Marketaxess Holdings Inc (MKTX) 2020 Q2 法說會逐字稿

Ladies and gentlemen, thank you for standing by. (Operator Instructions) As a reminder, this conference call is being recorded on July 22, 2020.

I would now like to turn the call over to Dave Cresci, Investor Relations Manager at MarketAxess. Please go ahead, sir.

Good morning, and welcome to the MarketAxess Second Quarter 2020 Conference Call. For the call, Rick McVey, Chairman and Chief Executive Officer, will review the highlights for the quarter; Chris Concannon, President and COO, will discuss automation and our trading volume; and then Tony DeLise, Chief Financial Officer, will review the financial results.

Before I turn the call over to Rick, let me remind you that today's call may include forward-looking statements. These statements represent the company's beliefs regarding future events that, by their nature, are uncertain. The company's actual results and financial condition may differ materially from what is indicated in those forward-looking statements. For a discussion of some of the risks and factors that could affect the company's future results, please see the description of risk factors in our annual report on Form 10-K for the year ended December 31, 2019, and our quarterly report on Form 10-Q for the first quarter. I would also direct you to read the forward-looking statement disclaimer in our quarterly earnings release, which was issued earlier this morning and is now available on our website.

Now let me turn the call over to Rick.

Good morning, and thank you for joining us to review our second quarter 2020 results. Our second quarter results reflect accelerating market share gains, robust credit market volumes and a global shift among dealers and investors towards fixed income trading automation.

Revenues were up 47% to $185 million on the back of a 44% increase in fully electronic trading volume versus Q2 '19. Earnings per share of $2.20 was up 73% year-over-year. Operating margins reached a new record of 56.4%, up from 48.5% last year. Record new issue activity contributed to the increase in market volumes during the quarter. Estimated market share on MarketAxess bucked the normal trend during the quarter with accelerating share gains in spite of the heavy new issue calendar. U.S. high-grade estimated share jumped to a record 21.5% during the quarter, up from 18.7%, and U.S. high-yield estimated share grew to 14%, up from 10.4%.

Open Trading volume grew to a new record of $241 billion (sic) [$243.8 billion] in Q2, up 87%. With significantly higher volatility and price dispersion during the quarter, Open Trading estimated transaction cost savings reached a new record of $312 million exceeding company revenues for the second quarter in a row.

The value proposition to our clients has never been more clear as we deliver substantial transaction cost savings, essential new market liquidity and greater trading efficiency. I'm also thrilled to announce the appointment of Kourtney Gibson, President of Loop Capital to our Board of Directors. Kourtney's breadth of experience in both equity and fixed income markets and her investor-client relationships will bring valuable perspective to our company.

Slide 4 highlights market conditions. Market volumes measured by TRACE remain high due to the combination of higher credit spread volatility and record new issue volume. Since the COVID-19 market condition started on February 24, high-grade average daily TRACE volumes have increased 28% and high-yield is up 12% versus the levels earlier this year. New issue volumes are at record levels with high-grade issuance of $691 billion in Q2, up 150% year-over-year.

Credit spreads peaked in March and declined throughout the second quarter, following the Fed announcements on liquidity programs. Credit downgrades continued in the second quarter. Year-to-date, approximately $196 billion of high-grade debt has been downgraded to high yield. While credit spread volatility has fallen from extreme levels in March, our expectation is that it is likely spread volatility will remain higher than normal in the quarters ahead due to the increased economic uncertainty that is likely to be with us for some time to come.

Slide 5 provides an update on Open Trading. Open Trading delivered essential liquidity to market participants throughout the last 4 months. During the second quarter, Open Trading grew to 33% of our trading volume, up from 25% one year ago. On average, our Open Trading marketplace had 32,000 orders and over $18 billion in notional value available per day for trading. Dealer-initiated orders into Open Trading reached record volumes during the quarter and investors reached new records for providing liquidity on the system. We believe this robust all-to-all marketplace represents a substantial improvement in fixed income market structure, which is most evident during volatile times.

