Quantifying impact of retail off-exchange trading on U.S. equity market spreads

PETER HAYNES: Welcome to episode 37 of TD Securities' podcast series, Bid Out, a market structure perspective from north of 49. I'm your host Peter Haynes and today I'm joined by Hitesh Mittal, founder and CEO of BestEx Research. Hitesh and I will discuss the controversial practice of payment for order flow in US equities, including quantifying the impact on equity market spreads in the US of retail off exchange trading.

Before we get started, a quick reminder to our audience. This podcast is for informational purposes. The views described in today's podcast are of the individuals and may or may not represent the views of their firm. And, of course, the content of this podcast should not be relied upon as investment, tax, or other advice. Thanks for joining us today, Hitesh.

HITESH MITTAL: Thanks for having me.

PETER HAYNES: Excellent. Well, why don't we get started by having you tell us a little bit about your firm, BestEx Research, and what is the mandate of BestEx Research?

HITESH MITTAL: Yeah. So the mandate for BestEx Research is helping institutional investors minimize transaction costs. We provide execution algorithms, transaction cost analysis, and a platform called AMS that helps institutions build their own execution algorithms and customize their own execution algorithms. And before starting BestEx Research, I was at a firm, a systematic hedge fund called AQR Capital, and that's really where I saw the need. I felt like the existing providers of execution algorithms, largely banks, their platforms were a bit opaque. And their approach was not as systematic when it came to minimizing execution costs, and that's what we're trying to change. Our clients include hedge funds asset managers as well as other banks and brokers who also white label our platform and provide to asset managers.

PETER HAYNES: Well, I'm going to suggest that your firm was plunged into the spotlight last week as you published a paper, which caught my attention, entitled, quote, "The Good, the Bad, and the Ugly of Payment for Order Flow." The report is timely given that the new SEC Chair Gensler appeared on Capitol Hill as part of the third GameStop hearing, and clearly, based on the questions and based on his answers, the topic of payment for order flow is on the short list of action items for the new SEC leadership. What prompted you to dig into this topic of payment for order flow?

HITESH MITTAL: Yeah, as a startup, we really sort of are pretty careful about where we spend our time, but we are passionate about market microstructure and especially things that have an impact on investor costs. This practice has obviously been around for decades, but retail volume was not as big in the market, and recently due to COVID, or whatever the reason might be, the retail volumes have spiked up. And so I thought that this would be something that would be interesting to sort of dive into, especially with a lot of the data now available that wasn't available earlier, the 605 reports and whatnot. Also, secondly, I've read a lot of reports, and generally speaking, all the news articles and other articles that I came across were pretty extreme, right? Like, they were-- the guys were saying wholesalers are frontrunning retail, and for some reason, retail brokers are doing something unethical by ascending order flow to the wholesalers. And on the other hand, wholesalers and retail brokers are saying things like, this is the best thing that has happened, and they're giving away billions of dollars to retail in price improvement. And frankly, truth is always-- as we say in the paper, truth is always in the middle, and we wanted to sort of put a pretty balanced perspective out there, that there are some good aspects of it and there are some bad aspects of it. So that's really what sort of triggered us to write the paper, that we didn't really see many articles out there that were providing a balanced perspective, and we tried to do that with this paper.

PETER HAYNES: Well, it is timely because, as you mentioned, the amount of retail activity in the US market, and for that matter in Canada as well, has gone up exponentially over the past six months. And what I think we lose sight of here in North America is the fact that there's other markets around the world where retail dominates market activity, particularly in Asia. So to suggest that we're going to go back to the point where retail doesn't have an influence on price is, I think, an unrealistic expectation. It might happen, but it might not. It might be the case that we see a lot more retail flow. And in your paper, as I say, you mentioned the good, the bad, and the ugly. Why don't we start with what you define as the good of payment for order flow. Break down what's good about the practice.

HITESH MITTAL: I mean, the good Is that a lot of the stuff that is said about wholesalers and retail brokers is not true, right? I mean, yes, there have been some enforcement cases against wholesalers and retail brokers, but leaving those aside, there's a reason this practice exists, right? And it's not illegal. The regulators have blessed it, so they're doing it within a legal framework. The retail flow at the end of the day is not toxic. It's benign flow. If they go to exchanges, they would get the same execution as everybody else. So if they send a market order, they would-- in an exchange, a market maker can't really differentiate between the flow is coming from retail, or it's coming from an institutional investor, or if it's coming from a toxic frequency trading firm, or a startup firm. So they can get better prices by segmenting that order flow and give that order flow to wholesalers. So there's no denying that when retail brokers are sending market orders to wholesalers, retail clients, ultimately, of these brokers are getting better prices than what's available at exchanges. Yes, that's-- how much is that price improvement that's amplified in the marketing from wholesalers, but there is price improvement over what they can get on an exchange. So that's the good part of it. Relatively speaking, they get better execution on their market orders than an institutional firm can get by going to an exchange.

