[Editor's note: ETF.com 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: Howdy folks, welcome to ETF.com Live!
I can see we have some BIG MEATY questions this week, so let's jump right in. I will type as fast as I can.
420Investor: Is MJ safe from an operational perspective? Are US Bank and NYSE standing behind it? I don't want a big tax bill if it has to liquidate early.
Dave Nadig: So for those playing the home game: MJ was MJX, and before that it was something entirely unrelated, and the issuer just morphed it into a cannabis ETF on the fly, which took a lot of its service providers (and investors!) by surprise.
So, I believe that IF U.S. Bank wanted out of the fund, they would have to give 90 days notice – which, given the fund became a cannabis ETF in January, puts it in the “any day now” zone.
Of course, we don't have all the contracts; it could be 120 days, which would be more like this summer. But certainly the more time that passes, the less likely there's an issue.
The reality is nobody knows for certain what the regulatory status of this thing is. The custodian could get cold feet. Heck, the exchange could get cold feet.
But honestly my bigger concern here is just the holdings profile – you get this somewhat-odd mix of licensed Canadian firms, and some tobacco and biotech.
I suspect what people really want to invest in is “hey, pot’s going to be sort of legal in half the country (by population) – how do I profit?” and I’m not convinced any of the listed products are the answer - here or in Canada.
Sean R: Do you think market makers should be compensated directly by ETF issuers to be more active in smaller ETFs?
Dave Nadig: This is a bit of a thorny subject, and since Cboe Global Markets is the parent company of ETF.com, I don’t want to get into the weeds of the market maker rebate system and the recent SEC action there. But ...
in general, the “small ETF problem” is a real thing. If you’re an upstart ETF, getting enough attention to have a decent advertised spread is a genuine, real problem, and one that issuers and exchanges have really worked to try and solve.
The reality is that the ETF ecosystem depends – enormously – on a system and players that they have essentially no control over, which is the willingness of big market makers to actually facilitate buys and sells. If you can’t incentivize them – somehow – with dollars, I’m not sure what the solution is.
Curious and Confused: Are there any advantages to having ETFs in a retirement account? ETF growth seems to be lagging behind in the retirement savings space.
Dave Nadig: Awesome question.
In your 401(k), a few key advantages of ETFs go away.
You don’t probably care about intraday liquidity, and you certainly don’t care much about tax efficiency. However, ETFs are still generally cheaper in most cases than a comparable traditional mutual fund, and I think the baked-in transparency is always a boon.
The big reason you don’t see more ETF-centric 401(k)s – although they exist, ask Schwab – is that mutual funds actually have two big advantages.
The first is fractional shares. You can buy an odd dollar amount of a mutual fund, but you have to buy a whole share of an ETF. That makes dollar cost averaging in your $500 a paycheck into 4 funds much trickier in ETFs.
The second is that mutual funds generally have 12b-1 fees, which can be used to offset the costs of running the plan.
That’s money you’re paying, but since it comes out of performance, most employees don’t know, or complain. 99.99% of ETFs don’t have 12b-1 fees, and are far too cheap to kick in 25-35 basis points for recordkeeping anyway.
(Oddly, the sector SPDRs do, but they are small).
Todd Rosenbluth - CFRA research: Increasingly ETF firms are going the self-indexing route. How can an investor know if this happening with their fund? What, if anything, differently should investors understand about these strategies than an MSCI- or S&P-based fund?
Dave Nadig: Hi Todd! Thanks for joining. Great question.
The biggest self-indexer is probably WisdomTree – the majority I believe are self-indexed.
The “how you find out” is pretty easy – they disclose it all over the place: in the prospectus, on the website pages, and so on.
(In general, I actually find most self-indexers are better at getting good documentation on things like index methodology in front of investors than the big, name-brand indexers. I've wasted so many hours chasing down index docs.)
I guess the big question is, should investors care?