Unlike the challenging experience in credit trading in 2008, market turnover actually increased throughout this recent credit event, demonstrating that new innovations in fixed income markets are creating important new tools for risk transfer in secondary markets. There are encouraging signs of credit market turnover or trading velocity is increasing.

Open Trading is creating opportunities for new market participants, leading to record active client firms with over 1,500 institutional firms completing open trades during the quarter. These new participants are important contributors to the Open Trading liquidity pool in credit products.

Now let me turn the call over to Chris to provide an update on automation and our trading volumes.

Thank you, Rick. Slide 6 demonstrates the growing momentum of automation and credit trading. Automated trading volumes rose to over $32 billion in the second quarter, up from $19.3 billion in the second quarter of 2019. 83 firms used our Auto-Execution functionality in the second quarter, up from 55 the prior year. The average trade size conducted through Auto-X is also rising. In the second quarter, the average trade size using our Auto-X functionality grew to $222,000 versus $184,000 the year prior. Clients continue to increase the size of their orders as they gain comfort with the execution quality of our Auto-X solution. The use of dealer algorithms continues to grow on the platform with approximately 3.5 million algo responses in the second quarter, resulting in over 300,000 trades.

Despite recent market volatility, we continue to see strong growth in our automated trading solutions as both investors and dealer clients look to improve their trading efficiency. Inquiry volume and count reached new highs, demonstrating greater willingness to automate trading workflows.

Slide 7 provides a summary of our trading volume across product categories. Our U.S. high-grade volumes were up 56% year-over-year to $415 billion for the quarter, largely due to market share gains and the increase in market volumes. Estimated U.S. high-grade market share increased by 2.8 percentage points year-over-year to 21.5%, while estimated U.S. high-grade TRACE volumes were up 36% year-over-year.

Our other credit category volume was up 32% year-over-year, led again by significant growth in U.S. high-yield trading over our platform. U.S. high-yield volume was up 85% as estimated market share reached a record 14% and estimated TRACE market volumes rose 37%. Our emerging markets and Eurobond volume each grew double digits in the second quarter, with virtually all of those gains due to higher estimated market share.

We also had another solid quarter of trading in municipal bonds. In the second quarter, 315 unique client and dealer firms traded $3.2 billion in municipal bonds on the platform, up 93% from the prior year. Our rates business maintained its revenue and market share as compared to Q2 2019, and we are continuing to invest in new rates trading solutions. We are excited about the successful launch of our click-to-trade client solution during the second quarter, and our client onboarding for this unique solution is very encouraging.

Also, our treasury auto-hedging solution recently crossed $2.8 billion in volume since launch and hit a record of close to $1 billion in volume during the month of June. We also plan to launch our net hedging solution during the third quarter of 2020.

Our green bond trading initiative continues to support clients' ESG-related investment mandates. In the second quarter, over $6 billion worth of green bonds were traded on our platform, resulting in over 30,000 trees being planted in critical regions across the world. We have now planted over 60,000 trees since the beginning of the year.

Before providing color on July, please note that there are 8 trading days remaining in the month. July market volumes have declined from Q2 levels as they often do around the 4th of July holiday. At this point in the month, July trades volumes for high-grade and high-yield look similar to last July, and our high-grade estimated market share is running similar to Q2 levels and our high-yield estimated market share is running above Q2 levels.

Now let me turn the call over to Tony to provide an update on our financials.

Thank you, Chris. On Slide 8, we provide a summary of our quarterly earnings performance. Overall, revenue was a record $185 million, up 47% year-over-year. The 44% increase in credit trading volume and the inclusion of U.S. Treasury trading resulted in a 51% uplift in commissions.

Information services revenue was up 18% in the second quarter and includes onetime data sales of approximately $600,000. Operating income was up 71% year-over-year. The leverage in our model came through clearly in the second quarter with operating margins expanding to a record 56%, while we continue to invest in a variety of growth initiatives. The effective tax rate was 19.7% in the second quarter and reflects $5.7 million of excess tax benefits related to share-based compensation awards. We expect that the full year effective tax rate will be near the lower end of the guidance range of 20%.

