Last week I described how an ETF’s market price and net asset value (NAV) can diverge. Now it’s time to look at the three main reasons why this occurs. Our guides for this discussion—which I’ll warn you is quite technical—are Pat Chiefalo, director of ETF research and strategy for National Bank Financial, and Steven Leong, vice-president at BlackRock Asset Management, provider of iShares ETFs.
1. The ETF has low trading volume. In most cases, low trading volume is not a major concern with ETFs. However, thinly traded funds are more likely to display short-term differences between their market price and their NAV.
Official end-of-day prices on the TSX are based on the last board lot to be traded in a single order. There may be some days when all the orders for an ETF are less than 100 shares. “In that case, there would actually be no official end-of-day price, even though there were transactions,” says Steven Leong. “Less extreme would be a case where the last board lot traded at 1:30 in the afternoon and the market moved on from there.” When the NAV is calculated at 4 pm that day, it would be quite different from the last transacted price, which is now two-and-a-half hours old.
This is also why you should always ignore the “last” price in an ETF quote. When you place a limit order to buy, use the ask price as your reference point. If you’re selling, base your limit order on the bid price. “It’s frankly unnatural for us to consider the last price as the benchmark for an ETF trade,” Leong says, “although I know a lot of people grow up trading like that.”
2. The underlying securities are illiquid. Sometimes the problem is not that the ETF is thinly traded: it’s that its underlying holdings are illiquid. “You can get into a situation where the prices of the stocks and bonds used to calculate the NAV are stale,” says Pat Chiefalo. “If a stock, or a preferred, or a bond has not traded in a few days, the net asset value will tend to keep the price of the last trade.” In this scenario, it’s the market price, not the NAV, that is likely to reflect the true value of the ETF’s holdings.
Leong says this is most likely to happen with fixed income ETFs, because bonds are traded over the counter. “Determining the NAV is based on a series of rules and a third-party pricing source,” he says. However, Leong says this only happens when an ETF’s trading volume is dramatically higher than that of the bonds themselves, and that rarely occurs in Canada. “It is certainly a phenomenon that can happen with a handful of US ETFs, but I’m not sure it’s terribly prevalent here.”
3. Time zone differences. International equity ETFs hold stocks that are traded in European and Asian markets, which are usually closed when North American exchanges are open. (European markets overlap briefly with Toronto and New York.) So when market makers are posting live bid and ask prices for those ETFs, the prices of the underlying securities are likely to be stale because they’re not trading.
Chiefalo explains the market makers may use American Depositary Receipts as proxies (ADRs trade in New York and represent the shares of overseas companies). “But when the underlying basket has very few ADRs, then the traders make their best estimate. So if there is a lot of market movement compared to the previous day, the NAV and market price are going to be very different. The price is actually going to reflect a fairer value than a stale NAV from the night before.”
These time zone differences can also cause an international ETF’s tracking error to appear unusually large, an issue I wrote about in an earlier post. For a more up-to-date discussion, see this recent article from Vanguard.
Which price is right?
So if you want to know how your ETF is performing, should you look at the market price or the NAV? “I don’t know if there is one universal way to do it,” Chiefalo admits. “I think by default people go to the NAV, but you should appreciate that discrepancies may happen and it may not necessarily indicate something is wrong.”
Leong agrees there’s no easy answer. “You shouldn’t look at one number and ignore the other: they are both part of the story. If you’re a long-term investor, the NAV is really the right thing to look at, because that is showing you how the economic value of the portfolio is changing. But the specific prices you transact at are going to dictate your own performance.”
If you’re holding your ETFs for the long haul, it ultimately makes little difference which performance number you use for individual calendar years. But you should follow some simple rules whenever you trade:
- Always use limit orders: no exceptions. Ignore the last price and place your order within a couple of cents of the bid (if you’re selling) or ask (if you’re buying).
- Don’t place an order during the first few minutes or last few minutes of the trading day: differences between price and NAV are likely to be widest at these times. And never place an order during off-market hours expecting it to be filled at the market open.
- If you’re making a large trade in an international equity ETF, consider placing the order before 10:30 am EST, while the European markets are open and price/NAV discrepancies are likely to be lowest.