With a shop like WisdomTree, you’ve got some of the smartest investment minds in the business on staff: Luc Siracusano, Jeremy Schwartz, Jeremy Siegel from Wharton, and so on. They do real, hard-core, nerdy research, which they then bake into their products through custom indexes.
The COULD certainly just outsource the actual work to MSCI or Solactive, but to some extent, why?
You can rent calculation services from those types of folks, so it’s not as if WisdomTree needs to all of a sudden staff a giant corporate actions or data team to do the job.
And since they control all the IP, they can effectively do the index part “for free,” which lets them keep their costs down, which theoretically means they can keep their fees lower than they would otherwise.
In short, I think it’s not a bad thing – you just need to do the work to decide if you like the self-index!
Chooch: Will I see approval of a non-transparent ETF in 2018 (or in my lifetime)?
Dave Nadig: Heh - i wonder the same thing myself sometimes. I have another question here about who might get the approval first.
There are several folks with their hat in the ring here. I won't go through all of them. I think what's actually happening is that the SEC is slowly driving all of the different applications toward a similar model, which means the differences between approaches will get narrower until, effectively, they all end up approved.
I've heard rumblings that with someone who really understands ETFs on the top of the food chain (Dalia Blass) that the process should speed up.
But I think I'm giving up on trying to pick the timing window.
I've been wrong for 2-3 years already!!!
Anonymous: Does wholesaling still work for ETFs? All info is widely available now, on ETF.com, for example, and other outlets. Seems advisors know where to go for info and ideas. Especially ETF users seem more early adopters and can competently find out about new and existing products. What do you see as trends in the way ETFs are distributed now to end users?
Dave Nadig: I think it absolutely 100% works.
We got a great question last week about First Trust, and the combo-platter of “so many funds something is always hot” and “great wholesalers” is their secret sauce in my opinion.
Advisors *can* dig in and do all the work. They can read ETF.com, and read prospectuses, and so on.
But advisors really have three main jobs: getting new clients, investing their money, and keeping existing clients happy. That's a wicked oversimplification, and of course GOOD advisors do a lot more than that, but bear with me here:
A good wholesaler doesn’t just say, “hey, here’s this product, please buy it” - attacking the middle problem (investing). They give advisors narratives that solve the other two problems.
Those narratives are what help get and retain clients. Maybe that sounds a bit cynical. The products still have to be fit for purpose, but assuming you’ve got three flavors of vanilla on the desk, it’s the wholesaler – and the story – that wins.
As for new trends in distribution? Obviously the robo space is for real, and more advisors will lean on tech for asset allocation, whether it's in a TAMP or some other structure.
I think most of how distribution is changing comes from that. But if you don't already, go poke at Michael Kitces' stuff. He's just the best on these kinds of discussions.
Follow up on the non-transparent active question:
Chooch: So in your opinion - which non-transparent model is the best? I have concerns around the added operational complexities of some of the filings.
And how does that even work? Doesn't Precidian have IP? So if they all get steered to the same model, then is there any IP?
Dave Nadig: Here's the thing - I think most of these structures are *clever* and I like clever. Like, the Precidian model, I find an ingenious set of solutions. To me the much bigger question is, does it actually solve an investor problem? I'm not sure it does. I think it solves an issuer problem.
If I had to make a bet, I would bet on Precidian, because I've gotten to know the actors involved, and I simply wouldn't bet *against* them on anything. But that said, their approach is a bit complex (as you mention) and thus might not get the out-of-the-gate nod, or as you point out, might get put into a "not original IP anymore" box if it all gets watered down by the SEC.
I honestly think at this point, we're just not going to know until we see it, without much warning.
Michael T. Kennedy: How long does it typically take for a firm to be granted exemptive relief to launch ETFs after filing an application with the SEC?
Dave Nadig: Boy, there are so many nuances to that question. I have heard that a super-plain-vanilla filing can get through in as little as 90 days at this point, but that doesn't mean you're done. That's just the actual relief filing (40-APP I believe is the form).
Actually getting to market? That can be a longer process, which involves half a dozen partners (exchanges, custodians, and so on).