PETER HAYNES: So we've used the word "wholesaler." We should probably define it for our audience here, although many people will understand what that term means. The firms that fall into the category of wholesaler are a short list. For our audience, could you just name some of those firms? And why doesn't a Goldman or Morgan Stanley or J.P. Morgan participate in this market?

HITESH MITTAL: So it's firms like Citadel, where these are the biggest wholesalers. We looked at the top five wholesalers in the US equity market, and they represent about 25% of the overall equity volume. And they represent about, more importantly, if you exclude-- if you just compare their volume to what happens at the NBBO on exchanges, which is basically what we call lit liquidity, which is the liquidity that sort of drives prices that we all look at, they're roughly 50% of that volume. So of that volume, Citadel is roughly half of it, and Virtu is another 25% of those five top wholesalers. And wholesalers are basically market makers. So retail brokers like Robinhood and Fidelity’s, they send their market orders to wholesalers, and wholesalers commit to sort of filling those orders for them so they don't have to route these orders to the exchanges. And they guarantee that the price that the retail broker would get would be the same or better than what's available at an exchange. So if the client's buying, they'll get a price lower than or the same as national best offer. Yeah, so these are the wholesalers. I think the banks, to your second question, they also do have some retail flow, but this game is really concentrated. And that's something we talk about in the paper as well that a part of the concern is how concentrated it is, that one or two firms are controlling roughly the same volume as what's traded in the 16 exchanges, which is attracting hundreds of market makers and institutional investors and other firms. And one or two firms can sort of make up the same amount of volume as the exchanges.

PETER HAYNES: I know that during the height of the GameStop activity, most of the list of wholesalers had, at some point, a technology problem and had to rod away or had too much order flow and had to rod away with, what I was told, the exception being Citadel, being the one firm that stayed alive the entire time, didn't turn any order flow away. The question that comes back is, what if Citadel had a technology problem? What would that do for the wholesaling world? And as you say, with the amount of flow controlled by a small group of parties, if they did have an unforeseen technology problem, which can happen to anyone, are they at that stage now where they're too big to fail or they're-- Citadel's role in making lit quotes as well as a designated market maker on the NYSE and its role in the options market as well? You put it all together and they are a very, very big part of day to day activity, and some will argue too big to fail. So we'll see whether the regulators try to rein that in. But as you say or as some pundits in the market will always argue, if somebody has it really good, then in a zero sum game, what's good about segmentation for this one constituency-- in this case, we'll call it the retail wholesaler dynamic-- must mean that it's bad for others. So in your paper, you also describe not only what's good about payment for order flow, but the bad and the ugly. Tell us what the bad and the ugly are of payment for order flow.

