Bonds

Automation, algos and odd lots — an evolution of electronification

Enjoy complimentary access to top ideas and insights — selected by our editors.

Transcription:

Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio for the authoritative record.

Lynne Funk (00:03):

Hello everyone and welcome to another Bond Buyer podcast. I’m Lynne Funk, executive editor at the Bond Buyer, and joining me today is Chad Wildman, executive director, quantitative strategies at FMS Bonds and Matthew Smith, founder and CEO of Spline Data. Welcome to you both.

Chad Wildman (00:20):

Hi Lynne. Thanks for having us.

Matthew Smith (00:22):

Yeah, good to be here.

Lynne Funk (00:25):

Excellent. So for our listeners, I’m going to give a little background on both of my guests and then we’ll get into the thick of how both their experiences are leading the muni market into a more technology driven one and why our listeners and the muni market at large should care. Chad has extensive experience in muni electronic trading and quantitative analysis. Prior to joining FMSBonds, Chad was co-head of electronic trading and the head of quantitative strategies for the muni securities trading desk at Morgan Stanley. Chad was also previously trained in the analysis of partial differential equations and quantitative finance. Chad has a bachelor’s degree in mathematics, a master’s degree in mathematics, a PhD in mathematics, and a master of financial engineering from the University of California.

And then Matthew grew on his deep experience with machine learning and muni bond price modeling to start Spline Data in 2022. After serving as the head of trading at Headlands Tech global markets, later acquired by TD Securities. Matthew created Spline data, which provides real time quantitatively driven pricing and yield curves for the muni market. Matthew created Spline to provide professional and institutional traders with actionable and innovative market data covering broad swaths on the market, including odd lots. Matthew is a graduate of North Carolina State University with a bachelor’s in finance. So yeah, your backgrounds are quite impressive and a little intimidating to be honest.

Chad Wildman (02:01):

Sorry that you had to read all those out loud. I went to school for way too long.

Lynne Funk (02:06):

Well, it’s also not often that I’m interviewing two muni quants and definitely not on a Friday morning as we are recording this.

Matthew Smith (02:13):

It’s not often that there’s two in the same room.

Chad Wildman (02:16):

Is true. Exactly.

Lynne Funk (02:19):

So then with that said, actually this works out well. I’d love for you to talk about a bit of how you two came to work together, what it exactly is that you have built together and why.

Chad Wildman (02:30):

Yeah, absolutely. So yeah, this is Chad. Back in 2016, 2017 type of timeframe, the muni quant world was pretty small. You kind of knew everybody in it. You certainly knew who your primary competitors were and Matthew was my primary competitor. But it’s a good asset class. People are nice, people are friendly. I think one of favorite things about munis is the people that work in it and all the different job functions. So you get to know folks and Matthew and I spoke on a number of occasions and became friendly. So it was kind of natural to work together later on and later on in the game.

Matthew Smith (03:21):

Yeah, I mean I’d say the same thing. It’s a small world for munis and I think the quants tend to somewhat seek each other out, but when they do connect, the rapport is kind of immediate and the conversations that we have are very different than the typical muni conversation. I think the first thing Chad and I spoke about in depth were databases of all things. So it’s definitely, we’re kind of on an island, so it’s nice when a new castaway joins

Chad Wildman (03:55):

Well put.

Lynne Funk (03:57):

Alright, that’s awesome. Yes, it is. The muni world is an island in itself, so yeah, you guys are an extra small island off the island. So can you talk about Chad, you’ve created a new algo for FMS Bonds and you’re using Spline data. Can you talk about exactly what that product is and how it’s working out?

Chad Wildman (04:17):

Yeah, really, really exciting stuff coming from a much larger shop like Morgan Stanley where you have a large dedicated technology workforce and all that kind of thing. This is a much different gig for me where I’m doing everything by myself. It’s kind of a one man show, and when I found out that Matthew was releasing this pricing product, I got really excited. So we spoke about it, I tested it out, kicked the tires, it’s an awesome product and was pretty much a no brainer to go with it because it allowed me to kind of jumpstart what I was doing here. And so what I’ve built is just built up around Spline as the direct pricing input. So there’s still plenty of work to do even if you don’t do the price formation part yourself in house. So a lot of my time has been spent working on fixed sessions for trading connectivity and message exchange between counterparties and platforms and a lot of the post-trade stuff, which I think we’ll get into a little bit later when we’re talking about regulations and stuff like that. But yeah, basically once you have a price — I mean I’ve said this before in the past when I’ve been on panels — price formation is possibly the most important part, but it’s not the only part. There’s a ton that you still have to do to solve for a very, very diverse set of workflows and modes of execution and ways that people want to trade muni bonds with you. So it’s been very fun, a lot of work in a short amount of time, but very excited to have gone live.

