Subtitles prepared by human
(piano music) - Financial markets operate at ever faster speeds. Critics say that opens up new possibilities for manipulation and prioritizes speed of execution above liquidity and quality. So is the current structure of regulation in the US up to the task? Welcome to "The Big Question," the monthly video series from Chicago Booth Review. I'm Hal Weitzman, and with me to discuss the issue is an expert panel. Eric Budish is professor of economics at Chicago Booth. He's an expert on designing all sorts of markets, from the markets for MBA classes and concert tickets to financial exchanges. His research proposing how to slow down financial markets won him several awards, and started a conversation among policy makers and industry professionals across the US. Sharon Bowen was a commissioner at the US Commodity Futures Trading Commission, the regulator for US futures markets, from 2014 to 2017. Prior to that, she was acting chair at the Securities Investor Protection Corporation,
following decades spent as a corporate lawyer in New York. She currently serves on the board of directors of Intercontinental Exchange. Steve Crutchfield is head of market structure at Chicago Trading Company, and a member of the advisory board at the Chicago Board Options Exchange. He was previously an executive vice president at Intercontinental Exchange, and CEO of NYSE Euronext's two US options exchanges. Panel, welcome to "The Big Question." Sharon Bowen, let me start with you. Do US regulators have the expertise and the technology to regulate these very-high-frequency markets? - I believe we do. We do have the resources and the expertise; but having said that, I've been a big critic of the fact that CFTC has been woefully underfunded for years. If we had more money, we could have better technology. - [Weitzman]: By administrations from both parties in the US— - [Bowen]: Yes. We've been flat-funded for four years, which is a long time, and with more money, we could invest in more technology.
But we do have the expertise to oversee these markets. - OK, Steve Crutchfield, you're in the markets, do you feel that they're well regulated? - Certainly, yeah, and I absolutely will defer to Sharon on the question of funding and resources. One question that we often wonder about, is about the focus. Is the focus of the regulators correct when taking a look at these markets? For example, particularly when we talk about some of the issues of latency-minimizing trading driven by very advanced technology. - OK, so lets go back a second. "Latency minimizing," meaning the fastest possible speed of trading. - [Crutchfield]: Right, right. And there are some questions around whether the regulators are focused on addressing the issues around incentives that arise from that type of trading, or whether they're more focused on, traditionally—in the case of the FCC markets, where I spent a lot of my career— looking at rules on a case-by-case basis. - [Weitzman]: —which are the securities markets on the other side. - Absolutely: equities, equities options, index options, and other products, ETFs— very popular with investors. Much of the regulation seems to be looking at individual rules as they're submitted and filed by exchanges,
which may, in isolation, be appropriate, rather than always looking at the entire market structure that's created as a result of these rules piling up over many years. - Well, Eric Budish, that leads neatly into your research about slowing down markets, but let's not go there quite yet because I wanna talk first about this question of capacity, and the technology that's needed to regulate markets. Is it your sense that markets are too fast to regulate or is it possible to regulate? - I won't take the bait on possible to regulate, and I defer to the commissioner, of course, on the funding question. What I will say, and this relates to my research a bit, is that the markets are more difficult to regulate than they could be because of the importance of speed, and because markets are trading literally as fast as technologically possible, which then creates these funny things, if you're trying to surveil markets, where you have to take special relativity into account
to figure out, did this thing in Chicago happen before or after this thing in New York. It's possible to do so, but it's more difficult, more complicated, than could be the case— - It sounds like that has more to do with, sort of, the philosophy of how the markets are approached rather than the capacity, which we talked about earlier. Sharon Bowen, part of the way in the US that this works is that the markets actually regulate themselves. To what extent do the regulators like you or at the CFTC depend on the exchanges, which you are now on the board of, or depend on these industry bodies to regulate themselves? Does that self-regulation work? - I think it does. It really is a good balance, I think, of resources, and expertise at the same time. I think most of the exchanges have great governance practices. For example, at the New York Stock Exchange, we have a separate regulatory oversight committee. —it's separate from the membership committee— that is responsible for doing
