- Hey, thanks for joining us. I'm Larry Tabb. I am head of Market Structure Research at Bloomberg Intelligence, and we've got a really interesting discussion going on today that really the the major topic is a next generation of tech in practice. And technology, of course, has really been dramatically changed the way investors and dealers trade, as well as how exchanges and folks access the marketplace and match buyers and sellers together. We've seen trading protocols, connectivity and technology completely really rewrite equity and derivatives markets and how they operate. And to that point, it's almost impossible for any equity or options trade to be traded, to be really matched by hand anymore. Everything needs to go through machines. That said, the fixed income world's been pretty different while sovereign debt markets have become somewhat automated. Once you get outside benchmark government securities. more than 60% of orders are still matched by phone, but that's changing. And to discuss how technology is changing the face of corporate bond markets, we have a great panel of experts. With me to discuss this is my friend and colleague, on the FIMSAC, the fixed income market structure committee of the SEC, Horace Carter. And Horace is the EVP and head of Fixed Income Capital Markets at Raymond James. Another long time friend is Jim Toffey. Jim was the ex-founder of Tradeweb, and he's now the president of LTX. And last of course, and certainly not least, Nadine Chakar, EVP, head of State Street Global Markets at State Street. So let's get this discussion started with Horace. You're head of fixed income for Raymond James and help with the daily execution of thousands of corporate bond trades across the vast array of financial advisors. What do you see as some of the challenges in today's market, Horace?

- Thank you, Larry. To do with what I think of as wasted energy. And a lot of it has to do with visibility into the market for especially our advisors. Visibility in terms of valuation. So by way of example, we have a client that wants to liquidate bonds. What they will do is they will put all of their bonds out for the bid. As bonds go into the market, they've received bids back and then the advisor and the client selects on the sell. So we have a conversion rate... Of inquiry that is actually traded. So we're spending a lot of time and a lot of resources on price discovery. That actually increases. In March, for example, the number of inquiries that we had approximately tripled and the conversion rate dropped to about 20%. So that's one of the big challenges. Other than the obvious challenge of having difficulty finding attractive yields for our investors right now.

- So basically, there's a lot of manual work, a lot of data that needs to be aggregated from a lot of different places and really just a lot of legwork and running around trying to get, find the best prices for your clients, I guess. That kind of sum it up?

- It does, but it's improving rapidly, because say 10 years ago, it was far more manual. And in addition, on the offered side, we're getting tools that help us sort of sift through offerings to find a real value for our client. There are thousands of bonds offered at any given moment, and we have to use these tools in order to find the bonds that our clients want. But also we need to be able to determine whether or not they're offered at an appropriate price. And so the tools that we have are good and they're getting better.

- Let's transition this to Nadine. Nadine, you're at State Street. You're one of the largest global custodians. You deal with mostly side of the business. How has it, and as well as you kind of represent Charles River as well. A lot that interacts with... You interact with Charles River, the order management platform that State Street bought about two, three years ago. How is the institutional side of this market? Do you have a good view on that? Is that similar to Horace with a lot of sneakernet and running around on the fixed income side?

- It is, Larry. I think the buy-side is struggling, if you will, to streamline its automation in trying to leverage more algos in this space. So what we're seeing is still a lot of buy-side firms still operating on legacy infrastructure. And if you think through what Horace said, the ability to aggregate data, map it, reconciling it, then do your modeling and trade, requires robust data strategies, and computing power, that not all firms have. And then if you add to that, the challenge of the lack of interoperability, the fragmentation of liquidity, non-centralized solutions, we're still not seeing... A lot of the ETS is trying to aggregate liquidity and help clients. So for the buy-side, we do need to figure out an easier... We need to make it easier for them to jump in with both ends. It's not lost on us that we're seeing a proliferation of electronic trading venues but yet still a very low percentage of bond trading done electronically. So we continue to see those same challenges.

- Well, let's bring Jim in here. Jim, you're kind of the grandfather of electronic fixed income trading. You founded Tradeweb, what? Almost 25 years ago. Shouldn't have this problem been fixed? Where are we and how are we moving forward? And I guess you're now with LTX. How is that looking to change kind of this process?