Lynne Funk (06:13):

Excellent. We’ll get more back into what makes algos different from different firms, but I want to take you guys back a step and the effects of automation on the market generally, on dealers. How much has it grown since pretty much I guess best ex took effect that be the marker of when things started to take off in the space?

Matthew Smith (06:36):

And I can speak to that first. Yeah, I don’t know exactly what the exact point in time was, but it seems like around 2017, 2018 is when everything started to gain to get some legs. I think it changed the market quite a bit and I think the way that the market has progressed until this point, and I’m sure we’ll get into this at some point, but SMAs and ETFs and some of these new vehicles for trading munis to some degree only exist because dealer automation has grown so much. And I would say the liquidity is probably pretty similar for blocks, but in the odd lot space at least, I think bid stacks in 2018 were probably six or seven for a slightly obscure mini, which is basically all of them.

And now, I don’t know, I would imagine it’s double that and a stat that I always kind of try and point out to people and I wish I could show it, but if you go on EMMA, there’s this market statistics page and you can see trade counts versus trade volumes. And if you go all the way back to 2014 or whatever, trade counts and trade volumes are flat, they have their spikes and their dips, but there’s been this massive explosion in trade count over the past three years and volume has remained relatively the same. And I think that that’s probably a good indication that automation in this space has made the market a lot more liquid, made it a lot easier to trade odd lots.

Chad Wildman (08:11):

Yeah, I would actually add to that any muni person worth their salt of course spends plenty of time on the EMMA website and market statistics and whatnot. Another thing in that same vein, I mean that’s absolutely right. The trade count has exploded in large part due to the increased SMA activity and the other automation. But something that I always found interesting, you would look to this and people sometimes will sort of talk about the portion of the MSRB trades that are printed on alternative trading systems, ATSs or ECNs as we sometimes refer to them, they’ll point to that as sort of a proxy for electronification, but I don’t really think that that’s right because that number is increasing of course, but it hasn’t really increased that much over a 10 year time period. And what is actually happening is that the automation, the increase in electronification is happening at dealers and in-house at clients to varying degrees of difference. But you don’t necessarily see that reflected that way in the MSRB trade data, which is something interesting, but it has gotten a ton better.

Lynne Funk (09:27):

How much of a saturation, or maybe that’s something that, how much a percentage of the market do you think is actually electronic at this point?

Chad Wildman (09:38):

It’s tough to say. It’s obviously going to vary a lot. I think it probably doesn’t vary too much amongst the major dealers. The rule is usually sort of an inverse one where you have a high percentage of trade count is electronic and a corresponding low percentage of trade volume, and then it sort of flip flops for voice. But I mean Matt, would I be crazy if I said something like 70 plus percent of ticket count is electronic? You think that’s too high?

Matthew Smith (10:12):

It’s probably a little bit high, but I mean I think that’s right. And that’s a good point on looking at what happens on an ATS. It’s not the best proxy for automation and ultimately depends on what you consider automated or what you consider electronic. Exactly. If it’s pure machine getting information, spitting out a quote, I would guess it’s closer to like 30%. But with any degree of automation I could see 70% being very likely.

Chad Wildman (10:44):

Matt, you make a really good point because people, again, just the words, they get minced quite a bit. People talk about electronification and I do like to separate automation versus electronification. I mean muni trading has been electronic for 20 years or something. You can have a trader clicking a button that executes a trade, it sends a message, trade gets booked and all that stuff, but they clicked the button, right? They’ve typed in a price and they clicked. And so you can’t really observe anywhere what the actual level of automation is. But yeah, I think you’re probably right.

Lynne Funk (11:19):

Alright, so can you talk about liquidity in this market as always, or rather, liquidity is usually the keyword. How do you go see automation aiding the market, making it more liquid?