the surveillance of our markets. As regulators, we never abdicate our authority, obviously, but for the routine registration, licensing, market surveillance. I think it is a good balance between the agencies and the SROs in that respect. - OK, the "self-regulatory organizations." For someone who comes from the outside who says, how can financial markets possibly regulate themselves? You think it works perfectly well? - Well, you have to also remember, these are really competitive markets, and so the likelihood that one player's gonna go out and game the system, it's not gonna happen for very long. These guys tell on each other, as well, so, yes. - OK, Steve Crutchfield, you were gonna say something? - Yeah, I would certainly agree from the enforcement standpoint, from the surveillance standpoint. I do think, also having seen at the New York Stock Exchange and other markets, very good separation of business decision-making, customer-service, sales, and then a regulatory arm, often there's a CRO— we used to have it at NYSE—
who reports up through a different channel to a different board to prevent those— - [Weitzman]: CRO would be "chief regulatory officer"? - Exactly, however, I think there are some questions about not the surveillance and enforcement, but actually the rule making itself. To what extent are exchanges promulgating rules that do what exchanges are really expected ultimately to do, which is— this is language from the Securities Exchange Act of 1934, which enabled the securities markets as they currently exist, talks about "investor protection in the public interest." And part of the problem is that exchanges, although they are expected to do that— and on a case-by-case basis, the SEC, for example, will ensure that the rules, at least, don't do the opposite of that— there's nonetheless the question: What are their incentives? What are exchanges actually incented to do? Unfortunately, in many cases, what exchanges are incented to do is to find the market participants who are the most elastic in terms of price, in terms of incentives, in terms of what technological advancements they can make relative to trading on a particular exchange. - [Weitzman]: And those are the mom and pop traders, you mean?
- And they're generally not, right? Those most-elastic participants, the ones with the ability to move their volume elsewhere because of that competitive dynamic, are very often large institutions. Those large institutions may have found a way— I wouldn't use the phrase "game the system," but— they may have found a way to really profit from existing at the very bleeding edge, that "special-relativity world" that the professor was talking about. Often, incentivizing rewarding the firms that have found a way to exist at that level, that can lead to consolidation. That can create oligopolies. That can reduce the number of market participants who are out there providing bids and offers for investors to trade with, and ultimately that's the function of markets, to encourage aggressive bids, aggressive offers, attractive prices that investors can look at on their retail-brokerage website screen and click and trade. Providing incentives that create the best environment for that small number of institutions that are willing to spend all that money to live in that world
of special relativity does not always have the result of ultimately providing the best prices, the best experiences for investors, and that's something that we're concerned about. - You're not suggesting that regulators should be creating the products in the market? - No, certainly not. Absolutely fair. - Is there a question about, for example, the high-frequency traders who now provide so much volume to the exchanges? Can the exchanges really regulate them as vigorously as they should be regulated given that they're such big customers for them? - A couple of things about the high-frequency traders. Yes, the market structure has changed, and we do have different people bringing liquidity to the markets than was the case a few years ago. Traditionally, when the banks were our liquidity providers with the capital rules, they're less likely to do so in periods of stress because of the capital constraints. When you take a look at things like flash crashes, for example, I think people presuppose that it's the HFT guys who flee the market,
and that just hasn't been the case. They tend to be there, and providing liquidity, because banks can't do it because they're constrained by capital costs. We find that with more liquidity in the market means tighter bid-ask spreads, which means the costs are cheaper for investors. I'm not trying to advocate for one player versus the other in the market, don't get me wrong. But I do think the market should really dictate the products that are acceptable, and with caveats, there should be a market for those products. - [Weitzman]: What about this question of regulation though? What you're talking about is them bringing liquidity to the market, so we can accept that. But just the idea of day-to-day regulating these firms, is it harder because they're such big customers? - No, it's not harder because of that. - [Hal] OK, Eric Budish. - May I chime in? I wanna make a few related points. One is on the incentives discussion we've been having. I think it's worth highlighting that competition among exchanges has worked
