US Videos

Vanguard's Sauter on How to Prevent Another Flash Crash

Paul Justice, CFA

Paul Justice: Hi there. I'm Paul Justice, Director of ETF Research in North America for Morningstar. Today I'm here to discuss the regulatory reforms following the flash crash of May 6th and 22 minutes of infamy. I'm joined by a central figure in the discussions Gus Sauter, Chief Investment Officer of Vanguard.

Gus, thank you for joining me.

Gus Sauter: Thank you, Paul for having me.

Justice: Now, while the cause of the flash remain unknown right now, it seems like we've really identified that we have some market structure problems that regulators are addressing and you've had several inputs as to some of the directions that should be chosen. Could you go over some of the most recent rulings? First of all, we'll discuss some of the circuit breakers that are discussed right now to help prevent future crashes from happening and how you think that these are going to perform and what other things need to be implemented?

Sauter: Sure. Yeah. So the SEC has implemented a pilot program with circuit breakers and any time a stock declines 10% in a five minute period, basically a halt is called, everybody has to take a time out and wait for five minutes before trading resumes. I do think that that's a good concept. I will admit that years ago when we started introducing circuit breakers at the market level that I wasn't really a fan of it, but I have changed my thinking over the years.

And the real reason is, because I recognize that at times people get a little hysterical overall, subject to perhaps panicking in times of duress, and when that happens we need to step back and ask ourselves is something really different or are we overreacting here. And so I think having this circuit breaker and stopping things for a few minutes, giving us time to reassess the situation actually will be a benefit and we'll see how that works out in the pilot program.

Justice: Sure. Now, speaking of kind of panicking and pulling out of the market, one of the participants that we saw in this were the market makers as we call them and often times these are the high-frequency traders that really enable the ETF marketplace to exist. Could you discuss their role in the marketplace and what reforms may be coming that could impact the market makers themselves?

Sauter: Yeah. So there are really are three different types of markets makers, if you will. There is a specialist in some cases where there is somebody who is a designated market maker required to make markets. There is also electronic market makers, who are designated as market makers on various exchanges where ETFs may trade, and then there are these high-frequency traders that you're referencing and these traders are making markets when they feel like making markets and perhaps disappearing at other points in time, but all of these are helping to provide liquidity to the marketplace.

What we saw was that they tended to disappear from this on May 6, but I would point out it wasn't just the high-frequency traders who decided not to make markets, it was the other market makers as well, who might be designated as market makers who had stepped back, and even on exchange traded products like stocks these specialist stepped back away as well. So, it's not just high-frequency traders that perhaps abandoned us, if you will, on that day.

Read Full Transcript

Justice: So, you have this first layer of market makers those are basically kind of required to be in position to place a bid on a stock. Now, the SEC today is stating that it may implement a 10% band rule on these market markers making sure that they have a bid available or an ask that's within 10% range of the current trading price, do you think that this could be an effective measure to complement the circuit breakers?

Sauter: I would be actually worried about a mandate that a market marker has to step up. If you've got a tidal wave coming at you and you're told you've got to stand in front of that tidal wave you can get crushed by it. And I would be worried that market markers could basically be forced into bankruptcy. I think it's actually better to figure out ways to prevent that tidal wave from really building and crashing. I'd rather see the circuit breakers implemented and I think the circuit breakers will be successful in preventing the buildup of the tidal wave and then you won't be required to have the market makers standing there in front of it.

Justice: Sure. You bring up a good point, because some of the market makers they really have been vilified in this circumstance, but they are essential for the ETF product to exist and for all these trades to happen very quickly. Can you talk about some of the benefits that they've actually brought to the marketplace over the last two decades?

Sauter: Sure, ideally we'd love to see ETFs and stocks for that matter trade just between investors and investor who naturally wants to buy a stock or an ETF and one who wants to sell the ETF. It's great to take out the middle man, but unfortunately there aren't always people ready to transact at same price at same time. So, you need these market makers who are willing to bridge that gap between the two investors.

