Understanding zombie ETFs

Due to their low assets, these products are thinly traded; their liquidity comes mainly from the work of market makers.

Some in the financial industry refer to these funds as “zombie ETFs” — funds considered the “living dead” due to their low trading volume and low assets. Others refuse to attach the zombie label to funds that are too new or that haven’t yet found an economic environment conducive to asset growth.

To find out more, we spoke to Daniel Straus, vice president of ETFs and financial products research at National Bank Financial Markets in Toronto.

Finance et Investissement (FI): How do you define a zombie ETF? 

Daniel Straus (DS): ETFs often launch with “seed capital” assets under management of $5 million. Usually the market maker or designated broker bank provides this seed capital. If the fund fails to attract inflows beyond that initial capital, then the ETF provider might delist and terminate the fund after a few years.

A zombie ETF is one that might get terminated in the near future, but I should point out that some might contest the designation of such ETFs as “zombies.” Especially in Canada, investors are naturally conservative, and tend to wait for a year (or two, or three) of live performance history before committing any money to a new, untested fund.

There are many interesting examples of ETFs that “came back from the dead,” so to speak: they languished at the seed-capital level for a long time, before investor interest caught up with the investment idea.

Instead of calling small, forgotten ETFs “zombies,” sometimes I call them “ugly ducklings” —you never know if one might grow up to be a beautiful swan. For example, Invesco Canadian Dividend Index ETF, originally branded as PowerShares, launched in 2011. For three years, it struggled to gather assets, remaining under $30 million. Then, in 2017 it started to pick up significant inflows, likely on the back of strong real-world returns that investors noticed.

Now it has over $600 million in AUM, and is the third-largest Canadian dividend ETF in a crowded field.

FI: At what level of AUM is an ETF no longer considered a zombie?

DS: Many variables affect an ETF’s termination risk. I’ve heard from some ETF providers that a fund’s break-even point for profitability is usually around $50 million, but this will depend on the fees and other costs associated with running the fund.

I would say that size alone does not determine zombie status — in our research, a combination of low assets, long lifetime since inception, poor performance and low trading volume all point to increased risk of fund termination, but none of these variables by themselves makes an ETF a “zombie.” Besides, examples abound of ETFs that satisfy all these conditions and have not been delisted (yet). At the same time, some larger ETFs have delisted.

FI: Does an investor take additional risk when investing in an ETF that has low AUM (e.g., closure risk)?

DS: The risk of a “zombie ETF” is more of an annoyance — not a true risk of loss. Worst case, if the ETF provider terminates the fund, any remaining investors will get their money back in a one-time special cash distribution. In that sense, it’s nothing more than a “forced sale,” which might have negative tax consequences for some investors.

However, risk is more likely to occur in products that have few actual investors in the first place. After all, if the ETF saw no inflows beyond seed capital, then very few real investors own the units aside from the market-making banks.

In general, fund terminations are announced well ahead of time, so investors don’t lose control entirely — they can choose to wait until the termination date to collect the proceeds, or they can sell their shares in the open market at any time before termination, because market makers are still committed to quoting continuously throughout the trading day.

Interestingly, if you choose to go this route, you will probably have to accept the “offer” price at the lower end of the ETF’s bid-ask spread. If the fund winds down and distributes its net assets in cash, it’s possible that the final investors will receive pricing that’s a bit better because of the scale of the final series of termination transactions.

FI: Is a zombie ETF be more difficult to manage for a market maker?

DS: When a new ETF is in high demand, the units that comprise the seed capital are sold on the open exchange out of the market maker’s inventory. After this point, market makers keep very low (if not zero) inventory of the ETFs they quote on the exchange, relying on the creation-redemption mechanism to balance their positions at the end of each trading day.

This ideal scenario is very capital-efficient for market makers. In fact, when we start to see true inflows into a new ETF, we generally know the market makers have sold out their seed. So, if an ETF has registered zero inflows, then market makers have to expend capital to keep the seed inventory on their books. This can be costly from an accounting and capital perspective, especially for banks, but not always — sometimes they enter into agreements with the ETF providers to allay or share these costs.

That said, quoting bid and offer prices for a low-volume, small, or unpopular ETF is not any more inherently difficult than for any other ETF — it all depends on the liquidity of the underlying market.

FI: If an ETF has multiple small transactions, these fees could result in a very high trading expense ratio. What do you think of this aspect with respect to ETFs with limited assets under management?

DS: Small ETFs are likely to report higher trading expense ratios in regulatory filings, because they might have fixed costs like one-time portfolio initiation trades — or the calculation is annualized according to a flawed regulatory formula. We take some of these numbers with a grain of salt, especially for funds less than one year old.

FI: What should investment advisors understand when investing in ETFs that have limited assets?

DS:

• Liquidity is determined by the underlying assets, and not by the ETF’s share volume.
• Just because an ETF has no assets, it doesn’t necessarily mean the idea behind it is bad. Successful investing sometimes means going against the crowd.
• Mutual funds may be a riskier fund structure from a termination perspective. If there is a true liquidity crunch on the underlying assets and the mutual fund experiences a “run” on redemptions, mutual fund managers may have no choice but the close the fund after a certain point. This is because they must fulfill the early redeemers in cash, and they finance this by selling the most liquid assets first. Investors who redeem later hold units with progressively less liquid and potentially toxic underlying assets, promoting a vicious cycle of runaway fund illiquidity. This has happened in the world of mutual funds multiple times, for example with New York-based Third Avenue Management. The creation/redemption mechanism of ETFs circumvents this problem.

FI: Is there a risk of an acceleration in the number of ETFs closing in Canada (whether zombie ETFs or not)?

DS: Every once in a while, we see waves of closures that have to do with fund providers streamlining their product shelf — we consider this a healthy “pruning around the edges,” and it’s indicative of a robust and growing marketplace. For example, in 2017, iShares delisted a number of redundant ETFs, some of which had non-trivial assets.

There may come a time when the lines between ETFs and mutual funds blur even further (mutual funds close and/or merge all the time, and no one seems to bat an eyelash), but that day could be a long way off. That said, smaller providers like Coin Capital and Equium Capital have pulled their ETFs. The marketplace is extremely competitive, but investors are the net beneficiary of this competitiveness because of the lower fees.