Matthew Smith (11:35):

So I think liquidity is another one of those kind of blanket terms. I think traditionally a lot of people kind of look to volumes as some kind of indicator for liquidity, but I also like to take into consideration when thinking about liquidity bid, ask spread and velocity of trading as a proxy for bid ask spread. Because if you’re an asset manager and let’s say that your asset class agnostic, when you want to go into a new asset class, one of the first things you look at, it’s like how easy is it to execute and how cheap is it to execute? And so I think that has gotten the opportunity cost, or let me say rather that the cost of doing business in Munis has gotten a lot lower and it is very telling that Chad is a one man show, but he’s probably quoting a massive amount of the muni universe.

I don’t think that was possible just five years ago. So volume hasn’t been gangbusters or anything, but I think bid-ask spreads a really tightened relative to the rate backdrop and especially since where they were at in 2016 and it is just bringing more people into the market, at least from the asset manager side, whether that’s existing participants creating new products that’s also new liquidity in the market, but also dealers becoming more efficient despite there being some consolidation in that space.

Chad Wildman (13:21):

And this is a place where the sort of net effect of regulation I think was good for liquidity. I mean certainly from the point of view of a retail investor for whom the regulations were designed, but it just sort of forced the market to change the way that it operates and exposed order flow to everyone. So that was the point in time I think at which electronification and automation really started picking up because it was retail sort of oriented ticket flow that means lots of tickets and smaller sizes and you kind of have to dedicate technology to prosecuting that trade volume. So yeah, the regulatory effect I think has been positive on liquidity and I agree with what Matt has said too.

Lynne Funk (14:18):

So you talk about how some of these regulations have helped with retail. Can you go a little more detail into that and explain to the listeners why?

Chad Wildman (14:27):

Yeah, absolutely. I’ll take first crack at that. So the regulation G-18, which was best execution for retail, that was the one that really opened it up because what it did is it said that for any appropriately defined retail investor, dealers have to use reasonable diligence to ascertain that they are getting the most favorable price. It’s a totally logical rule to have. The implications though, were very far reaching and it required pretty massive changes in technology and workflow for muni dealers across the board. So all of a sudden you have to, if you were only connected to one of the ECNs or if you didn’t have full straight through processing of trades, all of those problems had to get solved. So it kind of heralded this period of, okay, we have to dot all the Is and cross all the Ts and get all our connections in order, make sure everything’s working.

(15:37):

We got to look at data capture. So data is obviously Matt and I’s bread and butter. The first thing I think we ever talked about was databases and we love the data, but you have to have appropriate captures of data in all of your trading systems in order to after the fact demonstrate compliance with these regulations and stuff like that. So that’s kind of that one. And then two other ones that I think are interesting that are coming up. We have trade reporting shortening to one minute from 15 ostensibly, and we have a settlement changing regular settle changing from T +2 to T + 1. And so those ones are a little bit more kind of back office oriented, but nevertheless, there’s technology problems that are going to be sort of put into the forefront as a result. And trade allocations and all those sort of things, a lot of people don’t really pay attention to them at all because there’s some team somewhere that does it for them, but that’s another bonus. I dunno if it’s a bonus or a curse for me being by myself here, got to deal with all that stuff. And there are good solutions for these problems. There are CTM and DTC are doing all the right things in order to make it easy to integrate and process allocations and get trade settled on time. So there really should be no reason that people should have too much trouble with this one, but it’s not a non-trivial amount, non-trivial amount of work for folks.

Matthew Smith (17:25):

Yeah, I’m not as well versed in the regulatory space, but Best Ex was definitely a marker of change and at least if you look at the timeline, you see it kind of a lot of things trade in the muni space after that. Chad mentions T +1 and the reporting time cutoff. Those are interesting. I definitely see them forcing people to automate with the crowd. And I guess the way that I look at it is the entire market moves in that direction naturally people just get better. Software comes out, data comes out that make it easier to settle trades faster, to report the trades faster, and then what they’re doing just naturally kind of lends itself to that. But then it’s almost like the regulatory authority comes in and puts this cutoff. They’re like, Hey, this is where most of the market is or this is where we think most of the market is. Everybody needs to get ahead of this line to continue to participate. And I think it’s good to an extent it kind of establishes a bar, but that transition period can be kind of tough for definitely a lot of the tail end of the participants. But overall I think moving in that direction is kind of the right thing. You don’t want to get left behind and you don’t want to have a thousand different processes for every counterparty that you have.

