Surprise Exposures & Why Your NAV Is Wrong

April 19, 2018

[Editor's note: Live! with Managing Director Dave Nadig happens Thursdays at 3:00 p.m. ET, with the question window available two hours before and during.]


Dave Nadig: Hey folks! Welcome to another session of Live! Feel free to keep asking me questions, and I'll get to as many as I can before my tennis elbow makes me quit.
As a special note: Next Thursday, our Editor-in-Chief Drew Voros and I are doing a webinar that's just straight up ETF 101, with a live Q&A on video.
You can register for that here:
And as usual, don't worry if you miss something, we'll post a transcript of this session shortly after we wrap up.
With that, let's get started.

Bill Donahue: Dave, thanks for continuing to do these ETF Live sessions. The SEC released a series of proposals (1,000 pages) yesterday that I will collectively refer to as SEC Fiduciary Rule, which includes more disclosures around broker-dealer conflicts of interest, disclosures about services provided by investment advisors and broker-dealers, as well as a proposed fiduciary standard for advisors. One main criticism of these proposals is that they do not define "best interest." What is your preliminary viewpoint of these proposals and how they might impact ETFs should they be adopted? Thanks.
Dave Nadig: Hi Bill! Big meaty question to kick off with.
I feel pretty strongly in universal standards and level playing fields, so I applaud the SEC for wading into what can be a real briar patch.
As much as I have a reputation for being a regulation wonk, I have not read every one of the 1,000 pages:
But I've skimmed it and read the summaries from the poor reporters who stayed up all night. Importantly, the actual RULE is only 2 pages:
(Yes, I had those in my browser history. I'm a big hit at parties.)
The criticism that the rule itself is very, very vague I think is sound criticism.
I suspect the comments received from the industry will all talk about what the heck "best interest" really means.
But here's my bottom line: I'm not sure the proposed rule change makes a huge difference as it's sitting now.
The FIdicuary Rule, as written, I think would have and actually did drive some incremental assets.
But I don't think this one would have the same effect.
Worth nothing though: The SEC also suggested it's going to only allow the term "advisor/er" to be used by people who are operating as fiduciaries. That's a nice clarification.
But end of the day: This is just watered down.


Todd Rosenbluth - CFRA research: On, there are tools to understand how well an ETF historically tracks its benchmark. What happens when the index changes, like with MJ, and Xtrackers single-country ETF? First Trust plans to convert an ETF to an India ETF, I saw on
Dave Nadig: Hi Todd, welcome back!
Man, what a nightmare, right? The "reuse" of ETFs is a major pain, and not just for data providers.
The data on is primarily sourced from FactSet and Elisabeth Kashner's team there (FactSet bought it from years ago, so obviously we like it!)
The rule then, as I'm sure it is now, is that if the index change is "substantial," then the statistics around things like tracking reset.
So for example, Vanguard has done a bunch of changes over the years, say, from S&P indexes to CRSP indexes. It did those over a long period with transition indexes, and ultimately, large-cap US equity was still large-cap US equity.
In those cases, the indexes were mapped along with how NAVs changed, and the stats remained.
But something like MJ? Forget it. The history gets tossed out.
There's a judgment call from the FactSet team on what "substantial" means, but really it's usually pretty obvious.
There are some edge cases probably -- DXJ (WisdomTree Japan) made some pretty significant changes, but I BELIEVE keeps a long-term, sitched-together history.
The joys of being a data monkey...


Anonymous: Why does VWO fair-value its NAV but EEM does not? Is there a rule?
Dave Nadig: This is such an awesome question.
Simple answer: because VWO is a share class of the Vanguard EM mutual fund, and that's "just how those funds do it."
The complex answer is super interesting:
(grab a coffee!)
It's on the board of the fund to decide what the valuation processes are.
If you run a Japan fund in the US, well, the market in Japan is literally never open during U.S. hours.
So from 9:30 am to 4 pm, when you strike NAV, the "closing" price of the mutual fund or ETF will only move if the yen moves. The security values are just the close from last night.
Different fund accountants approach the resultant timing problem differently.
Some (like, as you suggest, VWO) will adjust the NAV at 4 pm based on an algo that looks at all sorts of proxies -- sometimes even ETF market prices, but also futures, proxy securities, ADRs, and so on.
They then adjust the NAV to make a best guess for fair value.
Probably should "air quotes" around that.
Why bother?
Imagine it's 3 pm Eastern and a tsunami hits Japan.
Well, fair-valuing prevents gaming that.
Someone can't come in and dump at 3 pm, knowing they'll get the stale, inflated closing prices of the stocks from last night.
There are actually some big examples of this happening back in the 1980s, where some funds did and some funds didn't, and there were lawsuits and everything.
So with a big mutual fund, where a lot of transacting is happening at end of day, that's a sensible move, fair valuing the NAV.
An ETF, on the other hand, is trading all the time.
The market price of an ETF would get hit when news of the event happened.
So since no money is really changing hands at NAV, there's not much point in going through the motions.
Also, most indexes also don't do any kind of fair value.
They just strike the close of the securities. So by NOT fair-valuing, they allow investors to measure apples to apples tracking difference.
The end result is that EEM, in this case, will show "real" tracking difference (very little) but "fake" premiums and discounts (closing price to NAV).
VWO will show the opposite -- very little premium or discount, but "fake" tracking difference.
And that's just about the most in-the-weeds thing I know about index and NAV valuation. Congratulations.