It’s really wonderful that you’re explaining this stuff. But wow, is there ever a lot to keep track of! It’s probably the greatest barrier to actually starting with investing in ETFs; much as I like the concept, I wish that I could just select a plan and throw $1000/month at it for the next forty years without having to be so concerned about the mechanics of everything. Does anything like this exist, or is there even a ‘getting started’ guide somewhere that provides specific instructions (e.g. using an online brokerage, managing rebalancing) on how to manage everything?
@Joe: A lot of keen DIY investors are interested in these things, and it’s hard to get good information about these technical subjects in the mainstream media. But you certainly don’t have to worry about these kinds of things if you’re a long-term ETF investor who rarely trades.
I was in a similar boat a couple of years ago Joe, but took the plunge and became a full-fledged potato head and haven’t looked back. I’m more than pleased with the results based on my individual situation, and I don’t worry about the technical aspects to my portfolio – just focus on the big picture. Thanks again Dan for your insight!
Joe Clark: I think it’s important to understand that Dan likes to go into the nitty-gritty, but that such a detail level isn’t required for most people.
It’s like bid-ask spreads: for a frequent trader, they can be important, but if you intend to buy and hold with a holding period in decades, such minor elements don’t account for much.
I’ve learned to skim over certain posts, which go into far more detail than I need to know. (no offense to Dan, whose blog remains my #1 reference for all matters relating to investing)
What is the best source of the NAVs to use when one is making any given transaction?
@Paul G: What!? You skim some of my dense, jargon-filled, 1,000-word blog posts? OK, I don’t blame you. :)
@Andrew: I would never worry about trying to determine an ETF’s intraday NAV: you’re likely to be working with inaccurate figures that will lead to a bad decision. See this article from Rick Ferri:
http://www.rickferri.com/blog/strategy/etfs-and-the-uselessness-of-inav
Again, excellent advice. Even when just skimming the article, every ETF investor should just remember the last three bullets. If I had known and followed that on a trade some months ago, I wouldn’t have been burned for a couple of $100.
Great articel! Yes the three last bullets are worth remembering.
And by the way, this is the kind of article I do not skim over, I find them interesting… what does that say about me!
In contrast to Paul G. I do skim over some of the more basic articles though since I have been following the blog for a while. I guess it shows how diverse this blog’s audience is.
@Philippe: “I guess it shows how diverse this blog’s audience is.” Or it goes to show that, no matter what I write, it’s bound to be boring for a large number of people. :)
Plus, there are only so many blog posts you can write about the basics. Those are definitely the ones that are most helpful to the largest number of people though. Hmm.. I wonder if next time you revamp the layout it might make sense to redesign the category listing a bit. Right now there are 20 different categories, so a person might look through there if they were interested in a particular topic, but newbies won’t be drawn to it. Done right, that could be a nice middle-ground between the basics of the FAQ and About pages, and the nitty-gritty of the day to day blog. You could highlight one or two “beginners'” categories, so there would be an obvious place for people to start off to get the key stuff down (once they’re ready for more than the FAQ level of info).
Anyway, keep up the good work! I wasn’t aware of your e-book before reading that comment 2 minutes ago, and have already recommended it to a couple people.
In fact, to start with, you could probably create a category called “Beginners”, and just make it bold and stick it at the top of, or just above, the main category listing, as well as linking there from the FAQ and such. (Since the Cat list is alphabetical, I guess the most logical would be an additional little section above it where that category is highlighted on its own.) I bet that alone would help a lot of new visitors.
Oh, and while I’m doling out advice, here’s something else I’ve thought of in the past, and considering the category thing just brought back up. Obviously you’re welcome to do with it what you will, or ignore it entirely! For me personally, the site would be easier to parse if some of the top navigation links were combined. There is a lot to explore there, and I can see it being overwhelming, especially for new users. For example, the Index Funds, ETFs, and Model Portfolios sections could be combined. So could some others, but you would obviously have to consider the tradeoffs. IE: FAQ & Resources, Advisors & DIY, About & Contact.
