Fixed income exchange traded funds (ETFs) will always have a degree of “imperfection”, according to Bluefin Trading.
This is because their underlying holdings are not an inherently transparent as equities, and the portfolio manager must select bonds from the index rather than opt for full replication in order to reduce tracking error. Managed in this way, it will take longer for fixed income ETF costs to fall, compared to equity ETFs.
But you can find ETFs whose bid-ask spreads are tighter than the underlying bonds they track, and the relatively cheap and transparent structure of the ETF is a good way to access the asset class.
ETF.com staff writer Rachel Revesz talks to Simon McGhee, head of ETF advisory for Europe, the Middle East and Africa and Lee Williams, partner and head of trading at Bluefin, about how these products work and investors’ common misconceptions.
ETF.com: How are fixed income ETFs different to equity ETFs and how do they trade?
Simon McGhee (SM): The main difference is that the underlying bonds trade privately over-the-counter whereas for equities they trade publicly on exchange. Bond prices themselves are negotiated bilaterally, and therefore their pricing is relatively opaque. Bonds also trade far less frequently than equities.
It’s actually often uneconomic for a fixed income ETF to hold all the bonds of an index it’s tracking. A physically backed FTSE 100 ETF will physically hold all the index stocks in proportion, but some of the bond indices have thousands of constituents, and if you have an ETF with, say, $50 million of assets at launch, it’s not possible to hold all of these bonds. The portfolio manager uses a sampling technique to select certain bonds that will mirror the return of the index.
Equity funds tend to be fully replicating and they can do that as it’s easy to execute all the underlying shares via a programme trade but you can’t do that with bonds as they trade OTC.
ETF.com: How are fixed income ETFs priced?
Lee Williams (LW): The conventional valuation methodology for fixed income ETFs also differs from equities.
Equity indices are “mid-valued”, while bond indices are mostly “bid valued”, but with a few using mid.
Because these bonds are so hard to price and the bid ask spread is so wide, the companies providing the indices said that using the mid-price i.e. the fair value, is not very useful. What investors want to know is where they can get out – the bid price. So, high yield might have a mid-valuation of 2 percent but it’s pretty meaningless as you can’t out near to that point.
The bond convention is to base the net asset value [NAV] of these assets at the bid mark of the bonds. It’s less common in more liquid products. For something like a rates ETF that tracks gilt futures with a spread only a couple of basis points (bps), it’s legitimate to use a midpoint valuation, but for credit, high yield and emerging markets, the trend has been to use the bid point.
ETF.com: What common misconceptions do investors have about fixed income ETFs?
SM: The misconception is that they are new products. The first treasury bond ETF was launched in 2002, a year after Bluefin was formed. Junk bonds came about in 2007. Now these products are being used more broadly. What’s new is that these products are being used more broadly by different types of investor.
People often think changes in premiums and discounts don’t work very well as there’s a difference in the ETF price and the NAV [net asset value], but that creates an arbitrage opportunity. This gap in price is exploited by market participants and that closes the arbitrage channel moving the price closer to NAV. The structure is incredibly well thought through.
ETF.com: How can investors measure fixed income ETF liquidity?
[This article continues...]