPS Growth %


On a first impression, it seems very value-y.
Let's compare this to the standard provider of deep-value ETFs (data from Morningstar; I don't know whether the numbers are actually comparable):

RPV (Guggenheim S&P 500 Pure Value) has a P/PE of 12.02, P/B of 1.30, and average market cap of $20,842M.
RFV (Guggenheim S&P 400 Pure Value) has a P/PE of 11.02, P/B of 0.91, and average market cap of $3006M.
RZV (Guggenheim S&P 600 Pure Value) has a P/PE of 13.54, P/B of 1.01, and average market cap of $701M.

So the value allocation of the large-cap portion of VFVA (top quartile) is close to that of RPV, but the mid-cap and small-cap Guggenheim funds have lower price/book ratios, and the small-cap fund has more micro-caps. VFVA may still have the advantage because it gives almost the same exposure for a significantly lower price.
Wiki David Grabiner

BetaTracker
Posts: 182
Joined: Fri Sep 21, 2012 7:57 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by BetaTracker » Thu Mar 08, 2018 11:39 pm

Wait, aren't these ETFs still building their portfolios? I read somewhere that most of these funds are nowhere near where their portfolio managers expect them to be at a more mature stage? If you're trying to measure degrees of tilts on actively managed funds, aren't things like trade execution and costs still going to be important to determine?
I'd rather be content than happy -- Lao Tzu.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Fri Mar 09, 2018 6:00 am

Theoretical wrote:
Thu Mar 08, 2018 10:26 pm
How'd you calculate the combined style box, @Angst?
I had those numbers in my tables, only horizontally. He did what I should have done and presented them properly.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Fri Mar 09, 2018 10:41 am

M* sector weights for VFVA (Vanguard U.S. Value Factor ETF)

Code: Select all

VFVA			001-200	201-400	401-600	601-785	Overall Rus 1K Val
*Cyclical*		43.49%	48.78%	51.96%	63.92%	47.81%	41.35%
Basic Materials		3.87%	3.60%	7.65%	7.36%	4.72%	3.04%
Consumer Cyclical	10.58%	18.78%	23.62%	17.64%	14.93%	6.48%
Financial Services	28.25%	23.16%	17.95%	36.36%	26.40%	27.47%
Real Estate		0.79%	3.24%	2.73%	2.56%	1.76%	4.36%
							
*Sensitive*		31.87%	33.78%	37.21%	28.99%	32.80%	30.73%
Communication Services	4.64%	0.97%	2.06%	1.09%	3.15%	3.22%
Energy			8.44%	7.89%	9.89%	6.50%	8.36%	10.34%
Industrials		8.80%	17.74%	12.08%	13.89%	11.64%	8.03%
Technology		9.99%	7.17%	13.18%	7.51%	9.65%	9.14%
								
*Defensive*		24.63%	17.44%	10.84%	7.09%	19.38%	27.93%
Consumer Defensive	3.60%	5.27%	2.98%	2.36%	3.74%	8.21%
Healthcare		12.02%	4.85%	3.26%	2.75%	8.32%	14.23%
Utilities		9.01%	7.33%	4.60%	1.98%	7.33%	5.49%
The *top line* of each segment is a subtotal for the M* category (Cyclical, Sensitive, Defensive).

VFVA has fairly typical sector weightings for a value fund. The last column is the Russell 1000 Value Index.

Jeff Albertson
Posts: 554
Joined: Sat Apr 06, 2013 7:11 pm
Location: Springfield

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Jeff Albertson » Fri Mar 09, 2018 10:51 am

Bloomberg does a similar comparison between Vanguard's value etf with the new factor value etf.
https://www.bloomberg.com/gadfly/articl ... ld-save-it

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Fri Mar 09, 2018 11:09 am

M* sector weights for VFMF (Vanguard U.S. Multifactor ETF)

Code: Select all

VFMF			001-200	201-400	401-558	Overall S&P 500
*Cyclical*		40.88%	48.48%	47.89%	43.56%	33.44%
Basic Materials		3.54%	4.29%	8.19%	4.29%	2.76%
Consumer Cyclical	11.29%	27.65%	17.06%	15.92%	11.50%
Financial Services	26.05%	15.10%	21.58%	22.87%	16.99%
Real Estate		0.00%	1.43%	1.06%	0.47%	2.18%
							
*Sensitive*		43.23%	37.34%	33.68%	40.65%	41.70%
Communication Services	0.00%	1.39%	1.73%	0.54%	3.46%
Energy			7.77%	2.63%	5.18%	6.22%	5.97%
Industrials		12.92%	18.91%	15.87%	14.72%	10.70%
Technology		22.55%	14.42%	10.90%	19.18%	21.57%
							
*Defensive*		15.88%	14.18%	18.42%	15.78%	24.85%
Consumer Defensive	4.69%	5.65%	7.80%	5.30%	8.03%
Healthcare		10.86%	7.63%	7.16%	9.63%	14.13%
Utilities		0.33%	0.90%	3.46%	0.85%	2.69%
As you would expect, the multifactor fund is closer to market sector weightings at the moment.

Angst
Posts: 1847
Joined: Sat Jun 09, 2007 11:31 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Angst » Fri Mar 09, 2018 5:31 pm

jhfenton wrote:
Fri Mar 09, 2018 6:00 am
Theoretical wrote:
Thu Mar 08, 2018 10:26 pm
How'd you calculate the combined style box, @Angst?
I had those numbers in my tables, only horizontally. He did what I should have done and presented them properly.
Of course you did all the real work; very much appreciated. Does this now mean you will be updating data on a regular basis until M* sees fit to take over? :wink:

In all seriousness though, I do hope these numbers are indicative of where they will end up, say 6 months or so down the road, but I don't really know how confident we can be of that. Do you? Fairly confident, I hope. 785 seems like a good number of companies. I don't suppose you have any context to add... such as how many companies there were say a week or 10 days ago? Or does this represent the first public release of any of this kind of data for these new funds?

