Last Updated: 18 May 2021
(If you haven’t seen Paul Amery’s blog on the rule changes that led to the flash crash, read it here.)
The London system that requires market makers to maintain specific bid/ask spreads on ETFs (within 1.5%, 3% or 5%) is enviable, but I don’t think it would fly in the U.S. At a minimum it would cause significant fallout and limit the number of ETFs that would launch in the future.
Market makers have already been stepping away from the ETF market for a number of years in the U.S., and for good reason: there’s not much in it for them.
When a new ETF is launched, the official market maker has to take on a number of risks. First, they must supply seed capital, putting up (typically) US$2.5 million to create the first shares of the fund. They must finance this cost and the cost of hedging their position while unwinding those first shares.
Then, they are required to maintain continuous quotes in the security, allocating time and money to monitoring it.
In return, they get a ten cent rebate on every 100 shares they trade. Given that many new ETFs trade less than 100,000 shares per day, you’re talking about a US$10/day kicker for all that work.
Of course, if you hit a home run and get to serve as Lead Market Maker on the next GLD or SPY, you make a mint over time. But at this point, the likelihood of that is small.
Did I mention that Lead Market Maker status can be taken away from you if you are not on the inside of the National Best Bid and Offer (NBBO) a certain percentage of the time? A savvy algorithmic trader can wait to see what ETFs are successful and then sit one penny inside your offer for a few months and steal the LMM status away from you.
Add in a new requirement to maintain specific quote quality and forget it. If you think it’s been hard for new ETFs to get seed capital up to this point, add in that new requirement without making the deal more profitable for market makers and it simply won’t happen.
It’s true that regulations are always fighting last year’s problems. But I think the single security circuit breakers that have been proposed – assuming they are extended to ETFs – will go a long way towards resolving the specific problems we encountered earlier this month.