Most Popular ArticlesSeptember 15, 2009
How to measure ETF liquidity
When evaluating an ETF, simply ask one question, “What is IN this index?” In other words, “Where does this ETF invest?” Is it a highly liquid and well-benchmarked index like the S&P 500, S&P 400 or S&P 600? Or is it an index that tracks a narrow slice of the market like solar energy, wind, or the ophthalmology industry? This should be the core of any ETF evaluation process, not relying on average daily trading volume to screen out potential investments.
Why is understanding what is in an index so vitally important? Because the liquidity of an ETF has everything to do with the liquidity of the underlying stocks that the ETF’s index tracks and little to do with the average daily trading volume of the ETF itself.
A clear cut example can be made using a well known, broad based, heavily traded ETF like SPDR S&P 500 Fund (SPY) and comparing it to RevenueShares Large Cap Fund™ (RWL). Both SPY and RWL track the same S&P 500 index and invest in the same 500 stocks; however, the weightings within the indexes differ slightly because RWL weights the S&P 500 by top line company revenue, not market cap as in the case of SPY.
Both of these ETFs invest in the same stocks, from the same well known and highly liquid benchmark, except they employ different weighting methodologies. Potential investors, following conventional wisdom, will look at the average daily trading volume of SPY (260 million shares) versus RWL (31,000 shares) and make a black-and-white judgment that RWL’s lower liquidity level could present a problem should they decide to exit their position. This is a case of the tail wagging the dog. Investors will incorrectly deem an ETF illiquid and therefore “uninvestable” because its average daily trading volume hasn’t reached a certain level. Dozens, if not hundreds of ETFs are disqualified by investors every day because of this misconception.
Bid/Ask spreads
Advisors need to take ETF evaluation a step further. Given the disparity of average daily trading volume between SPY and RWL, the bid/ask spread in SPY is 1 penny whereas the spread in RWL is 2-3 cents. Most investors would be perplexed by this. How is it possible an ETF that trades a small fraction of the average daily trading volume of SPY has a similar bid/ask spread? Both SPY and RWL track the same highly liquid index, the S&P 500. The price movements of an ETF’s underlying stocks determine its pricing, whether it is SPY, RWL or IVV regardless of the ETFs’ trading volume activity during a given session.
ETFs are not closed end funds and they do not trade, nor are they priced like closed end funds. This fact can be tested on any financial data system - Reuters, Bloomberg, or even Google Finance. Readers can launch this link and enter the symbol RWL.IV, to display the real-time indicative value of this ETF – roughly the equivalent of its NAV. Now compare this to the current bid/ask spread for RWL. Is the IV roughly in between the bid/ask? If the answer is yes, then the ETF is efficiently priced. If the bid/ask spread tight - within a few pennies - it cuts down on transaction costs. Consider using a limit order when buying or selling ETFs, placed within a few pennies of the IV, to ensure fair execution regardless of the bid/ask spread.
Testing pricing efficiency
In addition to running the IV test, I encourage ETF investors to watch the IV on the screen, on Google Finance or elsewhere, and cross reference it with the bid/ask over the course of a few minutes. An ETF like RWL may not even trade over this time frame, but rest assured the IV and Bid/Ask will change in real-time, ticking up or down as the underlying stocks in the S&P 500 index move.
This point is unfortunately lost on too many investors in our industry. Trading volume is not dictating the price changes in the ETFs. The driving force is changes in the market via the underlying securities. Unlike a closed end fund, which is driven by buyers or sellers in the market, an ETF’s value will fluctuate all day, every day regardless of volume.
Display article as PDF for printing.
Would you like to send this article to a friend?
Remember, if you have a question or comment, send it to .