Jiu Jitsu Fighter
Posts: 134
Joined: Mon Nov 21, 2016 10:22 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Jiu Jitsu Fighter » Fri Mar 09, 2018 7:15 pm

Very Awesome work jhfenton. Thanks for putting in the time.
whodidntante wrote:
Thu Mar 08, 2018 10:30 pm
Let's keep these threads going until the volume improves.
I imagine we'll see exponential growth in these funds, especially the multifactor fund/ETF barring a market crash. As much as I respect Larry Swedroe, he'll be saying, "for Christ's sake, don't get your factor exposure from Vanguard! That's impossible that the funds can provide meaningful factor loads. Only AQR can do that!" (At 8 times the expense). His new book is good BTW. I highly recommend it unless you are a dogmatic 3-funder.

Haven't seen much hoopla around these funds, but from what I understand, the ETFs are marketed toward FAs. That and Vanguard generally let's the funds/ETFs sell themselves.

This will be an interesting ride for the next few years to see if Vanguard can dominate the factor-based funds market as they have with indexing. I'm still unsure how their liquidity fund will turn out considering the size and scale of Vanguard.

Jiu Jitsu Fighter
Posts: 134
Joined: Mon Nov 21, 2016 10:22 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Jiu Jitsu Fighter » Fri Mar 09, 2018 7:20 pm

jhfenton wrote:
Thu Mar 08, 2018 1:14 pm
I'm going to do the multi-factor fund next. It's probably the only other one I care enough about.
Any chance you'd have interest in tackling the momentum one?

Beehave
Posts: 351
Joined: Mon Jun 19, 2017 12:46 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Beehave » Fri Mar 09, 2018 7:48 pm

I see that there are 6 factor ETFs and 2 mutual funds. I believe there are no special tax implications if someone holds these in an IRA or 401k fund.

But what happens if you hold these in a non-tax-sheltered account? I'd imagine that there is turnover in the holdings of these factor funds, and would think therefore that there would be tax implications depending on losses and gains. Is the holder of the mutual fund shares or ETF shares somehow sheltered from these "day-to-day" gains and losses and only subject to taxes on dividends and, when they actually sell their shares, the gains or losses? Or is there taxation also on the actual churn of the holdings of the fund even though you are simply holding onto your shares?

Hope this isn't an insultingly trivial question. I'm accustomed to holding only the broadest index funds in non-tax-sheltered accounts and they produce dividends but have never to to-date produced capital gains taxes and am wondering if I could expect the same from the factor funds. Thanks.

Theoretical
Posts: 1422
Joined: Tue Aug 19, 2014 10:09 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Theoretical » Fri Mar 09, 2018 8:18 pm

For what it's worth, IJS has 40-50% turnover and has never distributed a capital gain. Vanguard is just as good as Blackrock in the tax management game.

It might in its first year if there's a huge run up in the market, but is pretty unlikely to do so.

If you read how they explain active, this is a lot closer to a DFA style active that's focused on avoiding bad forced trades, managing trading costs, and to keeping factor exposure constant and high. This could be high turnover (and momentum should be) but it may not end up being that at all.

Theoretical
Posts: 1422
Joined: Tue Aug 19, 2014 10:09 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Theoretical » Fri Mar 09, 2018 8:27 pm

Jiu Jitsu Fighter wrote:
Fri Mar 09, 2018 7:15 pm
Very Awesome work jhfenton. Thanks for putting in the time.
whodidntante wrote:
Thu Mar 08, 2018 10:30 pm
Let's keep these threads going until the volume improves.
I imagine we'll see exponential growth in these funds, especially the multifactor fund/ETF barring a market crash. As much as I respect Larry Swedroe, he'll be saying, "for Christ's sake, don't get your factor exposure from Vanguard! That's impossible that the funds can provide meaningful factor loads. Only AQR can do that!" (At 8 times the expense).
Actually, I wouldn't be at all surprised if he loves them, because the information from Vanguaed is strongly indicating that these are not mainstream, low tracking error funds with modest to moderate exposures, but instead very high and consistent exposure.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Fri Mar 09, 2018 10:05 pm

Angst wrote:
Fri Mar 09, 2018 5:31 pm
jhfenton wrote:
Fri Mar 09, 2018 6:00 am
Theoretical wrote:
Thu Mar 08, 2018 10:26 pm
How'd you calculate the combined style box, @Angst?
I had those numbers in my tables, only horizontally. He did what I should have done and presented them properly.
Of course you did all the real work; very much appreciated. Does this now mean you will be updating data on a regular basis until M* sees fit to take over? :wink:

In all seriousness though, I do hope these numbers are indicative of where they will end up, say 6 months or so down the road, but I don't really know how confident we can be of that. Do you? Fairly confident, I hope. 785 seems like a good number of companies. I don't suppose you have any context to add... such as how many companies there were say a week or 10 days ago? Or does this represent the first public release of any of this kind of data for these new funds?
I'm not sure the exact day they started releasing the portfolios. They seem to be changing the "last updated" date every day, but at first glance the portfolios seem static. Both funds seem to have the same 100,000 shares that they went live with. I don't think they've had to create any new shares yet, and I assume they haven't refreshed the portfolio yet.

With the volumes on VFMF coming in higher, I wouldn't be surprised to see a new 25,000 share creation unit pop up along with another $1.9MM in assets. Based on etf.com, VFMO seems to be the only one of the factor funds with an extra creation unit so far. It's at $9.9MM in assets.

Of course, VFMF has its companion VFMFX, which has some assets (including mine).
Jiu Jitsu Fighter wrote:
Fri Mar 09, 2018 7:20 pm
jhfenton wrote:
Thu Mar 08, 2018 1:14 pm
I'm going to do the multi-factor fund next. It's probably the only other one I care enough about.
Any chance you'd have interest in tackling the momentum one?
I'll do the momentum one tomorrow. Now that I know what I'm doing, it doesn't take long. I may do liquidity too. I'm curious what that one looks like.
Beehave wrote:
Fri Mar 09, 2018 7:48 pm
I see that there are 6 factor ETFs and 2 mutual funds. I believe there are no special tax implications if someone holds these in an IRA or 401k fund.
There is only one new "factor" mutual fund, the multifactor VFMFX. There is a now-confusingly-named Vanguard U.S. Value Fund (VUVLX) managed by the same Quantitative Equity Group, but it is an older fund and has nothing to do with the Vanguard U.S. Value Factor ETF. :oops: (It's a tame fund with a value and multifactor mandate designed for low-tracking error to the Russell 3000 Value Index.)
Last edited by jhfenton on Fri Mar 09, 2018 10:13 pm, edited 1 time in total.