vic di marco: I own quite a bit of AMZA stock now and planning on purchasing more. the EX dividend date is usually the 20th of the month but I don't see an announcement as yet.
Dave Nadig: From the super big to the super small:
So AMZA is one of the Virtus MLP ETFs.
They've actually had dates ALL OVER the place, and to heck if I know why. Last year they did quarterly distributions. This year they seem to be doing them monthly.
I know they made an announcement, but my suspicion is that this has to do with when the payments come in on schedule from the underlying MLPs.
Because it's a C-corp structure, there's a weird double-tax issue that's involved too. It's worth making a phone call to make sure you really understand it. Taxing MLPs in ETFs is one of the most complex parts of the business.
And every individual fund has its own quirks.


J. Gross: Hello Dave! On a previous Q&A session, you had mentioned that most investors should allocate a small percentage of their portfolio to commodities. Can you discuss your favorite diversified commodity ETF that you would use for this purpose? In addition, given the new structure available for commodity ETFs (as a '40 Act fund), as well as commodities finally performing better after a long period of underperformance, do you expect a new wave of commodity ETFs to be issued?
Dave Nadig: So, our next issue of ETFR is on commodities, and I just wrote about the challenges here.
I could cop out and say I love all my darlings equally ...
The truth is there are a bunch of very solid approaches here.
But nobody agrees on the "default" way of indexing commodities.
Commodity indexes are HARD. Because of that, I kind of like the "just equal weight it" approach, which is what GCC (now owned by WisdomTree) does.
It's not the new structure.
The bigger downside is, at 85 or 86 bps, it's expensive.
The "cheap" comparable might be something like GraniteShares' COMB, which, while technically active, basically tracks the Bloomberg Index, which caps things so you don't end up with the huge energy overweights of the GSCI-based products that have a lot of the assets.
It's got two advantages. The first is, it's shockingly cheap: 25 bps, if I recall; maybe 27? Cheap.
The second is that it's in the new structure -- in short, the "new structure" for commodity ETFs invests through a Cayman Islands subsidiary, so you don't have to run the fund as a commodties pool and distribute partnership forms every year.
It's a bit of fine-hair splitting about whether there's really much of a financial performance difference, but there's no question it makes dealing with your taxes VASTLY simpler than the old system.
As for "will this drive growth?" -- I'd love to say that investors don't chase performance, but I have a brain and two working eyeballs, so: if the performance runs, the money will follow. The new structures will make that money move a lot more easily.
Great juicy questions.


Ross: What does the future of non-transparent ETFs look like? Despite not gaining assets, ETMFs (nextshares) keep rolling out - is there a better wrapper/loophole for these products that can work? Is there any demand from investors?
Dave Nadig: We covered this a little last week, but I'll add some nuance.
We made the decision at not to "cover" ETMFs (Exchange Traded Mutual Funds) because we believe they're really just mutual funds with an interesting order processing tweak.
(and we're, not
But that said: There's some cleverness behind the structure (Todd Braum and Gary Gastineau developed it ages ago, I believe), but I remain unconvinced it's solving a problem *investors* actually have.
So we've seen some adoption by issuers, but not a lot of assets.
That may end up being true when the nontransparent active ETFs -- the true ETFs -- are eventually approved by the SEC.
One of the various structures will, I believe, eventually be approved. But are investors lined up for it? I think that's a big if. No crystal ball here -- it'll be grab-the-popcorn time. But I'm pretty skeptical, short of a GIANT brand name (like contrafund or magellan or something).