Our diluted EPS was a record $2.20. The year-over-year increase in our diluted share count was largely due to the 146,000 shares issued as part of the LiquidityEdge acquisition.

On Slide 9, we have laid out our commission revenue, trading volumes and fees per million. Total variable transaction fees were up 61% year-over-year, driven by the increase in credit trading volume, higher U.S. high-grade fee capture and the inclusion of U.S. Treasury trading commissions. U.S. high-grade fee per million was $5 higher on a sequential basis and $20 higher year-over-year, mainly due to longer duration. Average years to maturity on bonds traded over the platform hit 9 years in the recent quarter compared to 7.8 years in the second quarter of 2019. Our other credit category fee per million increased by $6 on a sequential basis and $15 year-over-year, principally due to a shift among products favoring high-yield volume. Fee capture at the individual product level was very similar to the first quarter. Rates fee per million of $4.02 was slightly higher than Q1. Both U.S. Treasury's fee capture and agency's fee capture were similar to the first quarter levels. As a reminder, there could be some variability in our rates fee capture due to volume tiering under our treasury fee plans.

Total distribution fees were $700,000 lower than the first quarter level. One U.S. high-grade dealer and one U.S. high-yield dealer transitioned from distribution fee plan to variable fee plans during the second quarter.

Slide 10 provides you with the expense detail. On a year-over-year basis, expenses were up 25% for the quarter, with compensation and benefits accounting for close to 60% of the year-over-year change. The main contributors to the rise in compensation and benefits was an increase in head count of 81 personnel in support of our growth initiatives and an uplift in the variable bonus provision, which is tied to financial performance.

Clearing costs more than doubled year-over-year, reflecting the 87% increase in Open Trading volume and the inclusion of mass principal treasury trading volume. The increase in depreciation and amortization reflects the continuing investment in product development and the trading platform, along with the amortization of acquired intangibles. The biggest factor influencing the increase in technology and communication costs with higher software licensing fees, some of which are tied to trading activity and the uplift in professional and consulting costs is largely tied to higher recruiting fees.

We expect that full year 2020 expenses will end up near the high end of our guidance range of $314 million. Among other items, the most sensitive factor to our expense forecast is the level of credit market volumes, which impacts variable line items such as clearing cost and incentive pay. We are assuming that credit market volumes are likely to decline in the second half of the year.

On Slide 11, we provide balance sheet information. Cash and investments as of June 30 were $536 million and trailing 12 months free cash flow reached a record $280 million. During the second quarter, we paid a quarterly cash dividend of $23 million. We also repurchased 37,000 shares in total during the quarter, including 13,000 shares under our buyback program and 24,000 shares associated with the exercise or vesting of employee stock awards. We are approaching the go-live dates for our U.S. and U.K. clearing and settlement initiatives. We believe that our regulated businesses that handle match principal trading have sufficient liquidity and capital to cover any new deposit or reserve requirements. We also do not anticipate any change in our shareholder capital return programs.

Based on the second quarter results, our Board has approved a $0.60 regular quarterly dividend.

Now let me turn the call back to Rick.

Thank you, Tony. We are pleased with the record results we delivered during the second quarter. It is encouraging to see the acceleration of market share gains during these extreme conditions in credit markets. I want to thank our clients for supporting our vision for an open marketplace, our shareholders for believing in us and our employees for their dedication to allow the company to thrive throughout this health and economic crisis.

We wish you all the best and hope that we are on the right path to a full recovery in the coming quarters.

We would now be happy to open the line for your questions.

(Operator Instructions) And our first question comes from Rich Repetto with Piper Sandler.

My question is around automation and great results on the Open Trading and -- as well as the automated trading. And I guess my question is around market data. And how would -- I've heard from past panels and from calls, the importance of market data to continue to grow the automated trading. And so can you give us an update, either, Rick or Chris, on the market data offering and how important it is? And I'd also like to find out why Chris hasn't wrestled more this $312 million of savings from Open Trading back to MarketAxess?