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by lack_ey » Fri Mar 09, 2018 10:12 pm

jhfenton wrote:
Fri Mar 09, 2018 10:05 pm
I'll do the momentum one tomorrow. Now that I know what I'm doing, it doesn't take long. I may do liquidity too. I'm curious what that one looks like.
Actually, I have a script for pulling some of this Morningstar data (...scraping directly from the webpage, not via their actual tools), and I'm a bit curious as well, so I might do some additional processing/visualization anyway.

I'll probably run it regardless, so no need to rush. Though my method won't produce the same exact results, I don't think. The way Morningstar accounts for zero or invalid data and how it averages over stocks for a fund may differ from what I'd think to do.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Fri Mar 09, 2018 10:17 pm

lack_ey wrote:
Fri Mar 09, 2018 10:12 pm
jhfenton wrote:
Fri Mar 09, 2018 10:05 pm
I'll do the momentum one tomorrow. Now that I know what I'm doing, it doesn't take long. I may do liquidity too. I'm curious what that one looks like.
Actually, I have a script for pulling some of this Morningstar data (...scraping directly from the webpage, not via their actual tools), and I'm a bit curious as well, so I might do some additional processing/visualization anyway.

If you don't have it up tomorrow, I can probably run it, though the method wouldn't produce the same exact results, I don't think.

The way Morningstar accounts for zero or invalid data and how it averages over stocks for a fund may differ from what I'd think to do.
I'm exporting Vanguard's portfolio holdings from Vanguard.com, reformatting them properly in Excel/csv, and importing them 200 at a time (ordered by holding size) into Portfolio Manager. Then I look at each 200-stock portfolio's X-Ray, pull the stats, and value-weight them to get the overall averages. There are undoubtedly some rounding errors that creep in, but they should be close.

fennewaldaj
Posts: 219
Joined: Sun Oct 22, 2017 11:30 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by fennewaldaj » Fri Mar 09, 2018 10:26 pm

The value fund looks pretty great for those like myself that want to have value exposure across all caps. Thanks for taking the time to do this. I bought 3 shares this morning. I know that is a comically low amount but my taxable account was the only place I had free cash for purchases. I am contemplating partially replacing my Vanguard small cap value position in my rollover IRA with this though I might wait a bit longer to see what it stabilizes too before I do that.

User avatar
grabiner
Advisory Board
Posts: 22728
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Vanguard New Factor Funds Portfolio Statistics

Post by grabiner » Fri Mar 09, 2018 11:06 pm

Beehave wrote:
Fri Mar 09, 2018 7:48 pm
I see that there are 6 factor ETFs and 2 mutual funds. I believe there are no special tax implications if someone holds these in an IRA or 401k fund.

But what happens if you hold these in a non-tax-sheltered account? I'd imagine that there is turnover in the holdings of these factor funds, and would think therefore that there would be tax implications depending on losses and gains. Is the holder of the mutual fund shares or ETF shares somehow sheltered from these "day-to-day" gains and losses and only subject to taxes on dividends and, when they actually sell their shares, the gains or losses? Or is there taxation also on the actual churn of the holdings of the fund even though you are simply holding onto your shares?
The ETF structure, with creations and redemptions, allows ETFs to avoid most capital gains on portfolio turnover. However, this is less effective just after a fund starts; two of the three Vanguard stock ETFs which distributed capital gains distributed them only immediately after starting. (The third is REIT ETF, which shouldn't be held in a taxable account anyway.)

I wouldn't refuse to buy a new ETF because of the possibility of first-year capital gains. While there might be some gains, the dividend payment in the first year is also likely to be lower, because the ETF is likely to be growing rapidly. (For example, if a stock pays a dividend in October, the December dividend for the fund will include only the dividends on the shares held in October, not on new shares added in November and December.)
Wiki David Grabiner

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by lack_ey » Sat Mar 10, 2018 12:06 am

A preview. Click images for larger.

Image
Image

Dot size corresponds to the weighting within the fund. Do you think forward P/E is more useful or trailing P/E? Actually, doing any or P/B or P/S is just as easy, though I'd need to figure out what to do with some more of the NaN values. Maybe impute based on other stats.

How does one normally compute fund P/E and so on from single-stock data? Seemingly (total fund AUM) / (sum of earnings of all of the holdings) would make sense, but it seems common to drop stocks with negative earnings, do some other averaging, etc.

Some data sources just list nothing for some stocks; other times, negative values; other times, zero. And zero is obviously impossible. I'm beginning to see some stocks, even non-obscure ones like HP, can have negative book value.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Sat Mar 10, 2018 8:51 am

Vanguard U.S. Momentum Factor

Code: Select all

Pos#s	 VFMO 		P/PE	P/B	ROA	ROE	PEG	Yield	 Avg Mkt Cap 		LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	 $5,740,296.71 	21.13	5.57	6.28%	35.33%	15.24%	0.87%	 $28,812,870,000.00 	11	14	37	3	5	16	1	4	9
201-400	 $2,211,797.21 	20.29	4.27	-4.19%	11.87%	13.71%	0.44%	 $3,416,090,000.00 	1	2	2	4	9	18	5	18	38
401-600	 $1,523,463.92 	18.25	2.92	0.98%	6.35%	14.46%	0.73%	 $2,774,480,000.00 	1	3	2	4	8	9	15	19	36
601-710	 $574,841.53 	17.15	2.44	2.48%	0.95%	10.84%	0.97%	 $1,561,260,000.00 	0	0	0	0	2	1	17	40	38
Total	 $10,050,399.37 20.28	4.70	2.96%	23.81%	14.53%	0.76%	 $17,718,143,519.07 	7	9	22	3	6	15	5	11	21
P/PE = Price / Projected Earnings
PEG = Projected 5-year EPS Growth %