- Well, thanks Larry, for reminding me how old I am. The space has changed a lot in the last 20 years. I mean, RFQ, which powers both Tradeweb and market access was an important protocol, and really kind of took like the bond markets forward. But really, if you look at RFQ, it's really a protocol designed to electronify the phone call from before. And even though RFQ is very widely used, when you look at aggregate numbers like in the corporate bond market, over 70% of the market is still done, not electronically but over the phone or over IM or something like that. So there's still a long ways to go. And I think what you're gonna see here over the next several years is an advancement on two levels. One using AI and machine learning to make people smarter around the data of, for example, when is the right time to trade? Where's the most liquidity? Who are the natural buyers and sellers that should be invited to that trade? So I think there's gonna be a lot of emphasis on using data to help both the buy-side and the sell-side trade smarter. And then on the trading protocol side, I think you're about to see, and this is one of the things we're working at, LTX at Broadridge, which is the next level, the next generation or protocols that will go beyond RFQ, that will aggregate liquidity, that will create better price improvement in kind of the old RFQ model. So I think between these two trends in convergence, I think we're about to see a real step function change. And honestly, over the next three to four years, it wouldn't surprise me if the amount of trading done electronically across the fixed income markets finally gets up to 50% or higher from kind of the 20 to 30% that we are today.

- So you're talking... You brought up the idea of protocol and how folks actually interact with these platforms and how dealers and clients interact. Horace, what types of protocols, the traditional RFQ protocol is kind of clients put in a request to trade and the dealers kind of respond. Horace, are you seeing those protocols change? Are you guys embracing new protocols? How do you kind of interact with these protocols and how would you guys like to see them altered a bit?

- Well, we do, RFQ is obviously still the dominant protocol and we do a good bit of that, but we are looking at different protocols. Not to butter up Jim, but the LTX protocol is... It has the unique feature of being both dealer centric while simultaneously increasing transparency and competition. So, that's a new one, but in particular, what we've observed is the growth of the all to all or open trading. Obviously, as a dealer, we would be interested in that phenomenon. It's been steadily growing over time. And I would point out that in March, open trading grew very rapidly. And I think that the reason is, that people were trading execution quality for execution speed, because on an open platform, you get your RFQ in front of a very large number of potential counterparties very quickly. So you tend to get prices back more quickly as compared to a negotiated price where the execution might be better, but it takes a longer period of time. So that's a phenomenon, but I think we're going to continue to see increase.

- So, RFQ, as Horace said, kind of puts the dealers in competition. You mentioned something about LTX not putting the dealers in the competition. How does that kind of work?

- Well, as I said, it's dealer-centric, but it actually increases competition. Increases competition and transparency. So you need-

- Okay. Some of these challenges in terms of migrating the platforms forward have a lot to do with liquidity, and there's kind of a chicken and egg issue with liquidity, especially in fixed income. Things trade fairly frequently. Once they're issued they kind of go into the vault and kind of stay there. I don't turn over my portfolio because I don't necessarily see great prices or great, great prices when I trade. So I keep them in the vault, but because they're in the vault, there's not as much liquidity and competition to kind of get them in the circulation. How do you see this changing, Jim? Do you see this chicken and egg problem being solved? Or how do we solve this chicken and egg problem to drive more liquidity, create greater transparency?