HITESH MITTAL: Yeah, absolutely. And that's the tricky one, right? So generally speaking, if you go back to sort of the good part of it, the good part of it is, OK, retail investors are getting prices that are better than what's available at NBBO and exchanges, right? So NBBO itself is sort of used as a benchmark to figure out a good execution for retailers. If you look at that sort of narrative carefully, it makes sense if NBBO is fixed, right? But the problem is it's not fixed. If the retail flow itself moves to exchanges, that NBBO bid-offer spread that we call will itself go down dramatically. So the question is, if this retail flow were to move to exchanges, in absolute terms, would they do better than how they do today? So we use an example in our paper, and we posed the question, let's say that you wanted to buy a car. Let's say that you paid 20 grand to buy that car when your neighbor is paying 25 grand. You feel great about yourself that you're paying five grand less. But what if both of you or all of you could pay 18 grand for the same car? You're not doing relatively better than your neighbor, but you're all doing-- you're all better off than you were earlier, right? And I think that's the part that gets sort of ignored that, OK, how much will the bid-offer spread go down by if this flow were to move to exchanges. And if you think about it, it will go down dramatically because of many reasons. The first reason being that when market makers are making markets at exchanges, they are making market to what's not retail, which is basically the left overflow, which is the more toxic flow than retail. So they end up owning the spread, but they also end up losing a lot of the money because the prices move against them after the bid gets filled, the prices are trading down, and after their offers get filled, the prices are moving up, right? So they are interacting with more toxic flow, and that's why the bid-offer spread is higher at exchanges right now. So if this retail flow were to move to exchanges, that would mean that the mix of the flow, which is the retail flow and the other flow, would be less toxic than what it is today. And if it is less toxic, then it's simple math. The earnings spread, they are losing money in what we call adverse selection, which is the adverse price move after they get the fill. So if that goes down, the bid-offer spreads would automatically go down as well. And we actually quantify that in our paper that, hey, let's look at what's the average adverse selection? How much does the price move adversely against a limit order within exchanges, and how does that compare to the retail flow? And there is a difference of about 50% of the bid-offer spread. So you can see, like, it's basically-- on a round trip, if you add up the 50%, that adds up to the full spread. So it was actually quite interesting to sort of see that statistic that the bid-offer spread in exchanges is actually equivalent to the difference in the adverse selection between the toxicity from the flow that comes to exchanges and the retail flow. So if that adverse selection goes down, the bid-offer spread will go down. In our estimate, that bid-offer spread will go down by roughly 25%, which is a lot higher than the price improvement that retail gets by sending that order flow to wholesalers. And you have to now also think about, not only the retail investors are going to benefit if this flow somehow ended up in a centralized venue like an exchange. It's everybody else will also benefit from it because it's the institution and investors will benefit from it. So in terms of the total dollar amount of saving, it's not just 10% of the bid-offer spread that just went to the retail investor. It's everybody else. So the difference could be huge if this retail flow were to move to exchanges. And a lot of these institutional investors are really representing-- not all of them, but many of them, most of them, I would say, are representing retail investors. Like, when I'm putting my money in 401(k) with Vanguard or Fidelity, they are considered institutional investors. So everybody would benefit from it. So that's what we call sort of the bad part in our paper. And the ugly part is-- I think it goes beyond that, which we can't quantify within our paper because we don't have any data for it. But that has to do with the information asymmetry, and we can't quantify that. And the idea then is that the wholesalers, they are also making market within exchanges. And so now they have this information that they're getting because they're getting this order flow, and order flow is information. So they're making markets with that information at exchanges, and then you have other market makers who are making markets without that information. And this amount of the information that wholesalers have is huge, right? In some stocks, they could be more than the entire exchange volume that happens during the continuous hours. So that information asymmetry basically means that the other market makers are obviously going to lose a lot more money than just-- in exchanges than the wholesalers when they're making market in the exchange itself. And that is likely to drive the competition away, and we don't know how much effect that has on bid-offer spread. We can't quantify that, but that can't be good either. So that's the ugly part of it, that this information asymmetry, which is supposed to go away in a market that everybody is trading with everybody else and information is the same, is no longer the case with this concentration of liquidity with one or two wholesalers.

PETER HAYNES: Well, as you say, what I found interesting about your paper was that it is actually the first attempt that I've seen anywhere of trying to quantify the impact on stock bid ask spreads of off exchange wholesale retail flow. And we all know the NBBO is the lifeblood of price discovery. In fact, last week when Gensler was talking at the congressional hearings, he was mentioning some comments about the issuer community, and he remarked that, quote, "capital formation is better when there is intense competition in the middle of the market." And it certainly feels like, from the perspective of his perch, that there are some things that need to be cleaned up with respect to trying to make the market overall fairer and not benefiting one participant over the rest of the market. I don't think it's fair to criticize, and I certainly won't criticize, the wholesalers for the work that they're doing. And I won't criticize them for-- I certainly won't accuse them of anything untoward that the regulators haven't already found, because this idea of them frontrunning-- we know they're not frontrunning. They just have a lot of information. And I know, as you say, it's really hard to quantify the value of that information, but we can't begrudge the fact they have a lot of information and they have a lot of control. So does Google in terms of how you search or Facebook in terms of what you're looking at on social media. So obviously, these are concerns across multiple parts of the economy. But one of the misnomers that we hear a lot about retail flow is that it actually doesn't trade the same stocks. They don't traffic in the same securities that the institutional community that you're talking to in BestEx Research. However, in your paper, you did some analysis with respect to where the retail flow is active I think, in part, by looking at Citadel's breakdown of activity in the names that they're trading. Can you help our audience understand exactly where the breakdown of activity is with respect to retail and how much retail flow actually does trade in the same names that your typical institutions are trading in?

