DFA Vector ETFs

[phoroner]

Launching on September 10th
DFA US Vector - DXUV - ER 0.28
DFA International Vector - DXIV - ER 0.34

Anyone planning to use these?

https://www.sec.gov/Archives/edgar/data ... 85apos.htm

[muffins14]

Launching on September 10th
DFA US Vector - DXUV - ER 0.28
DFA International Vector - DXIV - ER 0.34

Anyone planning to use these?

https://www.sec.gov/Archives/edgar/data ... 85apos.htm

Not personally. My value ETFs are now just two: FNDA and AVGV. Going forward I will just buy AVGV in tax-deferred. I have a lot of FNDA purchases in 2020 that likely will never be at a net loss.

[phoroner]

I have significant factor weights in my portfolio, so these funds appeal as an efficient way to do so, although still slightly costlier than MCW+SCV blend.

[happenstance]

No plans to use these myself. As I understand it, the Vector series of funds are broad market funds with significantly deeper size/value tilts compared to DFA’s Core 2 series. These contrast to the component funds that target size/value/profitability or some combination thereof. These Vector series are meant to be core holdings for those who want deep size/value exposure in a single holding.

My opinion is that you can get that exposure for cheaper using a core-satellite approach, with more flexibility for tax-loss harvesting and options from different fund providers. Historical factor returns, which we will assume the DFA ETFs try to mimic:

DFVEX (DFA US Vector Equity) SMB=0.42 / HML=0.38
DFSVX (DFA US Small Cap Value) SMB=0.78 / HML=0.62

You can match that factor exposure with:

53% VTI (Vanguard Total Stock, ER=0.03) SMB=0 / HML = 0
47% AVUV (Avantis US Small Cap Value, ER=0.25) SMB=0.69 / HML=0.73

The new US Vector ETF has a listed ER=0.28, whereas the exposure matched portfolio would have an ER=0.13.

[nisiprius]

Anyone have a coherent explanation, preferably from a Dimensional source, of exactly what the word "vector" is supposed to mean in these products?

From the linked document:

The US Vector Equity ETF is an actively managed exchange traded fund and does not seek to replicate the performance of a specific index and may have a higher degree of portfolio turnover than such index funds.

So it's explicitly active.

The Advisor may also increase or reduce the US Vector Equity ETF’s exposure to an eligible company, or exclude a company, based on shorter-term considerations, such as a company’s price momentum, short-run reversals, and investment characteristics. In assessing a company’s investment characteristics, the Advisor considers ratios such as recent changes in assets divided by total assets. The criteria the Advisor uses for assessing a company’s investment characteristics are subject to change from time to time.

So it does, or at least allows itself to do, stock-picking based on vague and changeable criteria.

Seems like a yawn.

I wonder if there are DFA fans who are sitting around saying "I remember the good old days, the Golden Age when DFA really stood for something--precision tools that were as passive as possible while providing the highest factor loadings possible. It ain't the same."

[edge]

It’s not really active. Vector funds, last time I checked, are kinda small/value tilted total market funds governed by factor loading and other passive quantitative screens.

These have been around for awhile and are a substitute for the ‘barbell’ strategy of having a total market fund and a scv fund.

You can go to m* and look at the weights for dfvex.

I don’t use these things because of 401k and deferred comp investment options essentially forcing a 500 index centric portfolio construction. I might consider it otherwise…haven’t given it much thought.

Anyone have a coherent explanation, preferably from a Dimensional source, of exactly what the word "vector" is supposed to mean in these products?

From the linked document:

The US Vector Equity ETF is an actively managed exchange traded fund and does not seek to replicate the performance of a specific index and may have a higher degree of portfolio turnover than such index funds.

So it's explicitly active.

The Advisor may also increase or reduce the US Vector Equity ETF’s exposure to an eligible company, or exclude a company, based on shorter-term considerations, such as a company’s price momentum, short-run reversals, and investment characteristics. In assessing a company’s investment characteristics, the Advisor considers ratios such as recent changes in assets divided by total assets. The criteria the Advisor uses for assessing a company’s investment characteristics are subject to change from time to time.

So it does, or at least allows itself to do, stock-picking based on vague and changeable criteria.

Seems like a yawn.

I wonder if there are DFA fans who are sitting around saying "I remember the good old days, the Golden Age when DFA really stood for something--precision tools that were as passive as possible while providing the highest factor loadings possible. It ain't the same."