- Well, Larry, you're bringing up a great point. Number one is, as we've discussed, the overall penetration of electronic trading in the markets is still relatively low compared to other asset classes. And then when you look at the record-breaking issuance of corporate debt this year, and then look at secondary trading, the gap is actually widening between new issuance and secondary trading. So in some regards, you could actually say the markets, the secondary trading hasn't kept up with the new issuance. I do think that, for example, back to the theme of data and using technology, one reason people think about buying and selling is the probability of success. Will there be enough liquidity for the bonds I'm trying to sell at the price I'm trying to get it done at, for the amount? And there are ways to quantify that probability going forward. In talking to the buy-side over the last few years, we found that over 50% of the time, what the buy-side is actually trying to get done doesn't even turn into an order when they talk to their dealers. In other words, over half the time when they talk to the dealers and explain what they're trying to do with our market access or directly with dealers, there's just not enough interest, demand and liquidity. So one of the opportunities is how do we assess that liquidity before you trade? And we're using innovative liquidity scores with a quantitative analysis to help the buy-side assess that probability before they trade. Second thing, and in the OTC markets, there's always a balance between transparency and information leakage. And how do you encourage more liquidity, more transparency, but don't give away information leakage to the rest of the market that wasn't really interested in providing, and they got information for free. So you wanna look for innovative solutions that gives price transparency to buyers. I mean, often, one of the other things we've heard in electronic trading is RFQ is fine, but it doesn't tell me that I could, if I just bid a little bit more I could have won all the bonds. So I'm really bidding blind, I see nothing in return. And so there's not the ability to have another wave of price improvement because I can't see those prices and react accordingly like I can in equities, like I can in foreign exchange, where it's a very dynamic price discovery. So I think that's also an opportunity to provide that kind of price discovery to the market, which back to your original question will lead to more certainty, more certainty leads to more liquidity. More liquidity leads to better price discovery. And then you get this virtual circle breaking that chicken and the egg that you were describing.

- So, to reinforce this, so you're providing some... You're trying to create some sort of score that says that my bonds will be very liquid, moderately liquid, less liquid, so that I can have more confidence when I go to the market that the things I wanna sell can actually be marketable?

- Yeah. And again, in the traditional world of Charles River and OMSs and EMSs, and then in the marketplaces, there's always been the pre-trade happens on those screens. And then the execution happens on marketplace just like ours. What I think is you'll see tighter integration over time using these liquidity scores as a bridge. So for example, we're talking to Charles River about putting our liquidity scores embedded in the workflow to help all those customers understand on a pre-trade basis, what's the probability of success. And as you know, Larry, when you study data, the results you get is only as good as the data that goes into those equations. And the best data to have, really, if you think about it is all the data of the corporate bond market, you think trades and what's traded in the past, what people would like to trade now. Sharing anonymously what they're thinking about trading so that they can get better clarity on what's the probability that if I go to sell this illiquid bond now, there's actually buyers on the other side. And then obviously the final piece is the feedback loop from the marketplaces feeding back in. So I think you'll see an integrated title work cycle between pre-trade and trade as platforms like ours work tighter with OMS and EMS platforms going forward.

- Well, let's bring Nadine into the stadium. You guys are highly integrated or the CRD, you're highly tied to the buy-side. How do these new protocols and how are these new technologies kind of work through, I guess, CRD and to the buy-side to kind of get them more comfortable and get them better liquidity.

- Larry, I'm gonna have to sort of try to carry on on the themes that Jim raised. And if you think through our entire strategy of Alpha, which is really powered by Charles River was really to connect front to back the pre-trade, trade and post-trade environment for our clients. And starting with the EMS, for example, Charles River spent an incredible amount of time, I think over 10 years, rebuilding, well, building from scratch actually, the fixed income component of that. And we've spent a fair amount of time building the fixed income inventory, trade automation and most importantly, strengthened the open architecture by enhancing third party data providers' venues and application vendors. So really try to create one ecosystem where we can bring them all together. And while we've been connected to venues like Trumid and Tradeweb and market access and 30 others for years and years right now, I think what you're seeing is a shift and signaled by our partnership with LTX. Which we're really excited about that, because we we do believe that that's gonna bring more added value to our buy-side clients by providing them with better access to fixed income dealers, which is the theme that Horace had raised earlier. A more efficient trade workflow, and some of the AI tools. That's really important to be able to mine a lot of that data to provide better liquidity, better transparency. And throughout the whole process the pre-trade, trade and post-trade environment, which is critical. And then we've added, if you will, an extra layer to that to which we call a clear connect, which was really to add more simplicity to describe the complex process. And that's just another attempt for us to try to help clients with that post-trade process. But most importantly, giving them access to better data, better liquidity, and allow them to connect with any venue and any provider. But the LTX component of what we're doing sort of elevate the work that we have been doing on our own for these years. And it just gives the buy-side a much better experience, which hopefully would lead to a lot more usage of electronic trading. I mean, we're seeing an increased usage in algo trading, but it's hard to do when you don't have the infrastructure, the data to be able to execute in that environment.