extraordinarily well on some dimensions, and poorly on other dimensions. The dimension in which it's worked in the record extraordinarily well is trading fees. The fees are complex. You need a PhD in organic chemistry or something to navigate trading fees, but the average fee to trade a share of stock in the United States is tiny. It's about 0.01 pennies per share traded, once you work through all the complexity. If you go to StubHub and buy a $100 concert ticket, you're paying 20 bucks a fee. If you go to the New York Stock Exchange and buy a $100 share of stock, you're paying 0.01 pennies a fee. That's a big difference. The dimension in which exchange competition is a lot less fierce is the sale of colocation, and proprietary data fees. Selling speedy access to exchanges, most latency-sensitive customers, those fees have gone up noticeably over time. There's one debate between a high-frequency trader
and the New York Stock Exchange in which the high-frequency trader said, "This particular aspect of your fees has gone up 700 percent "in the last decade or so." It's because, the way I think about it conceptually, is exchanges have faced really vigorous, fierce competition on one piece of their market, which is trading fees. "If I can save one hundredth of a penny, "I'll go here not here," but they have market power over the sale of colocation and data. Only the New York Stock Exchange can sell fast access to the New York Stock Exchange. Now that dynamic, then I think, is a friction against market-design innovation because if you think about an exchange's incentive to adopt to new market designs that reduce the importance of speed in today's markets, such an innovation is good for the market as a whole, I would argue, but hits the one piece of an exchange's business that's any good, that gives exchanges market power, which is the sale of proprietary, fast connectivity to that exchange.
I think that's a friction against innovation. Something regulators could do that would be short of mandating a reform to market designs would be at least to proactively clarify that new market designs are allowed. But even that's been a step too far for US regulators. - I think it depends on how you define the market. I think you're right, speed costs money. Those are ways to make it much more efficient for people to do HFT, and it costs money to do it, but you agree it's definitely cheaper for the investor. Not every market is as liquid, and so you don't see the HFT in those illiquid markets. I don't think it necessarily stifles innovation. The pork belly market, I don't think you're gonna see a lot of people in those markets. - I defer to Steve on that. - I don't know, well no, maybe pork bellies. I don't know. Not every market is as standardized and liquid, in that sense. You're not gonna get innovation in every market. There's too many of them. - I think the objective, by the way, from a market-design perspective, is how to realize the benefits of algorithmic trading.
Computers are good at stuff. You don't need to be a professor to know that computers are good at processing lots of information. - But you would still agree, though, the market should create that. - What do you mean by the market should create that? - Competition. - As opposed to . . . - [Bowen]: Regulators. - [Budish]: Regulators saying this is exactly how your market should be architected? I think—. I believe in free-market competition. I'm a University of Chicago economist. I think markets need good rules within which competition takes place and flourishes. - [Crutchfield]: I think that's a critical point. It goes back to the question that you raised earlier, which is: Are regulators gonna create the perfect market? Do we expect that of our regulators? No, I don't think we expect it. I don't think we want it. The goal of the regulators is to create this sandbox within which everyone can play fairly, according to the rules. But I do I think, getting back to a point you made a minute ago, a challenge that we have right now is that the regulators are not necessarily— not only are they not encouraging, but in some cases they may not be allowing the type of innovation we're talking about. For example, there was a proposal from the Chicago Stock Exchange to introduce a speed bump for liquidity takers.
The idea is if a firm sends an order that is going to remove liquidity from the book, it's held up for a very de minimis period of time. It was 350 microseconds in the proposal. - [Budish]: It's one one-thousandth of the time it takes to blink your eye. That's small. - [Crutchfield]: It's a statistic I'm very fond of, exactly. So, a very short period of time. That proposal was suspended by a commissioner at the SEC, unilaterally, for an indefinite period. It's just kinda hanging out there. We're not sure if that will ever even be given a chance to become part of the competitive landscape. I would argue that for many liquidity providers, my firm included, the biggest obstacle that we see to providing better liquidity, and showing better prices for investors, is this risk of sniping or pickoffs, which is other advanced firms with very sophisticated technology are a microsecond or less than a microsecond faster than we are or another firm is at reading some event that's publicly available on a price feed somewhere and will cause us or another market maker to incur a trade that is immediately regretted. To the extent that that happens, market participants start to price that in, and they trade less aggressively, and the result is it goes counter to that improved investor experience.