HITESH MITTAL: Absolutely. I mean, I think that, as you said, is a misnomer. I've heard some pitches from wholesalers that, hey, retail guys are trading this one stock, and I'm sure institutions are not trading that stock. So they take these one or two stock examples and try to make you feel like you don't need to worry about it because retail is trading something completely different. But that's not what the data shows, right? And we are looking at the data, the data report. We are looking at 605 reports that are created by wholesalers. And when you look at the data, I mean, retail is trading-- yes, I mean, they trade S&P 500 stock. They trade Russell 2000 stocks. I mean, they're trading the same stocks that everybody else is trading, right? And it's law of large numbers if you think about it. If they are 50% of the volume that trades at NBBO in all the exchanges, they can't be trading some stocks that are not created by institutions. So I think that that's completely a misnomer. Yes, there is some difference, and their concentration might be higher in certain types of stocks than other stocks. But we look at Russell 2000, which is obviously an area of concern for most market participants because the trading costs are pretty high, the bid-offer spreads are pretty high. And for Russell 2000 stocks, retail concentration is actually a bit higher than institutional concentration. So going back to if that retail volume moved to exchanges, for S&P 500, you probably don't care as much. Your average bid-offer spread is 2 and 1/2 basis points. But for Russell 2000, if that flow were to move to exchanges and if they are more than 50% of the lit volume, that could have a significant effect on bid-offer spread. And that's going to help institutional investors, but that's also going to help retail investors because the NBBO itself will go down. So the idea that retail is trading stocks that are completely different, I just don't see that in the data.

PETER HAYNES: Well, as I mentioned earlier, you're in the spotlight now. I've seen a lot of social media activity talking about your paper, which I know you've engaged in a little bit. Overall, tell us about the reaction that your paper has received from the various types of participants. We'll categorize the wholesalers are one community, regulators are another, exchanges are one, and then the institutional community. Talk to us about the reaction that you've gotten so far and maybe some of the compliments and criticisms that you've heard to date.

HITESH MITTAL: Look, I mean, overall, the reaction has been extremely positive. We mostly engage with the institutional investors community and institutional broker dealers, and the feedback has been, I would say, very good. It's mostly been about this is the first time they've read a paper that sort of doesn't really sort of call one party evil, but at the same time sort of talks about what will be the impact on NBBO spread. And I think that's been pretty much the feedback that we've received from institutional investors. And overall, as far as comments on social media, et cetera, that's exactly what we wanted to see. We wanted to present a different perspective, a balanced perspective, and I think it should be a constructive debate. We have accomplished that a little bit, I would say that we wanted to accomplish.

PETER HAYNES: I would agree that you've put this in the spotlight. And hopefully it is something that everybody should be reading and critiquing, and I'm sure there'll be counterarguments that will come from others. And that's exactly what we want to do is to stimulate discussion and debate in the space. You did publish a follow-up blog, and it was on the topic of why haven't-- if retail would be better off if their order flow is routed to the lit market, why haven't they done it so far. So you provide in your follow up blog some of the reasons why none of the brokers have chosen to route to lit, even if, ultimately, it would be better for their clients. And in the blog, you used the classic prisoner's dilemma dynamic to answer that question. Can you explain?