- That's interesting. So you're seeing increased use of electronic tools and algorithms and fixed income data. Let's bring this to Jim. Do you at any point in time kind of see the fixed income markets becoming more like the equity markets and exchange traded derivatives markets becoming more electronic and more algorithmic? What do we need to do to get there? And are we gonna get there? Or will it just be different?

- I mean, think it will be structurally different, because the fixed income markets are OTC based. They're based on dealers. I think the key is actually instead of putting dealers in comm, is more about helping the dealers do more with their own network. Raymond James, for example, has a massive network of 800 or 900 institutional clients. Every fixed income dealer has a massive network. The key, I think, is using technology to harness that dealer with the liquidity of their entire customer base to aggregate liquidity. So what does aggregate liquidity mean? In an RFQ, if you think about it today, and Horace mentioned all to all trading, it's great that it goes out to a wide open marketplace of users, but at the end of the day, you're getting liquidity just from one buyer. So if you're trying to sell a block of bonds, you're still getting liquidity just from one, even though you reached an audience of 25, 50, 75 buy-side customers. In equities as you know, and other markets, they go out to the same universe, but they aggregate liquidity. They take five from one person, 10 from another and they use BestX to create not only aggregating liquidity across multiple customers, but also price discovery. So if I realize I need to pay a little bit more to buy the bonds I want, I can. And I think those dynamics have been missing in fixed income. And I think as those dynamics start to appear through our protocol and others, the fixed income markets will go to another level of liquidity, and again, get close to the kind of what you experience in more liquid asset classes.

- Interesting. Horace, how do you see electronic execution kind of carrying out through Raymond James and through other dealers?

- Well, I do agree with Jim. I think that the electronification is going to continue. I think that algorithmic trading is going to become more and more a part of the business. I think that in 20 years, you will see fewer human bond traders and more computer bond traders. And I would liken it to... Imagine the manufacturing industry going forward without robotics. There's a certain amount of inevitability to it, but I don't think it's going to get to the level or to mirror the structure of the equity market, because as Jim pointed out, it's an OTC market, whereas equities is more order matching. You get a match order book. That is not the way fixed income securities play. So the long and short of it is that they will... Ways for market makers in the fixed income business. And there will always be a role for these pools of capital. We have to work much smarter now using the type of AI that Jim and Nadine have described in order to service our clients better, because we don't have the massive pools of capital that we used to. So the dealer inventories are less than 10% of what they were in April of 2013, for investment trade, corporate bonds. That means that liquidity is done through... Liquidity is obtained through distribution rather than balance sheet.

- Well, we're getting towards the end of our discussion. So let's end up with Nadine. Since the order management platform, and certainly CRD is one of the biggest connectivity providers to the buy-side, how do you see the buy-side fitting in? Certainly, with retail and financial consultants and advisors, it's a little easier to kinda aggregate small pieces, but if you're trying to buy big blocks, how do you see this fitting in to the buy-side?

- I think, Larry, the buy-side doesn't really have a choice but to jump on the bandwagon of automation and aggregation. And it's mainly because of the pressure on the business models. We're seeing margin compressing, obviously MiFID is pushing towards best execution. So there's definitely a big push on trying to figure out more efficiency, the ability of trying to find efficiency and automation and scalability, which are gonna be key to that process. So we will continue in our partnerships with LTX, and Jim and others to continue to find ways to help our clients with sourcing that liquidity and most importantly trying to help them get access to the right data and the right protocols for them to execute efficiently. That's our entire business model here to really help the buy-side move forward on that front.

- Well, thanks, Nadine. Well, it's certainly an interesting time with that. I think our time has come to a close. I wanna thank Nadine Chakar, EVP, head of State Street global markets. Jim Toffey, the president of LTX, and of course, Horace Carter, the EVP and head of fixed income capital markets at Raymond James. And I wanna thank everybody for joining us and for a very interesting discussion on really, the future of fixed income technology and trading. So thanks very much. And I thought it was a very good discussion. Thank you so much to our panelists and all our speakers today. Thanks for our sponsor, Broadridge. Be sure to check out their survey under the handouts tab in the Q&A box. And special thanks for all you guys for tuning in. We look forward to seeing you at the next one. And I'm Larry Tabb, head of Market Structure Research for Bloomberg, and we'll see you soon.