[firebirdparts]

Is this the substance in question?

strongly increased exposure to smaller capitalization, lower relative price and higher profitability companies as compared to their representation in the U.S. Universe

One might ask what they were doing before, I guess. I'm sure you could do worse than this. I don't plan to do it, but I certainly don't plan to insult anybody doing it.

[nisiprius]

It’s not really active.

The prospectus says it's "active," please don't suggest ignoring the prospectus.

It's no sin to be actively managed--I have a big holding of the Vanguard Inflation-protected Securities fund, which is active--but it's something to know.

When a fund is indexed, it's indexed. But when a fund is active, it's kind of hard to judge how active it is. Still, activeshare.info tries.

Obviously there's no data for the ETFs yet, but according to activeshare.info, DFVEX, the DFA U.S. Vector Equity Portfolio...



...is more active than FCNTX, Fidelity Contrafund! Would you call Contrafund not "really" active? Fidelity explicitly ascribes the fund performance to its long-time manager, Will Danoff.



...More active than DFEOX, the DFA US Core Equity Portfolio.



...and, of course, more active than DFUSX, the DFA U.S. Large Company Portfolio.

[edge]

These appear to be based on a definition of active which is deeply flawed.

Many funds don’t have an appropriate index benchmark so they choose whatever is closest. But then at the same time don’t manage against it. The vector funds don’t really have good benchmarks. I guess DFA could create them but that would open up other problems.

I define active as a behavioral trait of the fund related to its underlying strategy. Ie does it try to pick winners using firm specific analysis or is it passively using quantitative screens and rules.

In other words without the s&p 500 index would an s&p 500 fund be active? I would say not. I think at that point the definition used in the post above becomes problematic as it requires the presence of some external ‘thing’ to detect what is really imo an innate behavior.

[nisiprius]

edge, it seems to me your are saying something rather close to

I shall not today attempt further to define the kinds of fund I understand to be embraced within that shorthand description, "active," and perhaps I could never succeed in intelligibly doing so. But I know it when I see it, and the fund involved in this case is not that.

Can you go beyond asserting your personal judgement? If you don't like "active share," what would you put in its place as an objective measure that would let people agree whether or not a fund is really "active?"

Are you, or are you not saying that the document linked by phoroner, above, is inaccurate?

Do you agree that Fidelity Contrafund is correctly described as "active" or is it, perhaps, not "really active" either?

[edge]

Sure. It’s pretty basic.

Contrafund, as per its strategy in its prospectus, is highly ‘active’. It states it tries to beat the market by picking stocks that are ‘undervalued’ by the public and performs fundamental analysis to unearth these opportunities. To know for sure one would need to analyze what the fund actually did. Perhaps they ended up being cowards and just did closet indexing, I wouldn’t know but it is possible to verify.

DFA does not say this. Instead DFA tries to capture the market return given a set of quantitative factor loading - size / value being the primary. This is not the same as total market investing to capture total market return, but it’s not that different than SCV index investing. Vector is more complicated because it covers more of the market cap size spectrum but the theory is the same.

Many index products have gotten complicated enough with niche coverage and new approaches to construction that they have started to look more like dfa-type approaches over time (eg rules that look a bit like momentum by not instantaneously selling when a stock has moved out of strict index inclusion rules, other techniques to avoid front running, etc). But the difference remains that dfa does not adhere to a published index and the exactness and precision of what they do is less transparent. In any case, because of this evolution of indexed products and because of the cost gap, I personally own very few dfa funds and don’t think they are at all necessary.

As far as the document that describes the ETF as active what I will say is that asset management is plagued by really poor vocabulary and definitions generally. ETFs probably should have been described as ‘indexed’ and ‘non-indexed' instead of where they landed.

[phoroner]

Of course DFA funds are active; marketing materials on their website frame it as something along the lines of active implementation of indexing. Then again, anything other than a total-market fund is active, including the SP500. See https://businesslawreview.uchicago.edu/ ... ses-sp-500. Vanguard has active equity funds. So the issue seems to be costs, turnover, avoiding speculative approaches, and staying the course. DFAs costs are reasonably low, if not Vanguard index-fund low; per Morningstar, the US Vector mutual fund DFVEX had 11% turnover, their Core 1 US equity mutual fund (DFEOX) was 8%, and VTSAX was 2%.

In US equity & focusing only on the value factor ETFs, DFA has Core, Core 1, Core 2, Marketwide Value, Targeted Value, and Small-Cap Value (progressively increasing factor exposure, and concentration). The Vector fund seems to fit in after Core 2, likely with stronger factor loadings but, unlikely Targeted Value or SCV, does not completely exclude stocks low on the desired factors. So the vector fund strikes me as a good tool for someone who wants heavy factor tilts. Simplicity may have lower transaction costs and be easier to stay-the-course with, even if the same factor exposure could be achieved at lower cost with a combination of a total-market and SCV fund.

