Julie Abbett: Quoting volume as a proxy for exchange traded fund (ETF) liquidity and similar misleading information is quite prevalent. There are a number of ETF bloggers who frequently write about how an ETF’s short trading history or low trading volume make it an unattractive investment. Additionally, commentators often look at where the ETF last traded, not at the fund’s Net AssetValue (NAV), as a basis for estimating the cost of trading. Both of these mistakes can harm investors if allowed to continue. The opportunity cost of shunning newer, more innovative ETFs because of a lack of understanding about ETF liquidity may actually exceed the true execution cost. Exposure to alternative asset classes through ETFs in particular is fairly new and can beused to create more efficient portfolios. Investors may be paying much more dearly by missing out on valuable asset allocation exposure and hedges provided by some of the newest and mostinnovative ETF products.
Institutional investors, RIAs, financial advisors and anyone else trading an ETF that is composed of liquid underlying securities can find immediate liquidity and execution for large trades in that ETF at very low cost, regardless of the ETF’s trading volume and bid/ask quote depth. Unlike a regular equity security in which there is a somewhat fixed number of shares outstanding, ETFs can create new shares and therefore additional liquidity as frequently as necessary. Institutional traders called “Authorized Participants” (APs) use the creation and redemption process to bolster ETF volume and satisfy market demand. Liquid portfolios make it easier for an AP to cover the cost of creating additional shares of the ETF, thus facilitating low cost execution. The more liquid the underlying securities are, the easier and cheaper it is for an AP to create more shares and meet demand. The APs will buy the underlying positions and deliver these securities to the fund’s custodian in exchange for shares of the ETF that they can turn around and sell to their clients. This process reinforces that printed volume does not represent true market liquidity and, in fact, can actually be a very misleading indicator of an ETF’s liquidity and in turn the true cost of trading.
Using the last traded price for an ETF as the basis for estimating transaction costs ignores the fact that the market may have moved quite significantly since the last trade. Instead, investors should be focused on the ETF’s NAV, which is the true real time value of the ETF’s underlying securities. APs, using the published basket file (often referred to as the “NSCC file”), base the price at which they are willing to execute off the NAV, not the last traded price of the ETF. Aslong as investors can execute at NAV plus a nominal cents per share creation cost, they have a good point estimate by which to determine whether or not they will receive good execution. Both last price and volume provide very little information about the price at which investors can expect to execute.
Clients who have larger orders above 10,000 shares can place a direct (principal trade) with a broker at a pre-negotiated price close to NAV. The broker that a client can trade with is either also an AP or has direct access to an AP desk, which will be able to execute the trade directly. Sourcing liquidity through a principal trade with a sell-side counterparty minimizes market risk and also helps the client to establish a price prior to execution. The sell-side wants to make aprofit which they can do as long as they make a spread that covers their costs. When the underlying securities are liquid, they can easily do so. Under these circumstances, almost anysell-side broker will facilitate a principal execution.
IndexIQ Live Execution
To illustrate that volume and last trade are poor indicators of expected trading costs, wesummarize a live trade in which a client wants to execute 50,000 shares of the IQ ARB MergerArbitrage ETF (NYSEARCA: MNA).
- Client wants to purchase 50,000 shares of MNA
- Client reaches out to two or more brokers and asks for a principal quote to buy 50,000shares of MNA rather than trying to work the large lot
- The broker prices the underlying securities in MNA using the published ETF basket file(NSCC file)
- At 11:50:34, the last traded price was 26.35 (Table 1)
- At 12:06:28, prior to the 50,000 shares trade, printed volume is posted at 5,590 shares(Table 1)
- At 12:06:00, the bid-ask spread is 26.07/26.21(Table 2)
- At 12:06, the NAV was 26.0696 (Table 3)
- Broker A was able to price the 50,000 share block at 26.0715
- Broker A must have had the best offer: 26.071 (Table 1) was the printed execution,resulting in a very low cost execution
Broker A may then:
- Immediately create 1 unit of MNA (1 unit = 50,000 shares);
- Go short 50,000 shares of MNA until it is ready to create 1 unit;
- Own long positions in the underlying ETF holdings to hedge its short position in MNA;or
- Purchase 50,000 shares of MNA itself on the open market to cover its short position
Typically, Broker A will create 1 unit of MNA either immediately or within 1-2 days, depending on its current inventory.
In summary, at 12:06, the NAV was 26.0696 (Table 3) and the bid/ask was 26.07/26.21 (Table 2). At 12:06:28, a 50,000 block trade was executed at 26.0715 (Table 1), very close to both the bid and NAV. Prior volume of 5,590 and the last traded price of 26.35 had no impact on this trade, as the client was filled very close to the bid and NAV.
This is a great example of how the printed volume or quoted bid/ask spread and quote depth foran ETF is not an indicator of how liquid the issue is. Once market makers see interest, they come in and try to fill the orders. Eventually enough demand will drive significant competition and the spreads will become thin.
Julie is Senior Vice President and Portfolio Manager forIndexIQ. Prior to joining IndexIQ, Julie was a Portfolio Manager at Deutsche Asset Management (DeAM)/DB Advisors for over 9 years. Julie was a Director and Portfolio Manager for various U.S. and Global strategies, which included the DWS Disciplined Market Neutral Fund, the DWS Blue Chip Fund, both the DWS Disciplined Long Short Growth and Value Funds, as well as a number of other institutional and sub-advised funds. The funds’ strategies used multi-factor quantitative models based on comparing companies on fundamental value and growth, as well as market sentiment signals in a disciplined risk managed process.
Prior to becoming a Portfolio Manager, Julie was a productdeveloper at FactSet Research Systems. In this role, she worked on developing and enhancing features on existing portfolio analytics products. Using the experience she gained in her earlier role at BARRA, Inc., she also worked with engineers on developing an open model equity risk analysis application that allowed clients to import user defined risk models.
Julie earned a Bachelor of Arts in Economics from the University of Connecticut and is currently pursuing a Master of Business Administration in Quantitative Finance and Economics from New York University Leonard N. Stern School of Business.