Lynne Funk (18:56):

And we’re going to take a short break, but we’ll be right back with Chad and Matthew and we’re back. So few both. You mentioned that data is your bread and butter. There’s a lot of data out there. How much of it is good? What does data even mean? Can you guys get into that for me and for our audience?

Matthew Smith (19:23):

Yeah, I’ll start with this. I’m the most biased. I kind of look at data as the fuel for automation. If you don’t have a wealth of data to draw on and software both, then to an extent mass market automation isn’t really possible. Obviously I’m focused more on the pricing side. I think in terms of what’s already available out there. There’s a lot of stuff off the shelf that you can build automation into your system. But in my eyes, and the whole reason I started spline was the piece that was always missing was quality baseline pricing to build off of everything right now out there is skewed towards round lots, which is great. That part of the market needs to be serviced and I think is pretty well serviced at the moment, but the odd lots were kind of always forgotten despite being 95% of the trade count or trade activity in the muni market.

(20:29):

So really what we set out to do was create a price not only for the round lots, but for the odd lots and with the goal of making it easier for people to traffic in odd lots. I’d have conversations four years ago where it’s like we wouldn’t touch odd lots, we’re not going to make enough money on those. And just the cost to get a price out into the market for them is not worth it. They’d rather focus on the round lots. And I just think that they created a lot of value in the odd lot space and subsequently attracted people. And now that I think that there’s a way to traffic in them a little bit more easily.

(21:16):

It’s kind of interesting, the data question more broadly. There is a lot of data out there and I kind of look at it on a spectrum. At the lowest level, you’ve got things like master or reference data just knowing what a bond is and then you have things as niche as ESG scores and stuff like that. So I think what data is useful in general is probably all of it’s useful if somebody’s willing to pay for it, it has some use in the market, but whether or not your firm is considering using ESG scores depends on really the maturity. So I think people start with the basics, reference data, pricing ratings, and then they kind of work their way towards needing more niche nuanced data to add edge to their strategy. But just getting over that first hump, it’s still kind of a challenge and hopefully we can make that a little bit easier for folks.

Chad Wildman (22:20):

Well said. Matt, having just kind of sat down and done everything myself, it brings into really sharp focus that you can’t really do anything without data, how you store it and what you do with it. There’s a bunch of different ways to skin that cap. But like Matt said, security master, if you’re going to do something electronic, the first thing that you’re going to want to do is understand the bond universe because irrespective of what you own, you have to sit there ready to quote a BIC or RFQ anytime that one comes in. And it could be for any bond. So there’s that. There’s pricing and like Matthew said, there’s lots of cool new stuff coming out. There’s ESG scores, there’s some companies that are doing even news, which I have thought was pretty cool. I looked at a couple of those in the past and they all come with challenges, something like a newsfeed in order to make it useful.

(23:35):

Well, the data itself of course is useful if you read it right, but in order to make a machine read it, you’ve always got to associate it to a bond identifier or issuer or something that you have in your data warehouse so that you can say, okay, now this is actionable, like machine actionable I should say. That’s I think one of the challenges that some of these new data sets have. ESG scores of course are already sort of I think associated to CUSIP, so it’s a little easier there. But yeah, I mean it’s really interesting. I haven’t gotten a chance to test any of those out in the wild, but definitely something that I think is kind of exciting. And like Matthew said, the spaces, I don’t think it’s overcrowded, but there are a lot of people electronically quoting bots now, so it will become harder and harder to differentiate as time passes and everyone gets better.

Matthew Smith (24:33):

And something I want to mention, just it always comes up as a topic of conversation is like I always get the question, what if everybody uses your price? And I don’t think that goes for really anybody in the market. What if everybody uses ice? What if everybody uses so-and-so other pricing service. And I think the point that I always try to make is that if you had theoretically perfect price for everything, if you could predict where everything is going to trade, does that mean you’re going to go in and trade everything? No, because everybody has different balance sheet limitations, they have different risk limitations, credit opinions they have, there’s a hundred thousand factors that go into why you might want to buy something. So I try not to give the impression that Edge is going to just evaporate in the market because what we’re trying to do is make it easier for you to express and realize your edge in this space and not so much try to sell your edge basically.