Well, I was planning on talking about that savings, Rich, because it's unbelievable that a company of our size can deliver that much savings in a quarter where the savings actually dwarfs our revenue. So when clients see that savings, obviously, they are migrating to the platform. But to answer your question directly, you're absolutely right, Rich. Market data is a key component of automation. If you look at all the automation innovations, both here at MarketAxess and elsewhere in the industry, this is highly dependent on a very solid, accurate market data and price information. And it has to be real-time in order for the automation to truly succeed.

If you look at our rollout of our automated trading solutions and our pipeline of new initiatives around those tools, they are all linked to our wildly successful CP+ market data feed. So that is a critical part of our automation strategy, both in terms of how to price the instrument as a guide for clients, but also as risk parameters to protect them against large market moves while they have automation launched. So we have a number of risk controls using that unique market data, that is AI-supported market data.

So to answer your question, we're continuing to advance our automation solutions, things like Auto-Responder is critically dependent on our CP+ data feeds and helping clients to automate how they become a passive liquidity provider in the credit market. Hopefully, that answers your question.

Okay. Yes. Yes. What about the $312 million -- the wrestling of the $312 million...

We're happy to provide that back to our clients.

Our next question comes from Jeremy Campbell with Barclays.

Thanks for the update on the capability side of the algo. Maybe this is for both Rick and Chris here. Obviously, it seems like Open Trading and trade automation are kind of critical to the market share gains that you guys have had year-to-date. It looks like both the number of firms as well as the volumes went higher. So I guess, from a client perspective, given the unique working circumstances we're in due to the virus, can you maybe just give us some color on the outlook for cementing both Open Trading and automation into trading workflows from maybe both a new user and deepening wallet share perspective? And then also, just kind of wondering have you seen any slackening engagements in either of these systems this past month or so, so some large dealers and asset managers have had traders move back into the office?

Sure. I'm happy to start, and Chris can follow-up as well. But the beauty of our Open Trading solution is its simplicity. And we've really leveraged the broad base of institutional order flow that we have on the platform to create an open network where more people can participate in that order flow. And I do think that the share gains first half of this year are largely attributable to the benefits that we see from the open marketplace. First, the cost savings that we've mentioned and secondly, by expanding market participation. And virtually every quarter, we see growing activity from a growing base of alternative market makers who are committing new capital to the credit markets. We see growing activity from ETF market participants. It's interesting to note when you look at the second quarter ETF, high-yield share trading was up 52%, very close to our own overall trading growth. So we're growing hand in glove with the ETF industry as well, and that base of market participants has been enabled for a new way of trading through Open Trading.

We're also starting to see very positive signs from quantitative fund strategies, which historically have not been active in credit. And then finally, I'd point to growth that we're seeing in the hedge fund client segment which is relatively new as well. So when you look at the volume and share gains for the first half of the year, it is a very healthy combination of more volume with existing clients and new entrants into credit markets, which also then drives our optimism about market turnover.

And on your other point about work from home, there's no question that, that has created a tailwind for electronic trading just out of necessity. Moving away from trading floors into a home office, it's much easier to connect with global liquidity electronically than it is any other way. However, given the results that we're delivering to our clients, we personally don't see that going back to where it was as clients return to their office environments. And this quarter, we do see some gradual return among dealers to their trading floors, although it's still at relatively low levels, but asset managers continue to trade primarily from home. And our belief based on lots of conversations is that, that is likely to stay with us through the second half of this year.