- In your mind, it's not just the question of manipulation. It's actually a question of, is the price the best price, because of the way that the market works. - Absolutely, yeah. I'm not speaking about manipulation whatsoever, but I'm talking about professional participants causing each other to lose money. The cost of that ultimately is worse prices for investors. Getting back to the technology point, and the point about competition: to the extent that a small number of firms have invested huge resources in being able to operate in that environment where microseconds and nanoseconds matter, those firms have an advantage. Now, one could respond with a Darwinian question which is, "Well, they deserve to survive." The problem is, ultimately, that goes counter to those words from the Securities Exchange Act of 1934, "investor protection and the public interest." In order to enter the market right now, and provide better prices to investors, a professional trading firm has to solve a massively expensive technology program. Tens of millions of dollars at a minimum. That's basically the equivalent of a regulator turning to market participants, and saying,
"Oh, if you want to trade in the market "and provide liquidity and better prices to investors, "every time you respond to a market tick, "you have to compute the first 1 billion digits of pi." We could make that requirement, but why would we? Why impose this massive technology burden on top of—? - [Budish]: It's a tax on liquidity. - [Crutchfield]: It's not about pricing or risk management, which is ultimately what we want investors to be able benefit from. - [Bowen]: Right, but at the end of the day, though, it's really relative. In the sense that, with speed bumps, which I think will be allowed, you're gonna find everyone else having speed bumps. - [Crutchfield]: Absolutely fair, yeah. - [Bowen]: If you make it whatever time frame you do— - [Budish]: If the New York Stock Exchange were to adopt a well-designed speed bump, I would be delighted. - [Weitzman]: Seeing as we're all talking as though we know what this speed bump means. Eric, that's a perfect segue to you to give us your platform. You've written research that proposes slowing down trading. - My specific proposal, and this is in research I've been conducting since 2010 or 2011 or so.
You can think of it as a very specific type of speed bump, where time is treated as a discreet variable. So, time might be broken down into, say, milliseconds, into thousandths of a second. 23 million units of time per day. Orders that reach the exchange in a particular millisecond are all processed in a batch process at the end of that millisecond using an auction. Most milliseconds, most stocks, nothing happens. Sometimes in a millisecond, an investor shows up and wants to buy 500 shares of the ask, or sell 200 shares of the bid. In that case, the auction processes that request to trade almost exactly in the same way as the current market. The difference is if there's a burst of activity in a particular millisecond, which to Steve's point a few moments ago is typically algorithmic traders responding to some price signal, then processing that burst of activity in a batch using an auction ensures that competition is on price and not speed.
It protects liquidity providers from being sniped at, and trading at an adverse price, instead trading at a price that is a market consensus price based on an auction. It's reengineering competition at the millisecond level. - [Bowen]: But it seems to me that that's, in fact, creating the kinds of dark pools we were trying to get away from. I think what we really want are transparent, real-time trading markets. - There's no such thing as transparent and real time because it takes time for information to travel even from you to me, excuse me. It certainly takes time for information to travel from one— - [Bowen]: But do you think the regulators should decide what the millisecond or what the right time frame is? - You can reduce any economic argument, "I'll let the market figure it out." - What's the optimal to get to the best price? - What I would argue is that you can use some combination of technology and engineering facts to identify what's a unit of time at which markets cannot go faster
than that unit of time without creating relativity problems. That unit of time might be a half a millisecond? I've heard some exchanges argue that even if we process trades once every hundred microseconds, every hundred millionths of a second—that's fast— that's still long enough to batch process requests to trade that arrive at roughly the same time in response to roughly the same economic news. - [Crutchfield]: I think part of what it comes down to is a question of priorities while for participants. From a regulatory standpoint, we don't allow, generally speaking, shares to be traded in increments of a single share. They're called odd lots. There are special rules around odd lots, of course, but generally you have to trade 100 shares at a time, or one options contract or one futures contract. Generally, you have to trade in certain price increments. Options trade in increments of pennies, nickels, and dimes. Shares, generally, in pennies. You can go out in decimals in some circumstances. Futures, often in quarters. But for some reason, we have this view that on the time axis, we should allow an arbitrary amount of compression even when there's no further benefit provided to investors, but
there's just a gaming opportunity generally created. - [Bowen]: But that's the question. - Sure, at some point— - That is the question. - I absolutely agree. I think there's some level where, just as we've said, and the exact amount might not matter too much. Maybe it's 350 microseconds, maybe it's a millisecond, but there's some level where additional time competition doesn't help investors. One could probably demonstrate that it may hurt investors. It just causes market participants to just snipe at each other, ultimately resulting in inferior prices. Just as we've done when we said, "You're gonna trade 100 shares at a time, not 50, not 10." - Right. - [Weitzman]: Sharon, you said there will be speed bumps in your opinion. - Yeah, I think people will compete based on speed. I don't think speed bumps. - Do you mean in your view, this sort of arms race will continue? - I don't think it's necessarily a good race, to be honest. I think it just makes it a little bit more artificial, and it creates dark markets. I don't think it adds much in terms of efficiency. I, as a regulatory, wouldn't want a regulator to set what that speed bump would be. - That's because nobody's tried my speed bump.