Chad Wildman (25:41):

Yeah, Matt, you’re absolutely right. The intention is not to take a pricing source that prices 75 or 80% of the universe like spine and just use it out of the box on every ticket. It’s sort of incumbent upon dealers to build around it and decide stuff to quote and what stuff not to quote. And if I just took it and quoted everything, I would blow through my risk limit in probably an hour. So there’s those considerations that are very important and that kind of dictate how you use the price.

Lynne Funk (27:01):

Very interesting stuff here. When we look at the muni market at large, a lot has shifted in the past two years, particularly with funds, and we’ve seen a tremendous growth in SMAs and ETFs and automation impacted that space. What are you both seeing there?

Chad Wildman (27:22):

Yeah, absolutely. I think I would touch on SMAs first, and I know Matthew has a lot to say about ETFs. So the SMA counterparties are exactly the type of clients that need to do lots of small tickets. So they are sort of the best users or best recipients of fully automated workflows on the dealer side. And I would say that there’s sort of a huge variance in what the different SMA counterparties do. Almost all of them have great solutions and are very good at doing things electronically because they have to be, it’s by the very nature of that type of account, you’re going to be doing a lot of tickets and you need fast responses, you need stuff automated, you need to not have to handle these tickets manually. So that has kind of been gone hand in hand with the increase in automation, I would say.

Matthew Smith (28:32):

Yeah, it’s hard to say it’s chicken and egg problem who drives what automation, but I’m happy to see both of them growing together. SMAs are cool. I think it’s representative of a broader trend in financial markets in general is that people are just kind of seeking out optimization. Passive investing for the longest time was the hot thing, but now people want the possibility of generating returns outside of that or doing something that suits their particular situation. So I think it’s really cool that this kind of product is entered to space, and I think that also applies to ETFs to some degree, and I don’t think any of these things would be possible without the degree of automation that exists in the market. ETFs for me, and everybody talks about how much both of these products have grown SMAs and ETFs, but ETFs are, I think they’re going to get a little bit bigger than where they are today in the muni market.

(29:48):

They’re pretty flexible. They can do both things structurally. They can act like a mutual fund go broad, or they can be kind of niche and capture value in some of the more obscure parts of the market. And my weird take on this, and it’s because this is how I get exposure to the market. I use a robo-advisor. I don’t want to think about it. It’s weird being a trader and not trading, but I just don’t want to think about it. Currently, they offer exposure to munis through ETFs and they will tax loss harvest with different muni ETFs. And I was talking to somebody a couple of weeks ago and they said they quoted somebody at one of the Robo-advisors and they said that we’ll never offer direct exposure to Munis, get your face ripped off in execution. The quote, as much as I butchered it was from 2014. I think that they would probably say something different now, but even if they maintain that, then that’s going to drive a lot of flows into ETFs because that’s a very high growth segment of the market is rowboat advisors all they’re going to drive money into the ETFs, and I’m sure many the existing ETF players and new ETFs will pop up to fill that void and offer more direct exposure if not direct to muni exposure.

Chad Wildman (31:36):

Yeah, the ETF space is super interesting, and again, lots of pretty ripe for technology and automation, and I think a lot of the major authorized participants in the space are looking at their business and realizing that they’ve got ETF trading with equity trading and they’ve got their bond trading and their fixed income trading that are two totally different deaths. And you kind of want to bring those things together, obviously to leverage to the maximum extent possible what you’ve got in technology for both sides. And you run into interesting things like Matt was saying, if you look at, you pull up MUB, I think by assets under management, the largest muni ETF by iShares, you pull up the holders for that and instead of your traditional clients that have longstanding relationships with the firm, all of a sudden you see these betterments and invest nets and wealth fronts in the top 1, 2, 3 holders of the outstanding shares. And that’s cool and exciting, but it’s interesting. It’s interesting because a lot of places don’t necessarily have a couple relationships with those types of counterparties. So there’s definitely a lot of room to grow and find efficiencies and synergies.