Code: Select all

   V   B   G
L  07  09  22
M  03  06  15
S  05  11  21
The portfolio lists one Russell 2000 e-mini future with a value of ~$78K, so I substituted an equal value of the Vanguard Russell 2000 Index Institutional Fund.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Sat Mar 10, 2018 9:04 am

Code: Select all

VFMO			001-200	201-400	401-600	601-785	Overall	MTUM
*Cyclical*		32.78%	30.61%	43.59%	39.01%	34.30%	37.88%
Basic Materials		3.66%	5.74%	10.92%	1.77%	5.11%	1.34%
Consumer Cyclical	17.78%	17.89%	20.38%	21.39%	18.40%	7.13%
Financial Services	10.95%	6.03%	9.80%	13.42%	9.83%	29.23%
Real Estate		0.39%	0.95%	2.49%	2.43%	0.95%	0.18%
							
*Sensitive*		40.60%	37.08%	37.40%	43.35%	39.50%	45.38%
Communication Services	0.25%	1.43%	0.00%	1.68%	0.55%	0.27%
Energy			2.94%	0.00%	4.15%	0.85%	2.36%	0.00%
Industrials		11.00%	17.93%	16.01%	22.22%	13.93%	15.31%
Technology		26.41%	17.72%	17.24%	18.60%	22.66%	29.80%
							
*Defensive*		26.63%	32.31%	19.01%	17.62%	26.21%	16.73%
Consumer Defensive	3.79%	5.33%	4.30%	3.83%	4.21%	1.05%
Healthcare		22.49%	26.98%	13.33%	11.44%	21.46%	15.68%
Utilities		0.35%	0.00%	1.38%	2.35%	0.54%	0.00%
Edited to add MTUM (iShares Edge MSCI USA Momentum Factor ETF) for comparison.
Last edited by jhfenton on Sat Mar 10, 2018 9:30 am, edited 3 times in total.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Sat Mar 10, 2018 9:25 am

lack_ey wrote:
Sat Mar 10, 2018 12:06 am
Dot size corresponds to the weighting within the fund. Do you think forward P/E is more useful or trailing P/E? Actually, doing any or P/B or P/S is just as easy, though I'd need to figure out what to do with some more of the NaN values. Maybe impute based on other stats.

How does one normally compute fund P/E and so on from single-stock data? Seemingly (total fund AUM) / (sum of earnings of all of the holdings) would make sense, but it seems common to drop stocks with negative earnings, do some other averaging, etc.

Some data sources just list nothing for some stocks; other times, negative values; other times, zero. And zero is obviously impossible. I'm beginning to see some stocks, even non-obscure ones like HP, can have negative book value.
Thanks, lack_ey. Those are interesting cloud plots. Eyeballing it, they seem consistent with the M* Portfolio Manager X-Ray data.

I usually end up using M*'s forward PE for comparisons because it's consistent across funds. It's hard to know whether other numbers are comparable across fund families.

User avatar
GreatOdinsRaven
Posts: 531
Joined: Thu Apr 23, 2015 8:47 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by GreatOdinsRaven » Sat Mar 10, 2018 10:51 am

jhfenton wrote:
Fri Mar 09, 2018 10:05 pm

I'll do the momentum one tomorrow. Now that I know what I'm doing, it doesn't take long. I may do liquidity too. I'm curious what that one looks like.
Beehave wrote:
Fri Mar 09, 2018 7:48 pm
I see that there are 6 factor ETFs and 2 mutual funds. I believe there are no special tax implications if someone holds these in an IRA or 401k fund.
There is only one new "factor" mutual fund, the multifactor VFMFX. There is a now-confusingly-named Vanguard U.S. Value Fund (VUVLX) managed by the same Quantitative Equity Group, but it is an older fund and has nothing to do with the Vanguard U.S. Value Factor ETF. :oops: (It's a tame fund with a value and multifactor mandate designed for low-tracking error to the Russell 3000 Value Index.)
jhfenton, thank you for doing this. Fascinating thread.

I’ve been using iShares Edge MSCI USA Momentum Factor ETF (MTUM) for my momentum exposure for approximately two years. It’s done well, though I’ve read postings indicating that it isn’t just targeting momentum according to standard long-only definitions (I can’t explain it, though. Alpha Architect May have posted a blog about this. Not sure). I’d be curious to see Vanguard’s momentum Fund construction compared to iShares Edge MSCI USA Momentum Factor Index (MTUM) and AQR’s Momentum Fund (AMOMX). If it’s similar to MTUM I’d switch new contributions to Vanguard in a heartbeat.

Thank you again for starting this thread. Looking forward to everyone’s updates as the AUM grows in each fund.

GOR

*edited to include a link to Alpha Architect’s blog post on why MTUM is only a “quasi momentum fund”.

https://alphaarchitect.com/2017/07/24/m ... -and-size/

Can anyone evaluate the Vanguard U.S. Momentum Factor ETF (VFMO) offering and compare it to MTUM as well as address whether it’s targeting the academic definition of long-only momentum? I actually participated in the Vanguard webcast, yesterday at 1pm, discussing their Active Factor ETFs but I couldn’t pay attention to it because I was at work. I did hear them mention that at least some if not all of these ETFs are not market cap-weighted and that they will have significant tracking error relative to the Russell 3000 (and that’s a great thing for people looking to load up on particular factor(s)).

As a person currently paying for platform access to invest taxable money in DFA funds I’m keenly interested in seeing how Vanguard will compare to DFA. Hoping it will one day allow me to move new contributions to Vanguard.

In an earlier posting I saw that the value ETF was similar to 50/50 DFA US large cap value/DFA US SCV (DFLVX/DFSVX). The weighted average of that portfolio is 39.5bps. Wishing, but wouldn’t it be great to get something that performs similarly for 13bps all while having the built-in tax efficiency of an ETF?
"The greatest enemies of the equity investor are expenses and emotions." -John C. Bogle, Little Book of Common Sense Investing. | | "Winter is coming." Lord Eddard Stark.

