In the last few weeks, ETF trading and liquidity issues have
been in the spotlight. There's much to be discussed-and underlying
liquidity is central to that discussion.
In his recent open letter to investors, Global Head of iShares
Mark Wiedman made a clear distinction between the primary market
for an ETF (ETF share liquidity) and the secondary market of an ETF
(liquidity of underlying securities in a fund). However, there's
one line in that letter I simply can't swallow:"more and more
are becoming the true market." If anything, that line should have a
HUGE asterisk next to it.
Indeed, there are
when specific ETFs can provide price discovery, but in no way
should we assume that ETFs are becoming the
market. Rather, we need to understand where ETF liquidity is
limited; namely, the underlying liquidity.
In the media circus regarding the flaws of ETFs, the focus has
really been on trading of fixed-income ETFs. However, the
fixed-income ETF space is particularly complex because of issues
relating to how net asset value (
) and indicative NAV (iNAV) are calculated.
In the fixed-income space, NAV is calculated off the bid of the
underlying issues, and is stale after underlying bond markets have
closed. This is one reason why investors have seen such a
disconnect between ETF share prices and NAV values in recent weeks.
However, to say this is the
reason doesn't provide the full picture.
Underlying liquidity is really the first place to begin when
understanding ETF liquidity. When I say "underlying liquidity," I
tend to think of it in two forms:the actual frequency of trading in
the underlying securities, and the level of access to those
If you're dealing with a particularly liquid underlying market,
both in terms of volume and access, there's a strong argument to be
made that that ETF can indeed be a viable way to establish price
discovery. But crucially, when either of those two components for
underlying liquidity is limited, the argument for ETF price
discovery becomes a bit flawed.
Something like SPY is a great example of this. It's a fund that
tracks the 500 largest companies in the U.S. All the underlying
securities in SPY are liquid, trade during U.S. market hours, and
are incredibly easy to trade for any one of the 45 authorized
participants (APs) mentioned in Mark Wiedman's letter.
Even a fund like the iShares MSCI Japan ETF (NYSEArca:EWJ) can
be used for price discovery. Although the underling basket in EWJ
doesn't overlap during U.S. market hours, the level of access in
the overnight markets is high for a majority of APs, and the
liquidity in the underlying basket is more than adequate.
According to our measures, EWJ has an underlying volume/unit
measure of 0.04 percent. This means that 1 creation unit of EWJ
unwrapped into its underlying securities accounts for 4 basis
points of the daily volume in those securities.
To put that metric into perspective, EWJ trades the equivalent
of 100 creation units per day. So even on a day when every trade in
EWJ can be considered an "inflow," the AP responsible for creating
new shares of EWJ would make up 400 basis points, or 4 percent, of
the daily volume in those underlying securities-a drop in the
However, in cases where underlying liquidity is limited, either
because of limited access or liquidity-you tend to see these
limitations materialize in the premiums and discounts experienced
by those trading shares of the ETFs.
For example, let's look at an ETF such as the Market Vectors
Indonesia Small Cap ETF (NYSEArca:IDXJ). The fund has an underlying
volume/unit measure of 8.77 percent.
In essence, if I needed an AP to create 100,000 shares of the
ETF on my behalf, that creation alone would account for over 17.5
percent of the volume in the underlying securities. Also, consider
that only a limited few of the 45 APs have direct access to the
Sure, they could opt for a cash creation, but in that process
the AP is essentially handing over the buying of those underlying
securities to a third party that isn't primarily concerned with
getting the best price for them.
As a result, something like IDXJ can see a significant premium
or discount, not simply as a result of stale iNAV or NAV pricing,
but also because APs will have to charge a premium for the risk of
executing the trades-especially in a situation where's there's less
competition among market makers due to lack of access.
So how does all this apply to all the problems with exiting
fixed-income positions that journalists and bloggers are so worried
Interestingly enough, State Street Global Advisors tinkered with
aspects of underlying liquidity. A few weeks ago, the firm halted
cash redemptions in its fixed-income ETFs. As a result, APs that
had to redeem ETFs as they were buying them back in the primary
market were given the actual underlying bonds to sell in the
secondary markets. That's a huge burden on the AP.
Not only does the AP have to account for slippage in selling the
bonds, but it's likely the group of APs willing to do such
redemptions was smaller than the normal group accustomed to cash
creations. As a result, investors saw significant discounts.
However, the benefit of the in-kind redemption process is that
investors currently invested in the ETF are protected, while the
APs-and consequently those selling the ETF-bear the burden of the
The key point here is that ETFs may offer liquidity, but that
liquidity often comes at a premium that expands during times of
market volatility as a result of issues that begin in the
Also, to believe that ETFs are becoming the
market is an oversimplification.
Unlike futures contracts that can settle in cash, with ETFs, at
has to deal with the liquidity of the underlying markets.
At the time this article was written, the author held no
positions in the securities mentioned. Contact Ugo Egbunike at
Permalink | 'copy; Copyright 2009 IndexUniverse LLC. All rights
Don't forget to check IndexUniverse.com's ETF Data
2013 IndexUniverse LLC
. All Rights Reserved.