Matthew Smith (33:07):

And Lynne, you asked about automation as it pertains to ETFs. The final point on this is all of these ETFs are seeing inflows or a bunch of money coming into the fund. They have to go out and put that cash to work and they have to do it quickly and more effectively than their competitors. So the only way that they’re going to accomplish that is through some degree of automation. You can hire out of that problem to a degree, but all of these ETFs operate in a very low fee space, so to some degree you have to automate

Lynne Funk (33:47):

What does it mean for mutual funds. We saw two years of, I know that they’ve really come into the SMAs and ETFs have come into that space. Where do mutual funds stand for the market?

Matthew Smith (33:59):

I don’t want to speak too definitively on this because I just haven’t followed it that much, but I saw something a little while ago that said some mutual funds were turning into ETFs, which is kind of interesting. It is effectively the same product. It’s probably a little bit lower fee, but it offers the end user a little bit more liquidity than maybe an SMA if it’s something that they want to get in and out of. I don’t know if that trend will continue, but it seems like mutual funds in general have been losing market share to ETFs in other parts of the market, so I don’t see that being any different in any munis.

Chad Wildman (34:41):

Yeah, I don’t think that mutual funds are going away. Nobody can deny that there’s been a flow of money from them to ETFs and potential conversions and stuff like that, but it seems to me that there’s still a reasonably solid base of these funds and people who buy them. But yeah, it’ll be interesting to see how it progresses.

Matthew Smith (35:11):

For the record, I own some mutual funds, so I’m not completely sure.

Chad Wildman (35:15):

That’s too.

Lynne Funk (35:18):

Okay. Could you guys maybe just talk about we came off these two years, well, 2022 was a mess, and then last year, the rally at the end of the year, what are you thinking just bigger, broader for the muni market coming down coming into 2024? Headwinds? Tailwinds?

Matthew Smith (35:37):

I think we’re in watch the interest rate environment mode. A lot of what’s happening in Munis is driven by the underlying interest rate environment, and then you see these oscillations of relative value just based on, in my opinion, I think it’s issuance driven. So if there’s not a lot of new bonds coming to market, munis are rich relative to treasuries, but cheap historically because rates are higher, I think the easier it is to transact in munis as we continue to grow in that direction. We’ll probably skew more towards the rich side of that, but at the same time, it’s really interesting to see where munis are rich and it’s really in the front half of the curve. It seems like there’s been persistently a lot of value, at least by the treasury ratio metric in the long end. So if somebody comes into the space and figures out how to transact there, maybe we see some flow shift out further along the curve. But with the weird call structures and just the general weirdness of Munis, it’s hard to say when or if that’ll happen.

Chad Wildman (36:46):

Yeah, I would’ve said almost the exact same thing. I think going forward, the macro factors and the economic factors are going to play a huge part. Not that they don’t always, but I think in particular this year, obviously what the Fed does, there’s going to be a lot of watching that. The issuance question or a point I think is really interesting too because the weeks where there’s a big primary calendar can just get dead in the secondary. I mean, for obvious reasons, people are focused on their allocations and they don’t really need any sort of additional bonds. So secondary gets pretty quiet. That can be frustrating, but it is what it is.

Lynne Funk (37:32):

Can I ask you both, and maybe this is, we obviously don’t want to see a March, 2020 ever again, but are there ways that automation might help in times of severe fell offs?

Chad Wildman (37:46):

Yeah, I think so. I mean, the problem and part of the problem, we all know what the problem was in March, 2020, but the effect on the market was that you just had the rapid and I mean almost snap of a finger, the bid just vanished, right? So that’s not true across the board. It was a chance for folks to define their value and really solidify relationships by not just disappearing, but you can have all the automation in the world, but if you turn it off when the proverbial stuff hits the fan, then that doesn’t help anyone and it further exacerbates the problem, which is what we saw then. I mean the consecutive days of, what was it, 65 basis point cuts or something like that, and then what two weeks later pretty much did just complete opposite direction once the sort of economic stimulus program became more well-defined.