The complexities are the reason we have firms like ETFS and ETFMG, who sort of rent the infrastructure baked in.
Nate Geraci: There’s a significant dislocation in the junk bond market. Give us the play-by-play. What happens to junk bond ETFs? What happens to junk bond mutual funds?
Dave Nadig: BIIIG question.
The important thing to remember is that no matter what happens in the markets, nobody suspends the law of supply and demand: if everyone wants to sell, things go down.
The ETF angle is just about “how.”
So imagine we have a rout on junk bonds because of event X – some sort of default or policy surprise. You can assume everyone will want to sell everything related – so you’ll see selling pressure on individual bonds, junk bond ETFs, and redemption orders into junk mutual funds.
Now, something like JNK or HYG is orders-of-magnitude more liquid than any individual corporate bond, so you’d expect them to react quickly and dramatically to event X.
You as an investor can choose to participate or ignore - you can buy low! You can panic and dump it all at the bottom. Your call.
Say they trade down 20%, boom, in seconds. So what happens in the underlying?
Well, the intraday NAV of the ETFs is based on a pricing service, which looks at all sorts of inputs – not just the last traded price of individual junk bonds.
So while the ETF is down 20%, the signals for “fair value” are just slowly starting to trickle into that INAV calculation.
A smart authorized participant with a good bond desk might see that 20% down and see, simultaneously, bids on the basket of bonds of “only” down 15%.
They can step in, be a buyer of the ETF at down-20, a seller of the bonds at down-15, and book the 5% arbitrage.
This will serve to BOTH bring the price of the ETF up from down-20 and push the price of the bonds further down. That’s exactly what’s supposed to happen.
But in a crisis, these gaps can get large, and persist for some time as the market starts processing the information.
But the reality is, price discovery is happening in the ETF first, and the bond market second. We’ve seen it happen over and over again since the early 2000s.
Now what about, say, a traditional fund that owns the same bonds?
Well, obviously at like 2 pm, nothing happens - all the fund can do is take orders. They're not participating in pricing.
If I put in a big redemption at 2 pm, I have NO IDEA what price I will be getting out at.
At 4 pm, the fund accountant will calculate an NAV, based on the trading prices of the bonds, based on pricing services for the bonds that didn't trade, and so on.
If ENOUGH people put in their orders, the fund may not have enough cash on hand to make redemptions.
They'll have to borrow overnight (they have lines of credit) to meet redemptions (which all shareholders will pay for).
Then tomorrow morning (really, tonight) the desk will frantically try and figure out how to unload enough to generate the off-setting cash.
They will inevitably end up dumping their most liquid, least risky bonds.
You can see how this could create a problem. And in fact, this is PRECISELY what happened to Third Avenue mutual funds a few years ago, where they had to close the funds for redemption.
In the ETF, no matter what, at least you could get a price! You may not like the price, but you KNOW the price.
(Phew, my fingers hurt. Time for a few more ...)
Mickey: The US leads the world in terms of ETF assets. Do you see other regions catching up?
Dave Nadig: Short answer - probably not.
The U.S. has an investor culture that's fairly unique. European ETFs are dominated still by institutional trading, and investors with some wealth tend to rely very much still on point of distribution - generally big banks.
Those banks control the products.
Those kind of quasi-captive systems make it hard for a real retail movement to take off. It's happening - but it's just slow.
We've seen a bit more catch in the UK market.
But it's also worth pointing out that the regulatory regimes are complex. Not that they're simple here. It's just hard to build a fast-growing, pan-European complex.
Outside of Europe - in Asia, ETFs are mostly an afterthought. If you remove the buying from the BoJ, there's very little focus from Japanese investors. Similar captive distro issues there as well.
It's not that it couldn't happen, it's just a lot of cultural and regulatory hurdles.
ETF Newbie: What niches do you see so far untapped in potential new ETFs?
Dave Nadig: I’m still shocked we don’t have real credit default swap products, or explicit spread products (we talked about those a few sessions ago).
