ETF Tradability: A Primer

Understanding how to trade ETFs means understanding a number of crucial metrics.

Reviewed by: Dave Nadig
Edited by: Dave Nadig

Understanding how to trade ETFs means understanding a number of crucial metrics.

The “T” in ETF often throws investors. After all, it’s the exchange-traded part that creates situations like the wacky premium Paul Britt noticed in MLP ETNs last week. It’s what everyone worried about in the “flash crash.” And if you’re used to investing in mutual funds, not stocks, it’s the scary new part of ETFs.

But it doesn’t have to be. That’s really the reason we have an entire tab devoted to Tradability in ETF Analytics—to demystify what’s really a pretty simple transaction.

The first thing you see when you go to any ETF (I’m using the iShares MSCI Emerging Markets ETF (EEM | B-98) here as the example) is the score:


That score gives you a rough guide for how well the ETF trades relative to the entire universe of ETFs. It rolls up all of our trading inputs, from simple on-screen volumes to much-more-finicky bits about how the underlying securities in the portfolio trade. To see how that “86” for EEM compares with the segment, you can look to the overall scoring graphic on every EEM page:


The orange line is where the average ETF in the segment—in this case, Total Market Emerging Markets Equity ETFs—trades. You can see that EEM does better than average not just in Tradability but in its Efficiency and Fit as well. That shouldn’t come as too much of a surprise, as it’s one of the largest and most successful ETFs on the market.

But what’s actually going on under the hood regarding that “86” rating in tradability?

Underlying Liquidity

Broadly speaking, we look at two things in our tradability methodology: measurements of real-world on-screen liquidity; and measurements of underlying liquidity.

Put another way, we look at how the ETF itself trades, and how what the ETF owns trades. The reason is simple: If you’re going into the market to buy a relatively small number of shares, all you’re really concerned with is how the ETF itself is trading.

If, however, you’re buying a very large number of shares—at least, relative to how much of the ETF trades on a normal day)—your experience is likely to be dictated by how well the securities the ETF holds trade as much as how well the ETF trades. That’s why we look at both.

On the right-hand side of the page, we break these measurements down in a number of ways:


This first block purely shows you how the ETF trades. It seems obvious, but there are a few nuances here. The first is that we look at median as well as average trading days. This matters because it’s not uncommon to find ETFs that trade very little for weeks at a time, and then trade 100,000 shares in one trade.


Volume Metrics

That large single trade usually signals an institution working a creation or redemption of a large, single block with an authorized participant (AP). That spikey volume is masked by averages, so we tend to focus on median here. The median answers the question, On a normal day, what does trading look like?

The last stats in this block are spreads. Importantly, we’re looking at both the percent spread and the price spread. The percent is by far the most important—after all, a 10-cent-wide spread on a $10 ETF is a whole lot more painful than a 10-cent-wide spread on a $100 ETF.

These stats can be very hard to find reliably online. We calculate ours by looking at the quoted bid/ask spread in one-minute intervals. We scrub out outlier periods, such as the market open and close. We also scrub out data that our algorithm assumes are wrong, like a “flash” pop of a bid that looks like a decimal slipped.

We also present spread as a chart, so you can see how the ETF reacts to different kinds of market conditions:


It’s critically important when looking at these charts (or any chart, really) to understand the scale. What looks like a fever-chart here is actually a fluctuation in spreads of less than a basis point; in other words, hardly cause for concern.

The second block on the right looks just at things that can impact how the ETF creation/redemption process works:


Here we’re keeping track of any issues that might affect a smooth creation and redemption of the ETF shares.

Any increased risks or costs will get passed on to investors in the form of a wider spread. A smooth creation/redemption process is what lets APs keep ETFs trading efficiently, and close to a fair value.


Watching APs Create And Redeem

“Market hours overlap” is important because if APs want to do creations, they need to buy underlying securities, and if they’re not available, they’ll have to wait a day to buy them, which increases risk. The size of the creation unit relative to how much the ETF trades shows how well an AP can fine-tune exposure and minimize risks. That way, costs are simply costs and, eventually, investors will pay them.

One often-missed data point here is the “underlying volume/unit” metric. That stat answers the question, If APs go out and buy up all the securities they need to make a creation basket, how much of the market are the APs going to be that day? The answer in this case—41 basis points—is tiny.

There’s little chance a big trade will move the underlying stocks involved. In more niche markets—say, small-caps or frontier markets—the numbers can be quite large, and should give large investors pause before they mash the trade button.

Finally, the “implied liquidity” metric rolls up these and other implied liquidity metrics into a simple 1 to 5 score, which we also present as a graph:


Premiums And Discounts

That’s really all there is to it—except for one thing that leaps off the page: premiums and discounts. We present premiums and discounts both in the right column of data within the "Tradability" tab as rolling 12-month "Median" premiums and discounts, and the full range measured in percentages in that period. We also present them as a chart like this:




The problem with premiums and discounts is that, most of the time, they lead you down a path that isn’t very useful. In this case, 80 percent of the stocks EEM owns are closed for trading during the U.S. trading day. That’s what that “market hours overlap” measurement showed us above.

When you measure whether EEM is trading at a premium or discount on any given day, you have to look at the closing price of EEM versus its advertised net asset value (NAV)—that’s what premiums and discounts are.

See the problem? If 80 percent of EEMs stocks aren’t trading, but EEM is trading, is it right to even call the deviations between the two numbers a “premium” or a “discount?” You can see from the chart that for EEM, the premium/discount centers around zero. EEM will respond to news while the underlying stocks can’t, so it will naturally, and always, trade off of its stale NAV.

So why do we bother presenting the data if we don’t think it’s useful? Well, we do think it’s useful, but only occasionally. Certainly in cases where there’s 100 percent market-hours overlap, like with U.S. equities, measuring how the fund trades versus fair value is valid. But even then, we worry people think they can just buy at a discount and sell at a premium as some sort of secret trading sauce. It’s rarely that simple.

So we decided to present the data—especially the chart—so you can eyeball for any kind of persistent premiums/discounts that might indicate something funny is going on, but we don’t algorithmically score a fund for it.

A Personal Decision

At the end of the day, the decision to pull the trigger on trading an ETF is always going to be yours, and it’s always going to involve some level of judgment on your part, whether that’s about on-screen volume; where to set your limit order; or whether to ignore the premiums and discounts.

We hope the “T” tab on each ETF page can help you make those decisions.

At the time this article was written, the author held no positions in the security mentioned. Contact Dave Nadig at [email protected], or follow him on Twitter @DaveNadig.



Prior to becoming chief investment officer and director of research at ETF Trends, Dave Nadig was managing director of Previously, he was director of ETFs at FactSet Research Systems. Before that, as managing director at BGI, Nadig helped design some of the first ETFs. As co-founder of Cerulli Associates, he conducted some of the earliest research on fee-only financial advisors and the rise of indexing.