And Rick, I'll just add on the automation front. We continue to see high levels of usage not only today, but it's consistent with the trend line growth that we were experiencing prior to the crisis and the work from home. And we have -- it's really a workflow efficiency demand that we're seeing from our large institutional clients. If you look at the size of the credit market and how large it's recently grown, the challenge is continuing to get more complicated to source liquidity and to execute both large and small-sized trades. So the workflow demands certainly are leading to a higher adoption rate of our automation solutions, both our Auto-X and our Auto-Responder. And I would say it's still early days on Auto-Responder, given some of the feature set that we have coming in the third and fourth quarter, things like high-yield, better enhanced OMS integration and even EM on our -- being launched on our Auto-Responder.

The other piece of automation that we're excited about, we plan to launch our first client algos in credit in the second half of this year, closer towards the end of the year. That's a unique offering where we're not only using our automated solutions, but we're wrapping orders together into a very sophisticated client algorithm for large institutional clients to use. So we're excited about the innovation that's coming in the second half of the year around automation.

Our next question comes from Kyle Voigt with KBW.

Just a question on the Fed, primary and secondary market corporate bond facilities, just when we're thinking about the direct and indirect impact from these. I guess, first question, have you seen any Fed buying directly on your platform? I'm not sure if that's happening to the dealers or BlackRock or somewhere else? And then on just the Fed facilities more broadly in an indirect way. Do you think that they're already helping to compress credit spreads and credit spread volatility? And if so, could that be a headwind to market turnover and the velocity of trade?

Sure. Kyle, I'll take a crack at that. But first of all, we would never comment on any individual client activity on our platform. So we don't have the liberty to answer your question about Fed activity electronically. And I would refer you to your corporate bond trading desk and others for further color on that. I think that the Fed has had an active role in the recovery of credit markets over the last 4 months, most importantly, the short-term liquidity programs back in March and early April that unlocked financing markets that really started the March toward more normal credit spreads that we observed throughout the second quarter.

What you see in the media is consistent with our understanding is that the Fed has been active in both ETFs and corporate bonds, but the levels are not outsized. And the reality is that with the changes in the liquidity and financing markets, the credit markets did a really good job recovering themselves and asset managers saw inflows through the quarter that they were putting to work, which caused credit spreads to compress significantly throughout the quarter. So I think the Fed is there when needed. It's clear that they have a vested interest in the capital markets and the cost of funding through this crisis. And they've had an impact on helping credit markets get back on their feet, which then allowed corporations to issue $700 billion in high-grade debt in one quarter. So they definitely have made a difference there. With respect to secondary trading, I don't think their activity has been outsized versus normal market participants.

Our next question comes from Ari Ghosh with Crédit Suisse.

Just circling back to your comments around Open Trading. The value proposition right now around liquidity cost saves, that's clearly playing out nicely and driving record order activity, share gains and new firms in the platform. So we're already seeing that. And I think you touched on this a little bit, but curious if you can just pass some of these recent (inaudible). How much of this is more environmental driven that could potentially dissipate on a more normal backdrop versus greater utilization from either underpenetrated clients or regions, be it (inaudible) participation, APAC, something like that, that could end up being a lot more sticky. So just trying to ask if these elevated levels should be viewed almost as a higher reset level driven by diverse and growing (inaudible) the platform?

Yes. No. It's a good point. There's no question that the cost savings from Open Trading benefited from the extreme volatility during the quarter and the much wider price dispersion the clients observed in levels coming back on order flow in any form. So I do think that the cost savings will ebb and flow based on volatility and price dispersion in each quarter. Having said that, we are confident that our Open Trading marketplace is unique. 95% of the order flow that I mentioned comes from institutional order flow and about 5% from dealer-initiated orders. And that's creating a tremendous opportunity for not only alternative market makers, but also for institutional investors to find more matching opportunities per day and drive their transaction cost down on a consistent basis. So we do believe that this is a long-term benefit to fixed income market structure. And it's great to see dealers embracing it more actively as well for their own liquidity with the levels of dealer-initiated order flow where they can end up transacting either with an end client or another dealer on those orders. And it's nice to see that the profitability of MarketAxess is growing, while at the same time, the dealers had very good quarters as well, which reflects how big these markets are. And we do think that longer term, you're going to continue to see market share gains and increased market participation that will deliver transaction cost savings booked back to our clients for many years to come.