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Sat Mar 10, 2018 1:10 pm

GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
jhfenton, thank you for doing this. Fascinating thread.
:beer
GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
*edited to include a link to Alpha Architect’s blog post on why MTUM is only a “quasi momentum fund”.

https://alphaarchitect.com/2017/07/24/m ... -and-size/

Can anyone evaluate the Vanguard U.S. Momentum Factor ETF (VFMO) offering and compare it to MTUM as well as address whether it’s targeting the academic definition of long-only momentum? I actually participated in the Vanguard webcast, yesterday at 1pm, discussing their Active Factor ETFs but I couldn’t pay attention to it because I was at work. I did hear them mention that at least some if not all of these ETFs are not market cap-weighted and that they will have significant tracking error relative to the Russell 3000 (and that’s a great thing for people looking to load up on particular factor(s)).
Interesting. Thanks for the link. I don't have a source for the momentum data on VFMO's holdings, but VFMO is using the same all-cap strategy (L38/M24/S37) as VFMF (L38/M25/S38) and VFVA (L35/M20/S42) (and I assume the others). So it won't have the mega-cap only limitation of MTUM. Whether it is using the FF 2-12 Momentum is something that I don't know. It's not data that M* presents, even indirectly, unlike measures of value like P/B and P/E. If I had to guess, I'd say that Vanguard is using FF 2-12 or something similar. These funds seem to be an attempt at a rigorous application of theory.

I wish I had known about the webcast. I would have tried to listen to it too, even though I would have also been frequently distracted by work.
GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
In an earlier posting I saw that the value ETF was similar to 50/50 DFA US large cap value/DFA US SCV (DFLVX/DFSVX). The weighted average of that portfolio is 39.5bps. Wishing, but wouldn’t it be great to get something that performs similarly for 13bps all while having the built-in tax efficiency of an ETF?
It would. I currently have one DFA fund in my 401(k)--DFA US Small Cap (DFSTX)--and my wife's new 401(k) will have DFSVX (DFA US Small Cap Value). (She's eligible to participate May 1.) Neither 401(k) has any other DFA funds. The only other 401(k) fund I own is VMVAX (Vanguard Mid Cap Value Index Admiral). Everything else is in our Rollover IRAs and Roth IRAs-- where everything is currently Vanguard--or my HSA, where I have most of the $50+K in EMGF (iShares Edge Multifactor Emerging Markets). I am very happy with that fund, but it's up in DFA territory at 45 bp. I'd love to have a Vanguard version at 25 bp.

User avatar
GreatOdinsRaven
Posts: 531
Joined: Thu Apr 23, 2015 8:47 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by GreatOdinsRaven » Sat Mar 10, 2018 2:39 pm

jhfenton wrote:
Sat Mar 10, 2018 1:10 pm

I wish I had known about the webcast. I would have tried to listen to it too, even though I would have also been frequently distracted by work.
GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
In an earlier posting I saw that the value ETF was similar to 50/50 DFA US large cap value/DFA US SCV (DFLVX/DFSVX). The weighted average of that portfolio is 39.5bps. Wishing, but wouldn’t it be great to get something that performs similarly for 13bps all while having the built-in tax efficiency of an ETF?
It would. I currently have one DFA fund in my 401(k)--DFA US Small Cap (DFSTX)--and my wife's new 401(k) will have DFSVX (DFA US Small Cap Value). (She's eligible to participate May 1.) Neither 401(k) has any other DFA funds. The only other 401(k) fund I own is VMVAX (Vanguard Mid Cap Value Index Admiral). Everything else is in our Rollover IRAs and Roth IRAs-- where everything is currently Vanguard--or my HSA, where I have most of the $50+K in EMGF (iShares Edge Multifactor Emerging Markets). I am very happy with that fund, but it's up in DFA territory at 45 bp. I'd love to have a Vanguard version at 25 bp.
Vanguard said they’d post a link to edited video from the discussion in a few months. Will return and post a link if I see one in the future.

GOR
"The greatest enemies of the equity investor are expenses and emotions." -John C. Bogle, Little Book of Common Sense Investing. | | "Winter is coming." Lord Eddard Stark.

User avatar
cfs
Posts: 4154
Joined: Fri Feb 23, 2007 1:22 am
Location: ~ Mi Propio Camino ~

Re: Vanguard New Factor Funds Portfolio Statistics

Post by cfs » Sat Mar 10, 2018 3:04 pm

Good conversation.

Thank you jhfenton and everyone contributing, good work.

I will continue to follow from the background.

Good luck, y gracias por leer cfs~
~ Member of the Active Retired Force since 2014 ~

cheapskate
Posts: 746
Joined: Thu Apr 26, 2007 1:05 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by cheapskate » Sun Mar 11, 2018 11:37 am

GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
In an earlier posting I saw that the value ETF was similar to 50/50 DFA US large cap value/DFA US SCV (DFLVX/DFSVX). The weighted average of that portfolio is 39.5bps. Wishing, but wouldn’t it be great to get something that performs similarly for 13bps all while having the built-in tax efficiency of an ETF?
My goto offering for US equities is DFA Vector (I hold it in taxable). I went back a few years and saw that DFA Vector has consistently distributed between 2.5%-3.5% of NAV as LT Cap gains each year, resulting in a tax drag. I too, am very interested in VFVA as an alternative to DFA Vector purely from a tax efficiency standpoint. While the Value and Small factor loadings of VFVA seem impressive, DFA Vector owns many more equities when compared to VFVA (~2600 vs ~750), only time will tell whether the deeper diversification of Vector is enough of a benefit to overcome its tax drag.

