ETF Capital Markets Desk: Misconceptions about Bid/Ask Spreads: Take Two

Refreshed look at the relationship between the ETF bid/ask spread and its underlying basket of securities.

Background Image

David Mann Head of Global ETF Capital Markets,Franklin Templeton Investments

Here’s a refreshed look at bid/ask spreads featuring David Mann, Franklin Templeton’s Head of Global ETF Capital Markets. He examines the relationship between the ETF bid/ask spread and its underlying basket of securities.

As part of an effort to revisit my old posts, I am going to review my previous discussion about the misconceptions around bid/ask spreads. And in case you missed the first part of the original series discussing assets under management (AUM), it can be found here.

The main point of the original article was that the bid/ask spread does not necessarily reflect where that spread is in relation to the value of its underlying basket of securities. That is an accurate statement but could use a bit of extra clarification. Specifically, ETFs with similar underlying baskets tend to trade at similar premium/discounts to net asset value (NAV)1 , irrespective of the bid/ask spread at the time of trade.

For ETFs that hold only US underlying stocks, the premiums and discounts to NAV tend to be smaller since the entire underlying basket is open for trading. For example, our Franklin LibertyQ US Equity ETF (FLQL), our multi-factor US large-cap equity ETF, has had an average premium/discount of approximately two basis points (bps)2 over the past 12 months through March 29. In spread terms, this is less than a penny.

Remember, market makers set buy and sell prices for the ETF shares and stand ready to buy or sell throughout a trading session, which facilitates liquidity. For ETFs that hold international equities, market makers will often use correlated securities as proxies to price the ETF since trading in the underlying stocks is closed.

Many of the high-volume international ETFs will serve as a guide for what should be expected in those closed markets the next day. That expectation then helps drive the spreads (and thus the premiums/discounts) of newer funds with similar underlying exposures.

For example, in the first year of trading in the Franklin FTSE Japan ETF (FLJP), there were 11 trades of over $2 million. When comparing the premium/discount to NAV of those trades to those of the largest US ETF that holds Japanese stocks, the average difference was only four bps. In spread terms, this is a little over one penny.

So the part I would have added to my original post on this topic was the expectation that funds with similar baskets would trade at similar premiums/discounts to NAV.

There is also one more point worth highlighting about bid/ask spreads: They do make a difference when investors use market orders. I have a lot of thoughts on the topic—including a new definition for the term “market order”—which I will discuss in my next article.


  1. Net Asset Value (NAV) represents an ETF’s per-share value. The NAV per-share is determined by dividing the total NAV of the Fund by the number of shares outstanding.

  2. One basis point is equal to 0.01%.