Somebody is willing to own the security for a period of time before they can turnaround and sell it to somebody else who does want to buy it. So, I think market makers serve a very useful function for us when there is not natural liquidity in the marketplace. They make tighter markets than we would otherwise see, bid-offer spreads are narrower, and they provide us with liquidity so that we can get out when we want to get out.

Justice: Now, one proposal that you've been a large proponent of is eliminating market orders and promoting the use of limit orders. Could you help explain your position on that and how it could possibly help market liquidity?

Sauter: I think a lot of investors don't really realize the risks of market orders. You think that a stock is selling at a certain price level, so you'll say, "Well, I am happy with that. I'll go ahead and sell it at the market or buy it at the market," and then it turns out that the liquidity really isn't there and by the time you end up buying it with your market order you've pushed the stock dramatically or the floor gets pulled out from under you and on May 6th the limit orders just weren't there. You enter a market ordering, you end up selling it to $0.01 a share. I don't think many investors realize that that risk was really there. For that reason I do like the concept of requiring everything to be a limit order. Now we would point out that you can in effect place an order that would function like a market order even by using a limit order. If you really, really want to buy a stock, let's say, you could offer to buy it or bid for it at $1,000. Well, you're going to buy it at whatever it's being offered at and which would be a reasonable price, but you're really assuring that you'll be able to buy it. So, the advantage of requiring all orders to be limit as opposed to market is that investors can still place the market order or something that functions like a market order, but they are also placing an intentional limit as opposed to one by default that could be $1,000 without realizing that it's a $1,000 on the buy side.

Justice: Now, once we've addressed the issues of fixing the market hoping it never breaks again, you still have to deal with the inevitable fact that maybe it will break again and then you have some trades to potentially cancel. We had a 60% ruling. If it traded off 60% during that timeframe, that those were the trades that were canceled, but several funds traded maybe 50%, 40% away from where they were and those were not canceled trades. What kind of rules need to be put in place regarding defining what trades are erroneous and when they should be canceled and what compensation should be just?

Sauter: Sure, there are certain trades that after the fact get designated as "clearly erroneous trades" and then somebody says, "Okay, well, those are going to be busted. Those trades are canceled." Well, historically, it's been somewhat of an ad hoc process of trying to say, "Well, that was erroneous or not erroneous." By having it ad hoc as opposed to a stated rule, it creates obviously a lot of uncertainty. When people are entering into the market, they don't know whether or not something might be canceled, and so it really creates many secondary problems that wouldn't be there otherwise. In fact, one of the reasons that many of the market makers exited the market on May 6 was because of the lack of clarity around "clearly erroneous trade." I think if you state a certain band for "clearly erroneous trades" like 10% outside of the current price, then market makers and other investors can know how to respond. They know when something is going to be designated "clearly erroneous," and you can keep better liquidity in the marketplace.

Justice: Well, can you think of anything else that needs to be done or should be presented in order to make this market more efficient?

Sauter: Well, I think the SEC is pursuing the right path right now. Trying a pilot program with these single stock and single ETF circuit breakers, I think, is an appropriate way to go, and we don't know that we have all the right answers and all the right dials tweaked at this point in time, so hopefully they'll be figuring that out over the next six months during the pilot program. I think they should continue to consider market orders versus limit orders. And one thing I think that really needs to be re-thought out is this "clearly erroneous trade" rule. Right now it's being applied what I think is incorrectly. Just the other day Citigroup traded at an appropriate price within that 10% band. that's now determined "clearly erroneous." However, after the trade you have to get the trade into the system to be reflected nationally in the trade book, and when it was entered in manually it was entered in incorrectly. Well, it was outside of this "clearly erroneous" level and so it halted trading in Citigroup stock. It looked like a trade had been executed outside of the bands, and so incorrectly the stock was halted. So I think they need to figure out a better way to implement the "clearly erroneous trade" rule not using actual trades that could be input incorrectly but using bids and offers.

Justice: Those are all relevant and interesting points and I'm sure we'll have more to develop on this discussion later on, but I appreciate you joining me today and we look forward to having the next discussion.

Sauter: Okay. Well, thank you very much Paul.