Fascinating thread. Thanks to jhfenton for the work and the data.

cheapskate
Posts: 746
Joined: Thu Apr 26, 2007 1:05 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by cheapskate » Sun Mar 11, 2018 11:38 am

cheapskate wrote:
Sun Mar 11, 2018 11:37 am
GreatOdinsRaven wrote:
Sat Mar 10, 2018 10:51 am
In an earlier posting I saw that the value ETF was similar to 50/50 DFA US large cap value/DFA US SCV (DFLVX/DFSVX). The weighted average of that portfolio is 39.5bps. Wishing, but wouldn’t it be great to get something that performs similarly for 13bps all while having the built-in tax efficiency of an ETF?
My goto offering for US equities is DFA Vector (I hold it in taxable). I went back a few years and saw that DFA Vector has consistently distributed between 2.5%-3.5% of NAV as LT Cap gains each year, resulting in a tax inefficiency. I too, am very interested in VFVA as an alternative to DFA Vector purely from a tax efficiency standpoint. While the Value and Small factor loadings of VFVA seem impressive, DFA Vector owns many more equities when compared to VFVA (~2600 vs ~750), only time will tell whether the deeper diversification of Vector is enough of a benefit to overcome its tax drag.

Fascinating thread. Thanks to jhfenton for the work and the data.

User avatar
AndrewXnn
Posts: 152
Joined: Thu Jan 02, 2014 7:55 pm
Location: Upstate New York

Re: Vanguard New Factor Funds Portfolio Statistics

Post by AndrewXnn » Sun Mar 11, 2018 12:28 pm

Formatting this is a pain.

Anyhow, here's are some relevant calculations/estimate of alpha.
Scaled up from past 22 or 21 days of trading data to 1 year.
Notice that Friday's performance tends to mirror the amount of alpha being offered.
Will obviously need more trading data to establish statistical significant values.


3/9/2018 22day 21 day
VFMV US Minimum Volatility 0.87% 2% -5%
VFVA US Value Factor 1.39% 0% -7%
VFMO US Momentum Factor 1.90% 30% 33%
VFLQ US Liquidity Factor 1.68% 30% 12%
VFQY US Quality Factor 1.25% 26% 3%
VFMF US Multifactor 1.92% 13% 5%
VFM US Multifactor Admiral 1.86% 11% 9%
VOO S&P 500 1.73% 0% 0%

User avatar
jhfenton
Posts: 3308
Joined: Sat Feb 07, 2015 11:17 am
Location: Ohio

Re: Vanguard New Factor Funds Portfolio Statistics

Post by jhfenton » Sun Mar 11, 2018 12:55 pm

While I was watching the Harvard-Penn basketball game:

Vanguard U.S. Liquidity Factor ETF:

Code: Select all

POS#s	 VFLQ 		P/PE	P/B	ROA	ROE	PEG	Yield	 Avg Mkt Cap 		LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	 $3,970,486.91 	19.18	2.65	5.96%	24.94%	11.88%	1.66%	 $19,220,610,000.00 	19	23	22	5	4	5	9	7	6
201-400	 $1,545,385.65 	18.29	2.15	3.20%	7.92%	13.80%	1.18%	 $2,020,520,000.00 	1	0	2	8	6	11	21	27	22
401-600	 $1,131,001.87 	17.78	1.87	-1.54%	-1.27%	18.92%	1.41%	 $1,391,170,000.00 	1	0	0	7	5	5	26	29	22
601-800	 $834,968.40 	18.73	2.15	1.25%	-4.19%	11.83%	1.34%	 $1,676,930,000.00 	0	0	0	5	6	8	26	28	25
801-902	 $364,001.06 	13.50	1.51	-2.11%	-3.39%	12.01%	1.94%	 $1,194,270,000.00 	1	0	0	4	9	1	41	30	13
Total	 $7,845,843.89 	18.49	2.33	3.46%	13.39%	13.27%	1.51%	 $10,559,218,250.92 	10	12	12	6	5	6	17	17	14
P/PE = Price / Projected Earnings
PEG = Projected 5-year EPS Growth %

Code: Select all

   V   B   G
L  10  12  12
M  06  05  06
S  17  17  14

Code: Select all

VFLQ			001-200	201-400	401-600	601-800	801-902 Overall 
*Cyclical*		46.75%	49.11%	49.32%	45.53%	47.04%	47.47%
Basic Materials		8.92%	3.29%	3.99%	6.67%	8.19%	6.83%
Consumer Cyclical	4.90%	7.50%	14.76%	13.11%	14.88%	8.17%
Financial Services	26.68%	28.47%	23.01%	15.82%	15.43%	24.83%
Real Estate		6.25%	9.85%	7.56%	9.93%	8.54%	7.65%
								
*Sensitive*		29.26%	35.21%	29.28%	33.37%	36.53%	31.21%
Communication Services	2.38%	1.14%	2.49%	1.02%	0.00%	1.90%
Energy			3.21%	1.37%	2.49%	3.01%	10.10%	3.04%
Industrials		16.25%	19.43%	14.69%	15.67%	22.32%	16.87%
Technology		7.42%	13.27%	9.61%	13.67%	4.11%	9.40%
								
*Defensive*		24.00%	15.67%	21.40%	21.10%	16.42%	21.32%
Consumer Defensive	5.00%	4.49%	4.24%	1.85%	2.86%	4.36%
Healthcare		11.21%	9.14%	13.50%	17.77%	10.37%	11.79%
Utilities		7.79%	2.04%	3.66%	1.48%	3.19%	5.18%
I'm still not sure what to make of this one. As you might expect, the average market cap is a bit lower, but it's still an all-cap fund. There are some mild sector tilts (less tech, more real estate) because of the liquidity screen, but nothing remarkable.

(Now it's time for my run.)

User avatar
AndrewXnn
Posts: 152
Joined: Thu Jan 02, 2014 7:55 pm
Location: Upstate New York

Re: Vanguard New Factor Funds Portfolio Statistics

Post by AndrewXnn » Sun Mar 11, 2018 1:41 pm

Hate to go off topic...

Is there a "how to" post on how to easily paste in parts of spreadsheets and the like?

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by lack_ey » Sun Mar 11, 2018 2:09 pm

AndrewXnn wrote:
Sun Mar 11, 2018 1:41 pm
Hate to go off topic...