And I would just add that Open Trading has a wonderful network effect that we're seeing play out in not only high-yield and high-grade, but obviously, in EM, where we're seeing participants that normally don't match with one another are able to match inside the all-to-all market. So the network effect of Open Trading will continue to grow and participants are benefiting greatly by that network effect.

We have a question from Ken Hill with Rosenblatt.

You mentioned earlier, I think on the Auto-X side, as far as the trades kind of moving into the larger scale kind of trades showing more confidence in that offering. Is there any sort of level you're seeing where clients maybe start to pull back or become more uncomfortable there? And what are the things you're looking at to kind of break down those barriers? Is it more information driven? Or is it more liquidity there?

It's really client comfort. We see a number of our largest users of automation, regularly increasing the size of their orders that they're putting on their lists and loading into automation. We're providing a great deal of analytics back to them of what is tradable and how liquid some of the products are. So it helps. We're guiding them on how they're finding quality execution at larger size orders across the Auto-X platform. So we are seeing our biggest users continue to grow their order sizes on the platform.

Obviously, I mentioned the launch of a client algo sometime in the second half of the year, that tool is also designed to take larger size orders and break them up into smaller size orders to allow clients to -- well, trade a large size to do it over the course of the day or week in much smaller trade sizes. So we're seeing a number of clients grow their order size in automation, but we're also trying to cater to those clients that want to trade with a lighter footprint in the market.

And I would just add to that, what we're seeing on Auto-Execution with investor clients is mirrored with the dealer community and the increase in size through their algos. So I do think it reflects confidence in trading automation in the system, and it makes both dealers and investors more efficient interacting with order flow on the platform. And in addition to the increase in trade size through Auto-X, it was also interesting through the credit event in Q2 that our overall trade size in U.S. credit increased during the quarter, and we had a record quarter for block trading volume. So you see really positive signs in trading automation and dealer investor confidence. You see more block trades coming in the system. Tony mentioned earlier, the increase in average maturity in the system, those are all very positive signs in terms of how our clients are now comfortable using the system.

We have a question from Chris Allen with Compass Point.

Wanted to ask a question on velocity. Rick, you mentioned earlier that you're seeing encouraging signs, losses increasing. Wondering if you can give some additional color there, how that may relate to the quant funds you mentioned. And then also what do you think Velocity can get to? It's still well below equity levels, not that I would have ever expected it to get there, but where do you think it could possibly approach over the next coming years?

Sure. Prior to bank capital, rent capital reform, it was common to see annual turnover in high-grade corporate bonds at 0.95% or 1% of debt outstanding. And we recently bottomed out at more like 0.65%, and we've recently come back up to about 0.7%. So we're still well below the levels you would see pre reg reform on capital requirements for the banks. So if I had to target what a reasonable goal would be with the extension of market participants into credit and the benefits of trading automation. And don't forget how important the ETF market is that, that was an active part of the solution over the last 4 months as both dealers and investors using ETF shares for critical liquidity needs, which then intersects with bond trading on MarketAxess. So I think it's market innovation, and it's the increase in market participation that's brought about through all-to-all trading that's starting to show a brighter picture on market turnover. So we bounced from the lows, and I do think getting back to something that we would have seen prior to 2008, it's more like 0.95% or 1% turnover per year would be a near-term goal.

And we have a question from Chris Shutler with William Blair.

Could you give us an update on your rough market share within high-grade by trade size bucket? And what kind of market share gains did you see in the quarter across each of those buckets?

Yes. So Chris, on the market share side, and I will give you the details, but the market share gains were -- we experienced market share gains across all trade sizes and all maturities. And when you look at -- if we just broke it up between block trading and nonblock trading, on the block side, our market share was up about 2.3 percentage points. And our overall market share was up 2.8 percentage points. And then the nonblock was also up about 2.8 percentage points. So regardless of trade size, market share was up. If you look at the heat map based on maturity, you would see market share up across all maturity sizes as well.

