Last week caused one huge flashback to the May 2010 Flash Crash as a trading glitch at ETF market maker Knight Capital flooded the New York Stock Exchange with orders. It resulted in a 30-minute roller coaster ride for some 150 stocks. It also caused spreads to balloon for less liquid ETFs.
The Knight debacle certainly reignited questions about the role of automated trading programs that churn through massive volumes at lightening speed. But it also shed light on one core component of the ETF space – the role of the authorized participant. APs like Knight Capital are critical in helping more illiquid ETFs trade. They’re chosen by the ETF sponsor to buy and sell the underlying shares in an ETF. When problems arise at APs like Knight, highly liquid ETFs can recover fast – in that case, other market players stepped in quickly to pick up the slack and mop up liquidity.
Not so for some less liquid ETFs which according to some are still having a hard time recovering. David Nadig at ETF Universe crunched the numbers and found out how much spreads have widened – prior to problems at Knight Capital, an ETF that normally trades like than 50,000 shares a day had an average spread of 56 basis points. Since the problem, Nadig says spreads have nearly doubled: 94 basis points.
It gets worse he writes: “For those low-liquidity ETFs for which Knight is the lead market maker on the NYSE, spreads have ballooned from 0.49 percent on average to a whopping 1.53 percent.”
To be clear, Knight Capital’s problems could have happened anywhere: mutual funds, pension funds, hedge funds – just about every large institutional investor relies heavily on a complex suite of technological products in order to trade shares and stay liquid.
But what Knight’s trading woes show is how careful investors need to be to look under the hood and understand how ETFs work and how different factors in the ETF space – market markers for example – work in tandem to keep shares moving (or not moving as the case may be).
In the case of Knight Capital, knowing who the AP was for those ETFs probably wouldn’t have helped – after all, Knight was the gold standard, the biggest and by all accounts the best. But it does show how important it is to understand the creation process of any ETF – and how active an AP needs to be to make an ETF viable (liquid). For that, more disclosure – and interest from investors – on both fronts is key.