HITESH MITTAL: Yeah, absolutely. And that really came from-- someone commented on social media that, hey, Interactive Brokers, which is a big retail broker in the United States, they send their order flow to exchanges and their clients bid-offers, clients did not get the 25% price improvement by going through exchanges, and that was just the wrong interpretation of the paper. Because in the paper, we said if all retail flow goes to exchanges and the bid-offer spread itself will go down by 25% doesn't mean that they will relatively do better than the NBBO. NBBO itself will go down by 25%. So I took this as an opportunity to sort of go with this other concept that we were debating whether to put that in our paper. And then the paper got too long, so we sort of decided to stay away from it, which is that one retail broker can't, in isolation, just decide that they're going to send their order flow to exchanges. Because if they do, yes, the bid-offer spread will go down by a little bit. It's not going to go down by 25%, number one. But more importantly, if other retail brokers continue to send their order flow to wholesalers, they will still get price improvement over what's available at exchanges. So it's a huge competitive disadvantage for that retail broker, that they're taking the high road and they want to get to what is going to be a globally optimal outcome. But in comparison with other retail brokers, they would be worse off, and that's why we sort of equate that to prisoner's dilemma where-- the classic game theory scenario where there are two prisoners and they're given a choice. They can be silent and they will only get one year sentence, or they can betray each other, in which case they get two years sentence. And obviously, they should stay silent, but they don't because there is the third scenario where if one were to betray the other and the other remained silent, then the silent one will get three years and the one who betrays will go scot free. So this game is very similarly set up. It's basically these are network effects, and that's why it's hard for one retail broker to just move to exchanges. It has to be coordinated and you can't really coordinate, right? The solution cannot come from one retail broker deciding to just move to exchanges. The solution has to be a regulatory solution that either incentivizes all retail brokers to move their flow to exchanges or uses-- it's basically characteristics, right? But the solution has to be a regulatory solution. It can't be just a solution that one retail broker can come up with. And also, I wanted to emphasize that that's why retail brokers should not be blamed for the fact that they're sending order flow to wholesalers. And It's important to sort of also separate the decision of sending order flow to wholesalers from the decision of getting payment for order flow. Yes, I mean, that is completely in their control. If they don't take payment for order flow, they will get a better price improvement for their clients. That's up to them. But that doesn't mean that the flow will not go to wholesalers. So just banning PFOF or not taking PFOF voluntarily-- and a lot of the retail brokers don't take PFOF in the US equity market and they still send flow to wholesalers-- the end result which we want, which is, OK, let's get the competition going within exchanges, let's get better price discovery, and let's get the bid-offer spread down for everyone, that is not going to happen just by banning PFOF.

PETER HAYNES: Like you said, I think Doug Cifu from Virtu has commented about just this. Out of the 150 or so retail brokers, only a handful actually take payment. So it is important to understand it's not necessarily-- in many cases, it's for the convenience and execution simplicity of routing to those wholesalers. But one of the topics that comes up a lot when we talk about PFOF, or Payment for Order Flow, is the accessibility of order flow. And actually, last year at the fall STA Market Structure Conference, virtually, SEC Commissioner Roisman floated the idea that retail trades should be marked on the trade blotter, or the TRF, so that other participants can explicitly know what flow is not accessible and therefore separate that flow as part of TCA analysis. I'm sure that's something that your firm talks about with institutions. When I spoke to one institution about this idea, their answer was pretty direct, and it was as quote, "we don't want disclosure. We want access to that flow." Now, how would you envision an equity market ecosystem absent payment for order flow, where retail flow is executed on exchange?

HITESH MITTAL: Yeah. I couldn't agree more. It's like watching someone eat in a great restaurant, right? So it's not just about transparency. It's about how can you access that order flow. I mean, it's a very complicated subject. And we have a lot of rules in US equity markets, and these rules have been implemented in various parts to address certain issues. And they have created a lot of complexity. And at the end of the day, you now have two sets of rules, basically, which is partially responsible for the market structure that we have right now, right? And so if you think about it, like, wholesalers, for example, they can provide price improvement and that doesn't have to be a multiple of pick size. They could give you a price improvement of 0.0023 and that's absolutely legit. A regular market maker in an exchange or in an ATS, they can't do that. They can't quote the limit prices that are not multiple of pick sizes. So that's not a level playing field. The second is, the wholesalers can segment order flow, but exchanges cannot do that for the market makers or institutions that are providing those limit orders within exchanges. So there are different rules for different venues, if you will. And it's not like these wholesalers are just some market makers. They have basically become venues. They are bigger than every single exchange, or many exchanges together, but there are different rules being applied to how they can provide liquidity and how exchanges can bring in multiple market makers and institutions to provide liquidity within exchanges. So that's an area that regulators are going to have to look into. They really have, like, two ways to address it. One is, like I said, to basically ban things. And the second is, OK, let's maybe unwind the rules that we have and see if we can make them more similar across venues and let sort of the competition in the market sort of naturally get to a point where this problem goes away. But it's a complicated topic and I think if regulators are interested, they're going to have to think about all of those different types of solutions, and engage the industry, and get their thoughts, and come up with something.