And we have a question from Jeremy Campbell with Barclays.

Just a couple of follow-ups on that. One on just the prior question. Again, you had some pretty impressive market share gains in the quarter, especially in the -- on the back of a lot of new issues that came out during the quarter. So kind of by our (inaudible) looks like your market share of new issued bond trading was actually higher than in prior years. I'm wondering, one, if that's correct. And then two, if that's true, whether you think this is a bit of a structural shift, so maybe it'd be less of a headwind in future years? Or is it maybe more transitory due to the kind of the work-from-home conditions?

Yes. I think I would agree with all parts of your comment that we did do better in trading of newly issued bonds during the quarter. And hard to know how much of that was the work-from-home environment versus a longer-term change in behavior. But we also did even better than the high level share gains when you look at season bonds and trading after the first 4 weeks. So a nice combination of both. But our progress on new issues has been encouraging, but we also think this is going to be a perfect place for our new live markets protocol. And that was set back because of the extreme levels of volatility and some of the key market makers that are supporting that initiative, having plenty to deal with through the crisis and the work-from-home environment. But we continue to hear very encouraging things about new issues and liquid bonds through our live markets protocol, and that's something that we will continue to work on developing during the second half of the year.

Got it. And then, Tony, can you just update us on the self-clearing initiative for the U.S. market? And maybe when that might now be expected to go live and likely to shut down the variable clearing system?

Yes. So on the clearing side, right now, we're expecting to transition to self-clearing in the U.S. and to a new clearing broker in the U.K. in the third quarter. So a little bit later than what we had anticipated, but probably understandable given the pandemic impact on us and on our vendor partners. What we're doing is, once we do go live in the third quarter, we do expect clearing cost as a percentage of Open Trading revenue for our credit business to decline. So remember, we have 2 -- really 2 components to clearing costs. We have it on our Open Trading credit business. We also have clearing costs on our match principal treasury business as well. What we've talked in the past about what kind of savings we can deliver once we do go live and typically clearing costs on the credit side as a percentage of Open Trading revenue have been 11% or 12%. And we do think we can drive that down to the single digits, and it's even more important today because if you look at Open Trading revenue in 2019, our credit revenue was about $100 million. If it goes down, for example, 3 percentage points, that's $3 million in savings. You look at the first half of this year, and we're on $170 million run rate for Open Trading revenue. So the cost savings are much more significant given what we're pushing through Open Trading today. So we've got our go-live date targeted right now. And we would expect some savings as we go into, say, the fourth quarter, we do expect -- that's part of our expense guidance. We do expect some favorability on the clearing cost line into the fourth quarter.

(Operator Instructions) Our next question comes from Sheriq Sumar with Goldman Sachs.

This is Sheriq filling in for Alex Blostein. Can you update us on the portfolio trading side? And like what's the traction that you're seeing over there in this quarter? And any color on how the volumes have been for the portfolio trading in 2Q?

Sure. As you recall, we launched our portfolio trading solution near the beginning of the year. We have seen growth in that product. I'll remind you that today, 3% of overall trades fund is estimated to be portfolio trade. So it's still a small subset of the overall market, but most of the portfolio trade solutions are really targeting a single dealer solutions that provide more of a trade processing solution and not really a full trading solution. Our trading solution, our portfolio trading solution does both single dealer but also multi dealer. So you can put your portfolio in competition for price among a number of dealers, and that's something that many of our clients find attractive. We are providing additional enhancements and improving how those portfolio trades seamlessly move through processing into the owned assets which is a critical piece of the portfolio trade. So since our launch, we're just over $1.3 billion in volume in portfolio trades. There was a slowdown, obviously, in portfolio trades during the most recent market volatility, given the difficulties in pricing portfolios in a fast-moving market.

Our next question comes from Rich Repetto with Piper Sandler.