I think there are interesting and arcane things that could be done in fixed income.
I'm surprised how few high-conviction active management products we have as well.
Last, I think we need – and will see – a lot more approaches on the ESG front, particularly impact-focused products. Kind of like what UBS has done with PRID and HONR.
Hugh Syme: Does it matter what kind of trade order I use for an ETF?
Dave Nadig: An easy one. YES!
The only people who should ever use a market order are investors who care more about speed than price. I don't really know any of them, but I'm sure they exist.
If you care about price, you should ALWAYS use a limit order. If you want to be super sure you get an execution, you can use a marketable limit.
What that means is if the ETF is bid 100-101, you could put in your buy order at 101, or even at 101.25. You're essentially telling the market, "hey, I'll take the market price right now, but just in case, don't fill this if between when I it "enter" and when it hits the book, the price EXPLODES."
The way your broker deals with that order is they internally view it as a market order, and will get you the next available fill. But you're protected from the rare case of a flash crash in either direction.
Ryan Hessenthaler: A number of firms go to S&P, Solactive, or others with investment strategies and have them create and calculate an index. As the fee war continues, do you expect custom index providers (i.e., those who use S&P or Solactive or others as a calculation agent) to be squeezed further and further in terms of basis points compensation from ETF providers?
Dave Nadig: Great question. There's no question that prices are really only going one way, and that's down.
It used to be, for instance, that if you had a great idea for an index, you went to one of these firms, paid a decent chunk of change up front to develop it and run backtests and so on, and then paid them a basis point fee on top of that for any assets that target against it. Maybe there was an additional negotiation about using the brand as well.
Now, from what I hear, you can go in with a flat fee - pay X dollars a year, period, and do whatever you like with the index, if you negotiate it well.
That's a huge shift, and I think you'll continue to see far more "fee for service" arrangements as the bpS get squeezed out of every corner of investment management (see question above around self-indexing - same thing applies!).
OK, last question - another biggie.
Concerned Investor: I've read about when ETFs lend out the securities they own to short sellers. Is this legal/ethical?
Dave Nadig: Many ETFs – and mutual funds, and separate accounts, etc. – do lend out their securities.
In general, this is a good thing. When some hedge fund borrows the shares of AAPL that are locked up inside, say, IVV, they put up collateral – generally 104% of the value of the shares.
That collateral is trued-up daily, so essentially the fund always has excess.
Depending on how desirable the stock is for short-sellers, there's also a fee of some sort baked into the deal, sometimes as high as a few percent, annualized.
There’s also some small overnight-level interest on the collateral itself.
That additional revenue accrues in part or in whole (depending on the issuer) to the fund's shareholders. You can actually see the split (what you get vs. what the issuer keeps) on the ETF.com fund pages.
Overall, it’s a small number, maybe 20 basis points on the MOST aggressive end of things, maybe a basis point or 2 more normally. In terms of its impact on the fund as a whole, obviously funds with hard-to-borrow stocks (think solar stocks a few years ago) can make a big number.
In the end, it all has the effect of helping offset the costs of the fund, so it’s a good thing.
So - is it risky? Well, the person borrowing the stock has to welch on the deal (going bankrupt) and never give you the stock back.
And then, the 104% collateral has to somehow not be enough to buy the stock back in the market.
So you have to have two things go wrong at once - a huge intraday run in a stock, and a default.
I’ve never found a single instance where a '40 Act fund or ETF has had shareholder-borne losses from someone borrowing a stock, and then magically going bankrupt the same day and the collateral not covering the loss.
I actually put this out on Twitter a few months ago, and NOBODY could come up with a historical example.
It’s theoretically possible, but so are asteroid strikes in Central Park.
OK, that's going to wrap it up for this week. We've settled on 3 pm Thursdays going forward, and you can always ask your questions starting at 9-10 am on the Thursday just before one of these sessions. We'll have an edited and prettified transcript up shortly.
Thanks for all the questions, and everyone have a great day.