- Well, there's gonna be speed bumps of all sizes out there before you know it. - [Budish]: There's some—. In market design, the devil's in the details. - [Bowen]: We'll see. If it were to create better prices, the proof will be in the pudding, in that sense. I just think it creates another opportunity for a dark pool to be out there, because— - [Weitzman]: Your point being that it would just go off the trade— It would go away from these regulated, transparent exchanges to somewhere we can't see it. - Where we can't see it. That bothers me. - [Crutchfield]: I absolutely agree that the proof is in the pudding. I would love to see somebody try it. Right now, at least on the SEC side of the markets— the equities and equity derivatives market— it's impossible because this stay has been placed on this one piece of rule filing. I would love to see somebody try it, and we'll see. It's certainly possible that the events will turn out differently, but my expectation would be some degree of speed bump would provide sufficient protection to liquidity providers that they would no longer need to price in these adverse trades that you referenced. - I've heard the phrase "unwanted trades." So if you're making a market, you have to worry about
having a trade that you immediately regret, and immediately measured in really immediately. Much faster than the blink of an eye. A trade that you're trying to get out of the way of, but someone else is a millionth of a second faster to snipe you. That professional-on-professional combat might seem like it just affects, it's zero-sum trading among professionals, but ultimately it becomes a tax on the rest of the market. It makes it harder for professionals to provide liquidity to real investors. - This is one very important issue that we've been focusing on. I will say, a note for the record, there are other regulatory concerns aside from the question of speed and technology. One of them that we spend a lot of time worrying about right now, the most important issue that nobody's ever heard of, in my opinion, is the way bank-capital rules adversely treat cleared options positions. In many cases, you can have a position with very defined risk, and the capital that a bank-affiliated clearing firm that houses professional options
traders trades at the end of the day have to set aside for that position is much larger than the maximum amount of money that that position could really lose under any circumstances. That's another form of tax on professional market participants that falls under the bank regulation rather than the CFDC and the SEC world. There are other areas where continued education of regulators is very important. - That raises another issue I wanted to ask you about, which is the separation between futures regulation and securities regulation— which, as you said Eric, this is so arbitrary a distinction nowadays. Presumably dates back to the days when pork bellies were completely different from shares in manufacturing companies. Nowadays, one market can ricochet into another. Does that make sense that division that you have in the US, which I think is unique to the US? - I think the division definitely makes sense, being our equities markets are about capital raising, and a way for companies to finance operations and growth. The futures markets have been historically used for
risk mitigation and price discovery. I do think they have different purposes, and having said that, of course, people do trade across asset classes, across time zones, across currencies, and I think that's fine. I think having the two agencies, I think they actually complement each other in some respects. I think, notwithstanding the history of it. if you kinda see where we are, if you look at Europe, they're still grappling with growing their capital markets union. Their companies basically rely on banks to lend them money to fund their growth. So it's different. In Asia, they're talking about lowering their standards to get more people to list. I frankly think our capital markets are the envy of the world. I think the system really does work. - Steve Crutchfield, what's your view on that? On the separation between securities and futures? - I actually agree with everything the commissioner said. I will say, though, that there are sort of