Chuck Mangione: How come ETF market commentators focus so much on a fund's trading volume when liquidity is driven by the underlying stocks? Even focuses on underlying liquidity in their analyses, yet every analyst just preaches onscreen volume. Its not that on-screen volume is unimportant but it doesn't tell the whole story. Advisors learning more about ETF trading for their clients would seem to be getting incomplete information unless they keep digging.
Dave Nadig: So this is a really great question too. To clear it up a bit:
On-screen volume = what you see when you enter "SPY" into your Schwab account. you see how much it's traded, or how much is quoted at different prices.
Underlying liquidity = the same thing, but for each stock in the S&P 500, which is what the SPY holds, after all.
The short answer is that both matter, but for different reasons.
Let's put asside the enormously liquid ETFs that trade like water. Nobody has to worry about SPY or EEM or anything.
And put asside the very, very smallest, forgotten $1million ETFs that haven't traded since I had hair.
In the middle, both on-screen and underlying liquidity matter, depending on what kind of investor you are.
If you are trying to put 500 shares through on an ETF, well, the only liquidity that will matter for you is the superficial, open-market liquidity. If it's trading 5% wide (like, bid 100, ask 105), that's just awful.
You can try and game it (putting your order in at 102.50) and hope, but really, the lack of on-screen liquidity CAN make your life hard.
But if you're an institution looking at that same fund looking to buy 50,000 shares, you don't care much about that.
Because you can ask your broker to help make new shares for you. And in that case, all you really should care about is the underlying liquidity. If it's large-cap U.S.? They'll MAKE you 50K shares at 102.50.
But if it's, I dunno, Canadian microcaps, maybe not? Maybe that spread is real there too.
So both CAN matter.
Hope that helps. Next one up ....


Wes: Why wouldn't you go with an ETF wrapper? I.e., choose mutual fund format?
Dave Nadig: See above! The ETF wrapper brings with it a bunch of advantages (liquidity, transparency, tax efficiency, etc.) but introduces "trading."
If you don't want to deal with all of the above, and placing orders and worrying about fair value, an apples-to-apples mutual fund could be just fine. The trick is in the "apples to apples."
For example, the Vanguard ETFs are generally the same price, and the same portfolios as the mutual funds (they're just share classes).
PIMCO's BOND is pretty much the same whether you buy the institutional mutual fund share class or the ETFs.
There's an undeniable convenience to the mutal fund wrapper. 
And there's also the fractional share issue, which we covered a bunch last week - you can't (easily) own half a share of an ETF in your 401(k). It's child's play in a mutual fund.
So there are a few cases where it makes sense, I think.
Few more here, and then we'll wrap up.


Gary Curtis: Should I have more than one Emerging Markets ETF in my portfolio? Do they overlap in coverage? Advantage?
Dave Nadig: In general, I don't quite get the "own two of every animal" version of investing, but I see a lot of people do it.
If you look at the reports from even some large institutions, they'll own IVV and SPY -- two funds with the exact same portfolios, minimal cost difference, etc.
When I've asked, the answers I get from them are things like "well, if one of them has a blowup, the other won't."
Which is a bit of a belts-and-suspenders kind of thing. I can see the logic, I just don't actually believe in it.
When you get into things less direct - like, say, the difference between EEM and VWO, the two big EM funds, there it's much gnarlier.
There are REAL differences in approach between the two. One has South Korea, for example; the other doesn't.
So. mixing the two really pollutes your exposure. As an investor, sadly, you actually have to decide for yourself whether you want SK in your portfolio.
And then pick one.
So in general, nah, I don't think it makes sense, and can actually create problems.


Not An ETF Mechanic: One of things I always hear is that you "should check under the hood" of what ETF you buy. But if one uses advisors, isn't that their job to tell me, a client, what is under the hood?
Dave Nadig: Yep! Certainly one of the reasons investors work with an FA is precisely to NOT think about these things.
Even if you're just using a roboadvisor, like, say Betterment or something -- you're counting on them to have done the looking under the hood for you.
However, having said that, I still think there's value in knowing what you own, because your FA doesn't know, necessarily, everything about you.
For instance, maybe you work at a semiconductor firm and get company stock. Well, maybe you want to lighten up on your tech exposure, since you're already career-invested there.
Betterment -- or your mom-and-pop advisor, won't necessarily get that.

OK, folks, that's going to do it for me.

Thanks for joining us this week. We'll have a transcript up shortly. Spread the word, and I hope you'll join me every Thursday at 3 pm Eastern. Also, as I mentioned at the top, as a special note: Next Thursday, our Editor-in-Chief Drew Voros and I are doing a webinar that's just straight up ETF 101, with a live Q&A on video. You can register for that here:

See you next week!

Find your next ETF