If you can get the area up there down to say three or four links, it would give people a really clear place to start. You could then even ignore my previous suggestion and just put the extra link to the beginners’ category right in the FAQ and About pages, keeping an eye on your Google Analytics to see how new visitors’ behaviour changes. My bet is that it will help them stand out more, resulting in more visits to those key pages (FAQ being the most key IMO) by new visitors.
Anyway, just in case any of that helps. Site is great as-is too. :)
Hi Dan
I was just wondering when you compare a mutual fund to a ETF on a graph, does the graph take into account the MER’s or do you have to figure it in yourself?
Thanks
@Ed: Mutual fund and ETF returns are always reported net of fees, so you never have to subtract MERs.
Hi Dan,
I didn’t follow this part:
“And never place a limit order during off-market hours expecting it to be filled at the market open.”
What’s the risk there?
@Patrick: the reason is that the price people are willing to pay changes based on news overnight, but the quote you see is based on the last action the day before. If there is still a bid and ask quoted, they will represent after hours trading, which has a larger spread. This can cause the price to “gap” up or down at the beginning of the day. Depending on which way it moves, your limit buy order could either go unfilled, because the bid and ask prices have gone up overnight, or could essentially act like a market order, because they’ve gone down.
I feel Nathan has prompted me to speak up. I love your site and have taken the time to read through all the entries. First, congratulations on the achievement, but also congratulations on your patience–so often answering questions which were so clearly laid out in some earlier post. I know you are probably crazy busy with your new project but what I would like to see is a book which pulls it all together. I know you already have one–Money Sense Guide to the Perfect Portfolio– which I have read. (By the way, my 30 year old son has just started an RRSP and I tried to pick up a copy of the latest addition at Chapters today to send to him but no luck. Any suggestions?) Back to my point, I think you have a kick-ass book lurking deep in this site. I can imagine something that goes a bit deeper than your current one. One that includes some of the qualitiy debate that your entries provoked. An example would be the debate that followed your comparison of the Permanent Portfolio and the Couch Potatoe Portfolio–including Justin Bender’s contribution etc. Also your evolving distancing from ‘tax advantaged’ etf’s. And the debate with dividend investors. Your entries on taxable accounts vs. sheltered accounts are illuminating. Or again how bond yields are misunderstood. It’s easy for me to say you should write a book but I appreciate your measured tone and your insights and your patience. Maybe you and Justin should team up together. Anyways, if you run with this idea, and I hope some day you do, I expect some kind of mention on the credit page. Again thanks for all your effort!!
@Jamie: Thanks for the suggestion. There may be a another book in the future, and if there is, it’s likely I would team up with Justin. Unfortunately, there are just so many competing priorities that’s it will be hard to find the time to make this happen.
Hi, speaking of NAV versus index returns, when I look at ZCN’s comparison on the prices and performance page (http://www.etfs.bmo.com/bmo-etfs/performance?fundId=72048) BMO lists the NAV since inception (over six years ago) at 6.35% while the index returned 7.79%. As far as I know this fund suffers from neither low trading volume, illiquidity, or time zone differences. In this case, what might account for such a large discrepancy? Thanks.
@Eva: When ZCN first launched it tracked a different index (large caps only). That’s almost certainly the reason for the discrepancy. Notice that the tracking error gets much smaller in recent years.
Dan
I understand that arbitrage activities should ensure small variability between market price and NAV – however is there ever a situation where there could be a significant long term divergence between the trading price of an ETF vs its NAV? ETFs are still a relatively new investing instrument. Markets are not always rational – could an ETF price collapse significantly below its NAV (beyond a temporary flash sell off)? Perhaps loss of confidence in a particular fund? Given that index funds generally hold the underlying stocks and are priced daily to NAV are they in a sense “less risky”?
Thanks
@Talisker: An investment fund (whether a mutual fund or an ETF) can’t be worth significantly less or more than its underlying holdings. Over short periods, as you’ve noted, there can be a disconnect between the market price and the NAV, and there may be periods where the bid-ask spread is unusually high. But an ETF’s price cannot “collapse” unless the value of the underlying holdings fall.