Is there a "how to" post on how to easily paste in parts of spreadsheets and the like?
One of the Excel export types "Change file Type" is "Formatted Text (Space delimited)" so that would work. Otherwise and perhaps if the spreadsheet application does not support it, you could just copy somehow and then do the spacing yourself in a number of ways (including manually in a text editor using monospace font), or via some script.

Then paste into a code block delimited by the code tags [ code ] and [ /code ] without the spaces.


btw I've now downloaded a bunch of Morningstar data and can generate these P/E vs. market cap style grid graphs for a number of funds. Any requests? (except DFA, unless you can find me a list of holdings with ticker symbols; they have pdfs—ew—with names and weightings, but importing data via stock names is messier than ticker symbols) edit: also, no international, as grabbing data for those tickers I think would be more error-prone and complicated

Additionally, with a bit more work I think I could run 12-2 return calculations for each stock and plot momentum as well, not just P/E. Actually, maybe I should create a new thread with this stuff, unless you think it should remain here.

Image
Image
Last edited by lack_ey on Sun Mar 11, 2018 2:41 pm, edited 1 time in total.

User avatar
triceratop
Moderator
Posts: 5655
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Vanguard New Factor Funds Portfolio Statistics

Post by triceratop » Sun Mar 11, 2018 2:20 pm

lack_ey wrote: btw I've now downloaded a bunch of Morningstar data and can generate these P/E vs. market cap style grid graphs for a number of funds. Any requests? (except DFA, unless you can find me a list of holdings with ticker symbols; they have pdfs—ew—with names and weightings, but importing data via stock names is messier than ticker symbols)
I think the factor tilters would be most interested in IJS, VBR, RZV. Foreign, maybe: DGS, VSS, IVLU, EFV ? Definitely don't do more than you have fun doing! I know how much work it is to do some of this stuff.

The plots look great. A new thread might be helpful to keep things on topic in this one.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

Theoretical
Posts: 1422
Joined: Tue Aug 19, 2014 10:09 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Theoretical » Sun Mar 11, 2018 2:36 pm

I'd add SFENX, SFILX, SFNNX (EM Large Value, International Small Blend-Value, and International Large Value) to the mix.

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by lack_ey » Sun Mar 11, 2018 2:40 pm

triceratop wrote:
Sun Mar 11, 2018 2:20 pm
lack_ey wrote: btw I've now downloaded a bunch of Morningstar data and can generate these P/E vs. market cap style grid graphs for a number of funds. Any requests? (except DFA, unless you can find me a list of holdings with ticker symbols; they have pdfs—ew—with names and weightings, but importing data via stock names is messier than ticker symbols)
I think the factor tilters would be most interested in IJS, VBR, RZV. Foreign, maybe: DGS, VSS, IVLU, EFV ? Definitely don't do more than you have fun doing! I know how much work it is to do some of this stuff.

The plots look great. A new thread might be helpful to keep things on topic in this one.
Theoretical wrote:
Sun Mar 11, 2018 2:36 pm
I'd add SFENX, SFILX, SFNNX (EM Large Value, International Small Blend-Value, and International Large Value) to the mix.
Oops, sorry, I don't think international stocks would work, or at least it would be actual effort at this point to try to grab that data. For now, just US funds.

edit: moved to new thread:

viewtopic.php?f=10&t=244016

User avatar
nedsaid
Posts: 10349
Joined: Fri Nov 23, 2012 12:33 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by nedsaid » Sun Mar 11, 2018 3:52 pm

Fascinating, Vanguard now seems to be gunning for Dimensional Fund Advisors. A criticism of Vanguard is that it seemed to take a half-hearted approach towards factor investing. I am thinking of all the discussion about Vanguard Small Cap Value Index in both mutual fund and ETF form, not loaded well enough for size or value. I do own the ETF and I have been very satisfied with it.

This new approach towards Factor investing by Vanguard seems superior to what they were doing before. Pretty much, they are offering DFA-like products that you can access directly without having to go through an advisor. I was very much impressed by DFA but I didn't want to pay Paul Merriman's former firm 1% a year for access. This will make do-it-yourself factor investing much easier. I wouldn't be shocked if Vanguard Advisory Service would offer factor based portfolio advice and directly compete with DFA at much lower advisory costs. Not the greatest at math but last I checked 0.3% a year in advisory fees is much lower than 1.0% a year. :wink:

I will have to take a lot at Vanguard's multi-factor fund too. They also have a Market-Neutral Fund. I know this gives John Bogle heartburn, but it looks like AQR is next in Vanguard's crosshairs. Vanguard looks more and more like the blob that ate the mutual fund industry. Watch out hedge funds, you are next.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 10349
Joined: Fri Nov 23, 2012 12:33 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by nedsaid » Sun Mar 11, 2018 4:01 pm

Jiu Jitsu Fighter wrote:
Fri Mar 09, 2018 7:15 pm

I imagine we'll see exponential growth in these funds, especially the multifactor fund/ETF barring a market crash. As much as I respect Larry Swedroe, he'll be saying, "for (Pete's) sake, don't get your factor exposure from Vanguard! That's impossible that the funds can provide meaningful factor loads. Only AQR can do that!" (At 8 times the expense). His new book is good BTW. I highly recommend it unless you are a dogmatic 3-funder.

I can imagine Larry including DFA in with AQR in saying that they provide superior factor exposure to Vanguard. The thing is, I don't think that is true any longer.

Another thing to keep an eye on is if the increasing amounts of monies under management will dilute the factor loadings, for example in a Value ETF I suppose the managers would have to migrate to higher and higher market caps. It seems to me that at some point you have to limit the size of the portfolios. A closed end mutual fund would solve the size problem but then you get into the premium/discount problem. Not sure you can close an ETF to new investors. Could somebody address this?
A fool and his money are good for business.

User avatar
GreatOdinsRaven
Posts: 531
Joined: Thu Apr 23, 2015 8:47 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by GreatOdinsRaven » Sun Mar 11, 2018 5:03 pm

nedsaid wrote:
Sun Mar 11, 2018 3:52 pm
Fascinating, Vanguard now seems to be gunning for Dimensional Fund Advisors. A criticism of Vanguard is that it seemed to take a half-hearted approach towards factor investing. I am thinking of all the discussion about Vanguard Small Cap Value Index in both mutual fund and ETF form, not loaded well enough for size or value. I do own the ETF and I have been very satisfied with it.