(38:53):

So if you had just done nothing, you would’ve been fine, but you’re talking about huge losses that institutions weren’t really ready to sustain, so they kind of sidelined. And with automation, you can react to these market moves. I mean, the trades that are happening are feeding most automated price formation mechanisms, and I’m sure Matt will have something more specific to say about this kind of thing, but you can be ready, it’s a matter, it’s a matter of buckling up and not running for the hills, I suppose. But that’s almost more of a policy thing than it is a technology thing, I think.

Matthew Smith (39:47):

Yeah, I think in types of turbulence, people tend to focus on what’s going to make them the most money for each unit of effort that they put into it. So I think a lot of people kind of shifted to the blocks in that space if they’re more of a manual desk and so odd, lots kind of lost some additional bids, but the algos that were there, the ones that didn’t turn off, they were the only bid on some of these RF qs. And so the more algos you have in the space or the more, it doesn’t even have to be algos, but just technology enabled desks that are there that don’t have to put a bunch of time into odd lots, you’re just going to see more bids in a time like that and it’s probably not going to be as severe.

(40:36):

The odd lot component of it is kind of interesting too. If you’re primarily trafficking in blocks and then something like March, 2020 happens, your P&L swings really wildly and you might not have an opportunity to get out of the one of 10 or 20 positions you have, but if you’re running an algo and you’re trading odd lots in that situation, you could be doing still thousands of trades, but they’re smaller and they’re a little less spookier. In a sense. It’s one thing to see negative 30 K on a single trade, but if those are pieced out into little pieces, people might have more of a stomach to ride it and commit to providing that liquidity.

(41:23):

ETFs are also interesting. In 2020, there were some big discounts that I think, I don’t want to speak ill of any data service out, but a lot of the evals that the ETFs use are end of day, and so the only intraday kind of price discovery you had to agree to a degree, there might’ve been other data services out there, but really everybody’s looking at these ice prints and then seeing or ice marks and then seeing where bonds are printing in your day. And so not having an idea of where that price is moving has put a lot of uncertainty into where their end of day P&L was going to be. But I think since ICE has a realtime eval service and just things like that in the market are just going to make people way less spooked when it comes to actually trading these and just having some idea of what the actual underlying value is.

Lynne Funk (42:22):

Excellent. Before we sign off, is there anything that we didn’t touch on that perhaps you want to leave our audience with?

Matthew Smith (42:33):

I’ll leave it with this. So I think it seems like big fancy words like AI and machine learning and this quant that have kind of dominated a lot of the chatter in the market, and I think it can be kind of daunting from an outside perspective, but when you really dig into it, it’s like at this point, it’s truly something that you can really easily integrate into your desk, whether that be adding some tools just to feel more effective, to do more things with your time during the day, all the way to coming up with an algo to maybe support what you’re doing. Something as simple as maybe buying matching odd thoughts to your position. As long as you start somewhere, you can kind of keep that ball rolling and make use or make your business much more effective and reduce the margin of your business as well, and without having to completely overhaul and rethink what you’re doing.

Chad Wildman (43:39):

I think that’s well said. And I think in closing, I would say something very similar. It’s just that automation and electronification are not just a market liking algo, right? The benefits that can come from just using any technology, it doesn’t have to be an algo, it can just be anything can be so far reaching, right? There are workflows everywhere that are super manual that people spend a ton of time on, and those same people can be so much more effective if there are things helping them not type a bunch letters and click a bunch of buttons in their day. And so identifying these is something that is very valuable and figuring out the right way to address them is the way that a lot of places are going to grow and remain relevant.

Lynne Funk (44:42):

Thank you so much to both of you for this conversation. I think it’s super important and it’s interesting and I don’t think it’s going to be stopping anytime soon, so I really appreciate you coming on, and thank you so much.

Chad Wildman (44:56):
Thank you for having us. Yeah, thanks for having us.

Lynne Funk (45:01):

Thank you for listening to this Bond Buyer podcast. I produced this episode with audio production by Adnan Khan. Rate us, review us and subscribe to our content at www.bond buyer.com/subscribe from The Bond Buyer, I’m Lynne Funk. Thanks for listening.

Articles You May Like

Labour promises to work with business
Top Wall Street analysts like these 3 stocks for long-term growth
Budget giveaways will still leave UK households facing hefty tax rises, think-tank warns
Barclays to return £10bn to shareholders
Israel’s economy shrank at 20% rate after outbreak of war