Paul Justice: Hi. I am Paul Justice of Morningstar, the director of our ETF research in North America.
Today, we're going to talk about the ETF structure, and when it works and when it might not, and some of the challenges that you might face.
Today, I am joined by Sam Lee, one of our ETF analysts, who has been evaluating the structure question and really digging deep in understanding what ETFs are going to work well for people and when you might want to look at maybe a mutual fund, a closed-end fund, or the underlying securities in particular.
So, if you don't mind, let's jump right off and really talk about, what are some of the attributes of underlying securities that makes an ETF work well?
Samuel Lee: Well, the first are the basics. The ETF has to deal with the securities that trade in deep and liquid markets, and what we mean by deep markets, we mean there is a lot of volume in there, a lot of active players, who are making sure that the price of the underlying securities are somewhat fair. So, the ETF doesn't actually influence the prices.
Justice: So, like large caps, domestic stocks will be a great starting point where the ETF...
Lee: International stocks, Treasury bonds, those are perfect for ETFs.
Justice: So, we get a lot of trading volume in there. It makes it easier for some of the institutions to go ahead and do the creations, redemptions necessary, really reflects the liquidity of the ETF. So, it could be a small ETF, but if you got the liquid underlying, it's probably going to work pretty well.
Lee: It's going to work excellent. So, even if it's $5 million in an ETF, if it deals with U.S. Treasuries, it's still going to be really good if you are simply entering your purchase orders with limit orders.
Justice: Pretty close to whatever the fair value is listed by the indicative value index published every 15 seconds.
Justice: Okay. So what are some areas that we should be concerned about, because we've got nearly 1,200 exchange traded products today, and seemingly new asset classes introduced all the time. What do you think are some of the ones that could face challenges down the road?
Lee: So these are not really a big part of the ETF market, thankfully, but they are ETFs that deal with less liquid, less trafficked securities, so on big example is micro-caps.
Justice: So, there is no market, wide bid-ask spreads…
Lee: Wide bid-ask spreads, not very many market players, so the ETF can actually influence the prices of the securities.
Justice: So, what's the general investor experience, then, in those funds?
Lee: It's been absolutely horrible. Micro-cap ETFs have had negative alphas or negative excess returns of anywhere from negative 3 percentage points annualized to negative 8 percentage points annualized.
Justice: Wow, that pretty incredible.
Lee: So that's extremely bad
Justice: So, if we're looking at studies, we could look and say, "micro-caps, you know the smaller caps can perform very well, especially illiquid securities tend to perform very well on a theoretical basis.
Lee: Yes on a theoretical basis, but ETFs are not the right vehicle to access those, because ETFs are basically index trackers, and index trackers buy and sell mechanically. They do not care about what price the underlying securities are bought or sold at. When you have those kinds of issues, it can lead to substantial deviations from what a fair market would create.
Justice: So, in some of these illiquid cases you might be better off in an open-end structure where the manager has much more discretion and the ability to carry some cash.
Justice: Or even in closed-end funds.
Lee: Exactly. What's great about the open-end fund structure where there is an active manager, or even a closed-end fund structure, is that the manager can play the role of a market maker. So, the investors in those funds can actually reap excess returns simply by lending out shares or making markets in these illiquid markets.
Justice: I think oftentimes the challenge for investors is that if they run a screen and they look at these comparable funds, oftentimes that expense ratio in the ETF is going to be much lower than some of the alternatives.
Justice: ... which, if you get the good execution and it can track well, that's going to be real saving for the investor and help performance out over the long term, but sometimes you are going to pay for what you get.
Justice: I think that's something that people need to pay attention to.
Lee: So, yes, absolutely a case of false economy to look at the expense ratio for these illiquid securities, when it might make sense to pay up for active management.
Justice: Now you mentioned before a great example of ETFs that perform well, and definitely some of the ones that have seen the most inflows have been international-stock focused ETFs, particularly emerging markets.
Now, the challenge is that we say we want a liquid market, but those markets aren't even open when the New York Stock Exchange is, when I'm going to be buying my ETF. So you can't really say that that's a liquid market at all points in time. Why would you give that, qualify that those are actually efficient markets and work really well on the ETF structure?
Lee: Well, I would actually say that they are highly liquid markets because their underlyings tend to trade extremely actively during trading hours and during the hours that the ETF trades, there is also a lot of volume there. So, the ETF, even if it trades at premiums and discounts, tends to reflect the best information about those underlying stocks.
Justice: So they are price discovery vehicle?
Lee: Yes, exactly, they are price discovery vehicles, and a lot of times they are better indicators of the true value of the underlyings rather than the NAV because the NAV is simply the…
Justice: ... the net asset value ...
Lee: Yes. The net asset value is simply determined by the closing price of those securities with some currency adjustments. So, if there is news that happens overnight, the ETF's market price can reflect all of the best information about it. So, what you see is that the NAV tends to converge on the market price the next day.
Justice: You ran some numbers showing how predictive, actually, if there is a premium or a discount, it's a very high R-squared or predictive power of premium saying what the index is going to do the next day, really highlighting some of the inefficiencies of today's indexes when you are tracking things that aren't trading at any point in time?
Justice: Well, all great stuff. Thanks for joining me today. Hopefully, people have gleaned a little information on what ETFs can and can't work.