This new approach towards Factor investing by Vanguard seems superior to what they were doing before. Pretty much, they are offering DFA-like products that you can access directly without having to go through an advisor. I was very much impressed by DFA but I didn't want to pay Paul Merriman's former firm 1% a year for access. This will make do-it-yourself factor investing much easier. I wouldn't be shocked if Vanguard Advisory Service would offer factor based portfolio advice and directly compete with DFA at much lower advisory costs. Not the greatest at math but last I checked 0.3% a year in advisory fees is much lower than 1.0% a year. :wink:

I will have to take a lot at Vanguard's multi-factor fund too. They also have a Market-Neutral Fund. I know this gives John Bogle heartburn, but it looks like AQR is next in Vanguard's crosshairs. Vanguard looks more and more like the blob that ate the mutual fund industry. Watch out hedge funds, you are next.
I don’t think AQR has anything to worry about...
And that’s why I’m not writing off DFA, yet...

Here’s the growth of $10,000 in Vanguard’s market neutral fund vs AQR’s market neutral fund:

Vanguard CAGR 1.55%
AQR CAGR 9.46%


Here’s what a monthly dollar cost averager would have experienced starting at $1000 and investing $1000 monthly over the life of the youngest fund:


Vanguard’s IRR was 0.8%. CAGR 221.87% (DCA’ing)
AQR’s IRR was 7.72% CAGR 232.89%
"The greatest enemies of the equity investor are expenses and emotions." -John C. Bogle, Little Book of Common Sense Investing. | | "Winter is coming." Lord Eddard Stark.

User avatar
nedsaid
Posts: 10349
Joined: Fri Nov 23, 2012 12:33 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by nedsaid » Sun Mar 11, 2018 5:31 pm

It is interesting how well AQR's Market Neutral fund has performed. The Market Neutral funds I have seen have returns comparable to Treasury Bills, which doesn't make me want to invest. On the other hand, AQR has done something right. Hard to argue with the results.

Note that I wondered how well Vanguard would maintain factor tilts once asset size grows. We will also see if they maintain the intellectual commitment to factors. Certainly the folks at AQR and DFA are true believers.

Thanks for responding.
A fool and his money are good for business.

fennewaldaj
Posts: 219
Joined: Sun Oct 22, 2017 11:30 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by fennewaldaj » Sun Mar 11, 2018 6:12 pm

nedsaid wrote:
Sun Mar 11, 2018 5:31 pm
It is interesting how well AQR's Market Neutral fund has performed. The Market Neutral funds I have seen have returns comparable to Treasury Bills, which doesn't make me want to invest. On the other hand, AQR has done something right. Hard to argue with the results.

Note that I wondered how well Vanguard would maintain factor tilts once asset size grows. We will also see if they maintain the intellectual commitment to factors. Certainly the folks at AQR and DFA are true believers.

Thanks for responding.
I would think that these funds could get pretty big as is. Current small cap value fund has ~ 30 billion. This value factor fund has a similar amount of names and a larger average market cap.

lack_ey
Posts: 6613
Joined: Wed Nov 19, 2014 11:55 pm

Re: Vanguard New Factor Funds Portfolio Statistics

Post by lack_ey » Sun Mar 11, 2018 6:37 pm

fennewaldaj wrote:
Sun Mar 11, 2018 6:12 pm
I would think that these funds could get pretty big as is. Current small cap value fund has ~ 30 billion. This value factor fund has a similar amount of names and a larger average market cap.
The new value factor fund may have a larger average market cap, but it reaches deeper into a lot of the small caps, having higher weightings than the small cap value index fund does.

Check in my companion thread:
viewtopic.php?f=10&t=244016&newpost=382 ... ead#unread

Image Image

Also the turnover should be higher. I would say that the small cap value index fund in particular has higher capacity than you'd think because it reaches into mid caps and has a significant weighting there, and the CRSP index rules have the packeting procedure to mitigate turnover.

I don't disagree that the new factor funds could get pretty big, though.

Jiu Jitsu Fighter
Posts: 134
Joined: Mon Nov 21, 2016 10:22 am

Re: Vanguard New Factor Funds Portfolio Statistics

Post by Jiu Jitsu Fighter » Sun Mar 11, 2018 11:23 pm

jhfenton wrote:
Sat Mar 10, 2018 8:51 am
Vanguard U.S. Momentum Factor

Code: Select all

Pos#s	 VFMO 		P/PE	P/B	ROA	ROE	PEG	Yield	 Avg Mkt Cap 		LCV	LCB	LCG	MCV	MCB	MCG	SCV	SCB	SCG
001-200	 $5,740,296.71 	21.13	5.57	6.28%	35.33%	15.24%	0.87%	 $28,812,870,000.00 	11	14	37	3	5	16	1	4	9
201-400	 $2,211,797.21 	20.29	4.27	-4.19%	11.87%	13.71%	0.44%	 $3,416,090,000.00 	1	2	2	4	9	18	5	18	38
401-600	 $1,523,463.92 	18.25	2.92	0.98%	6.35%	14.46%	0.73%	 $2,774,480,000.00 	1	3	2	4	8	9	15	19	36
601-710	 $574,841.53 	17.15	2.44	2.48%	0.95%	10.84%	0.97%	 $1,561,260,000.00 	0	0	0	0	2	1	17	40	38
Total	 $10,050,399.37 20.28	4.70	2.96%	23.81%	14.53%	0.76%	 $17,718,143,519.07 	7	9	22	3	6	15	5	11	21
Awesome. Thanks for taking the time to put this together.
nedsaid wrote:
Sun Mar 11, 2018 4:01 pm
Another thing to keep an eye on is if the increasing amounts of monies under management will dilute the factor loadings, for example in a Value ETF I suppose the managers would have to migrate to higher and higher market caps.
Agreed -That's the big question. It seems inevitable. Nothing a market crash can't delay for at least a bit, though.

Post Reply