these muddled areas in the middle. You look at S&P 500 index options, there are securities options, there are futures options that are basically the same thing. The CME can't list S&P 500 futures options that expire on the third Friday of a month because CBO has those listed on the SEC-regulated world. It would be nice to have a regulatory structure where some of those middle areas ended up cleanly on one side or the other. On the question of fragmentation, though, and confusion resulting from regulation, that's almost a separate question from SEC and the CFDC. We have just in the SEC-regulated world, I'm not sure now how many equity-trading venues. I know there are 15 options-trading venues. There's a lot of fragmentation right there, and it would be nice to see some steps taken to help to encourage a better integration across these many, many markets because I don't think there's another country on Earth where there are 15 exchanges all trading the same thing. - Eric, do you have a view on that? On the distinction between securities regulation,
and futures regulation? - I don't have a strong view. It's never made sense to me, and I think the products that, there are these gray-area products that are not quite pure. I think, to the commissioner's point, there is a large economic difference between capital formation for corporations and futures contracts. There are a lot of products that live in the vast gray area in between. We were debating, or discussing, what happened in volatility products a couple weeks ago. The XIV exchange-traded fund, for example, or exchange traded note, excuse me. That's a product that's not about capital formation for Apple computer or Microsoft. That lives in the gray area. - But recently, even the recent Treasury report on the financial market concluded that the cost of merging the two agencies outweighed the benefit. This discussion happens every year, pretty much. The decision is the same: it's a bad idea.
- It's never quite made sense. - [Weitzman]: I was talking more in theory. If we could start from zero, we wouldn't do it, presumably. Does it make sense? - [Bowen]: That's a different question. You have to start from way, way back, though. - The old joke, "Wouldn't start from here." Talking about things that fall in between, I must ask you about cryptocurrency, which are currently a regulatory black hole. Should they be regulated? Should the US make a bid to regulate them? What's your view? - I don't think they're in a regulatory black hole. At my former agency, it deemed Bitcoin and other cryptocurrencies to be commodities, and I think that's the right answer. I think that the CFDC and the SEC have been pretty proactive in making sure that the investing public is aware of a lot of the false advertising, and the false marketing, and the various schemes to invest in these cryptocurrencies. I do think we don't regulate the cash markets, and the spot markets. There's some question as to whether
there should be regulation there. I think the regulators are doing a good job of taking notice, being very aggressive, particularly in the retail side, where they see retail investors getting harmed, which is the right approach. At the same time, not stifling possible innovation that could come from the blockchain technology or detail, for example. I think they're taking a good, cautious approach without stifling innovation. - [Budish]: One point I'd add to that. I agree with everything the commissioner just said. If you think about the formation stories of futures markets and of stock markets that you teach your students, whether MBAs or undergraduates, in futures markets you always talk about wheat contracts, and a farmer who wants to hedge the future price of wheat. When we're talking about teaching about stock markets, you talk about an entrepreneur who needs to raise capital, and you go to the capital markets, and you can raise capital for your venture. For Bitcoin, I haven't heard those kind of stories yet.
That are the real economic uses of cryptocurrency aside from buying fake IDs and drugs on the dark web. I'm waiting. - I know. We're all waiting. I think we're all waiting. - I'm curious. I think the computer science is cool, but I'm waiting to see a little more about the economics. I worry that a lot of the trading's gambling, and that there's gonna be a lot of retail investors who will lose a lot of money. - [Bowen]: Yeah, yeah. That's when we have to be very vigilant on the enforcement side. - Unfortunately, on that note, our time is up. My thanks to our panel, Eric Budish, Sharon Bowen, and Steve Crutchfield. For more research, analysis, and commentary, visit us online at Review.ChicagoBooth.edu, and join us again next time for another "The Big Question. Goodbye. (piano music)
Download YouTube subtitles and read quietly. Ssh! © 2020