I am interested in others opinions about Vector vs MCW+SCV. The behavioral aspect about Vector funds appeals to me. As in much about this, it may just be a matter of personal taste.

[edge]

That seems a pretty reductive definition where maybe you end up with only a global all asset market index fund is the only truly ‘passive’ product.

Here is what dfa says about themselves as of 2022:

https://www.dimensional.com/gb-en/insig ... ally-passe



Anyway, vector is fine if you want a very heavy tilt and your portfolio has room for it and you don’t mind the expense ratio.

[nedsaid]

Sure. It’s pretty basic.

Contrafund, as per its strategy in its prospectus, is highly ‘active’. It states it tries to beat the market by picking stocks that are ‘undervalued’ by the public and performs fundamental analysis to unearth these opportunities. To know for sure one would need to analyze what the fund actually did. Perhaps they ended up being cowards and just did closet indexing, I wouldn’t know but it is possible to verify.

DFA does not say this. Instead DFA tries to capture the market return given a set of quantitative factor loading - size / value being the primary. This is not the same as total market investing to capture total market return, but it’s not that different than SCV index investing. Vector is more complicated because it covers more of the market cap size spectrum but the theory is the same.

Many index products have gotten complicated enough with niche coverage and new approaches to construction that they have started to look more like dfa-type approaches over time (eg rules that look a bit like momentum by not instantaneously selling when a stock has moved out of strict index inclusion rules, other techniques to avoid front running, etc). But the difference remains that dfa does not adhere to a published index and the exactness and precision of what they do is less transparent. In any case, because of this evolution of indexed products and because of the cost gap, I personally own very few dfa funds and don’t think they are at all necessary.

As far as the document that describes the ETF as active what I will say is that asset management is plagued by really poor vocabulary and definitions generally. ETFs probably should have been described as ‘indexed’ and ‘non-indexed' instead of where they landed.

This is a very good post. This ought to be in the Wiki explaining why DFA and Avantis type of factor funds are not really active but not quite passive either. Close enough because these type of funds have relatively low turnover compared to what we would traditionally have called active funds.

The DFA US Vector Equity I Fund (DFVEX) is categorized by Morningstar as a Mid-Cap Value fund and it has the same 0.28% expense ratio as the new ETF that will be introduced in about three days. It has a portfolio turnover of 11% compared to Vanguard Mid-Cap Value Index Admiral (VMVAX) which has 22% portfolio turnover but the Vanguard product has a 0.07% expense ratio. The Vanguard product has a very low expense ratio in comparison to the DFA product and follows an index (probably CRSP) whereas the DFA product has a higher expense ratio, doesn't follow an index, but has half of portfolio turnover of the DFA product. In Nedsaid's mind, one of the things you consider when placing an investment product somewhere on the continuum between active and passive is portfolio turnover.

When you look at the portfolios of DFVEX versus VMVAX, the portfolios look very similar but DFA has a much smaller average market capitalization. When comparing the valuation measures (Price/Earnings, Price/Book, Price/Sales, Price/Cash flow). So you can see the trade-offs between factor loading and expense ratios. In fairness, the DFA US Vector Equity fund is not designed to be a Mid-Value product but because of its tilts it sort of turns out that way.

Edge has also explained why Nedsaid believes that the Style Index funds are plenty well good enough, mainly because of the patient trading techniques that Style Index funds are using to minimize such things as front-running and market impact costs. The CRSP indexes gradually move stocks in and out of the indexes, so the patient trading techniques are built into the construction of the index itself. It is interesting that over 10 years, the Vanguard fund has outperformed the DFA product very slightly. It seems that at least in this example, what you gain with deeper factor exposure is offset almost exactly by Vanguard's cost advantage.

When you do a more apples to apples performance comparison between Vanguard Small Cap Value Index Admiral (VSIAX) versus DFA Small Cap Value I (DFSVX), Vanguard outperforms over a 10 year period by 8.43% annual vs DFA at 7.92%. On the other hand, DFA has outperformed over 3 and 5 year periods 7.96% vs. 5.49% (3 year) and 14.01% vs. 11.05% (five year), this makes sense as Value has been making a comeback since the Covid crisis. So if factors do well, the DFA/Avantis products should outperform the Style Indexes; if the factors don't do so well, the Style Index products will do better than the DFA/Avantis type of funds. This is about what one should expect.