PETER HAYNES: I think one of the most common refrains from political leaders on one side of the aisle during those three hearings was an introductory statement by saying, we have the best equity market in the world. Why would we want to try to put rules in place that might disrupt that? And I'll step back, coming from Canada, where we're obviously a lot smaller than the United States, and just looking from outside in, say, no matter what, the US is going to have the best equity market in the world. It's 50% or more of the market cap. It's 70% or more of the volume in the world. It's always going to be the best. But it doesn't mean the structure is the best today and it can't be improved. So the fear mongering I find that I hear from time to time, which is, if we change something, we might no longer be the best market in the world. We might have capital flow elsewhere as a result of that. I find that defensive refrain to be a little bit insular. And what I like about your paper is that it is a debate about the collective good as opposed to the individual good of one set of participants. Now, it's a very complicated topic. You mentioned that. We've learned that in the last 40 minutes. So to suggest that you just simply end payment for order flow and the related practices such as rebates on exchange, that's radical. And some pundits will argue that if there are going to be changes to payment for order flow, that they'll be staged, and the first stage would be improved disclosure. You've mentioned several times 605 and 606 reports. That's what people think can be the low hanging fruit. Do you agree that maybe a staged approach to payment for order flow reform, starting with disclosure changes, would be the first step, and is that a step in the right direction?

HITESH MITTAL: I mean, look, more disclosure is always better. Right? it's also the cheapest sort of form of regulation. And anytime you come up with regulation, you have to be careful. Ultimately, it's tax money from folks like you and I. So yeah, I mean, you can't really-- like, if you can solve something with more disclosure, absolutely. But I think disclosure alone is not going to fix this problem. Like I said, it's not about transparency. It's about access, that 50% of the flow during the continuous market goes to one or two wholesaler and they have that information advantage and everything else. That issue is less about transparency. It's more about market structure. So I think regulators are going to have to look into market structure reforms if they really want to address this issue.

PETER HAYNES: Just as we finish up here, again, I'm going to repeat you got the ball rolling here and let's see where it goes, whether it continues to roll and the snowball gets bigger. But you've put some quantitative data behind the impact of retail segmentation on the lifeblood of markets, which is bid-offer spreads. I'm sure that others in the industry will counter your research with their own, and that's great. And this becomes sort of a growing topic and one that more people are educated on. As you mentioned before, it's going to be up to the regulators. So as we sit back here and just finish up, what advice do you give to regulators as they try to mediate the debate? There's obviously a lot of vested interests, and do you think that there's any realistic chance that payment for order flow is banned during the Gensler administration?

HITESH MITTAL: I mean, I can't speculate. It's certainly on his mind based on his testimony, and I think I was kind of happy to see that he wasn't just talking about a conflict of interest in PFOF. I think he was talking about it holistically, more holistically, as he should be, because I think the issue goes beyond PFOF. The issue is more about is it OK to have one or two market makers get 50% of the market flow, and that goes beyond PFOF. And he talked about concentration. He talked about the effect of concentration and effect of information asymmetry on the entire market. So I felt like-- and he's an economist, so I felt like he is, and I'm sure the SEC staff that's working with him, is thinking about it the right way. So I'm optimistic, but it's a complicated topic so I can't really speculate.

PETER HAYNES: Well, I will say my feeling watching Gensler in front of the House Financial Services Committee, that he was thoughtful. He admitted where he'd only been on the job three weeks and wasn't going to make an opinion, but he was clearly well informed. And I thought he spoke incredibly well. So I think we should all feel confident that the decisions that ultimately get made under his leadership are going to be thoughtful decisions and done, I'm sure, with the best interests-- as you say, the holistic best interests of the market. Well, Hitesh, I really do want to commend you and your team at BestEx Research for putting some quantification behind this debate about segmentation and holistic go-to-the-market. So thank you very much for joining us today and we look forward to watching this debate unfold over the coming years.

HITESH MITTAL: Thanks, Peter. I appreciate the opportunity.

This material is intended to provide commentary on economic, political or market conditions.

Not Advice: The information contained in this material is for informational purposes only and is not intended to provide professional, investment or any other type of advice or recommendation, or to create a contractual or fiduciary relationship. Neither TD Securities (USA) LLC (“TD Securities USA”) nor any of its affiliates (collectively, “TD”) makes any representation or warranty, express or implied, regarding the accuracy, reliability, completeness, appropriateness or sufficiency for any purpose of any information included in this material. Certain information may have been provided by third-party sources and, while believed to be reliable, has not been independently verified by TD, and its accuracy or completeness cannot be guaranteed.

Not Securities or Derivatives Research: This material has not been produced, reviewed or approved by TD’s securities or derivatives research departments. The views of the author may differ from others at TD, including TD securities or derivatives research analysts.

Not Independent: The views expressed in this material may not be independent of the interests of TD.