Yes. I just wanted to get back to the sort of the impact of the issuance in the quarter. And Tony, you've done -- and I think this sort of mentioned earlier, but a calculation that sort of removes the issuance and sort of normalizes what the market share excluding the issuance. And could you go through that and what you calculate 2Q as? And then also, maybe one last thing, the variable fee capture. It's elevated because of the longer duration. Would you expect that to fall back somewhere between, say, this quarter and the first quarter? Is that a reasonable rate going forward? How would you view the variable fee capture?

Sure. Yes. So Rich, I can't say I did a whole -- a little science project here on what market share would look with and without the new issues. But this is what I can tell you is when you look at the portion of TRACE volume that related to newly issued bonds. And let's just take the second quarter, if you look at just the first 4 weeks of trading, it was about 1/4 of TRACE volume related to newly issued bonds. And if you look for any period prior to that, it may have been 10% or 11% or 12%. So clearly, the portion of the market related to newly issued bonds increased dramatically. And you look at our market share, our market share gains year-over-year were up -- for high-grade were up 2.8 percentage points. And again, without getting too scientific, if you pulled out the new issue piece, you would see our market share gains would have been either even healthier.

On one of the earlier questions that Rick responded to on, did our market share go up in newly issued bonds? It did, not accretively. But if you look at the first 4 weeks, our market share is typically around 5% or 6%. It was up a little bit, but not a lot. So you can see when you do the math around a big increase in the portion of the market that relates to newly issued bonds, the market share gains were even healthier.

And Rich, on the variable fee capture, there's so many factors, and this is particularly for high grades that we're talking about. The fee capture for high-yield, Eurobonds, emerging markets, very stable and not duration impacted at all. It's the U.S. high-grade plan where it is dependent on trade size. It is dependent on duration, which is years to maturity and yields. It's dependent on protocol. So there's lots of factors that can move it. The item we flagged was on years to maturity. You look over the past 10 years now, the years to maturity is ranged between 7 years and 10 years. We're not at the absolute high. We're in the middle of that range. It did extend out a little bit in the second quarter here. Hard to say what will happen going forward. We've given some color in the past that -- and this is just -- if you look at all else equal, a one year change in maturity could move the fee capture by something like $10 to $15 per million.

So if we move from, say, 9 to 10 years, all else equal, fee capture would move up, went from 9 to 8 years, fee capture would move down. Tough to just -- it's really, really hard to predict though.

We have a question from Kyle Voigt with KBW.

Maybe just one, just on that point on high-grade fee capture. Was it helped at all by lower yields as well? Or was that just a negligible impact in the quarter?

Yes. Kyle, it was much more on the years to maturity because even though you're right that yields were lower, spreads were higher. So -- where treasury yields were lower, spread to treasury was higher. So it was much more about the years to maturity. And if anything, we had a slightly negative offset because we did better in larger trade sizes. Not only our market share was up in larger trade sizes, the average trade size moved up as well. So if there was the -- more of the offset came from the tiering under our fee plan than it did from yields moving.

Right. Got it. And then just another question on the high-yield business. Just wondering if we could get an update on how sizable the ETF market makers were as a percentage of your high-yield volume in 2Q? And just wondering how that compares to where that's averaged historically?

Kyle, it's not -- I can't say it's -- again, sort of scientific project around this. We do look at the ETF participants. Typically, this is going back, say, over the past 5 years. The ETF participants for our high-yield business have ranged between 15% to 30% of our business. That 30%, actually, you may recall, it was the blowout in the energy sector going back almost 5 years ago now. In the second quarter, that percentage was below 20%. So what it does show is that our real money business and activity has increased significantly. ETF is critically important to our high-yield business, but you're seeing growth from other market participants and clients on the platform as well.

And I would just add to that it's becoming very difficult to target who is an ETF market participant anymore, given the banks have grown their ETF desks dramatically in the credit space and a number of new participants have come on to the platform. And obviously, we launched an effort late last year to target systematic fund strategies. And they do make use of ETF arbitrage strategies and other signal-based strategies. So we've seen a huge growth in the systematic fund side. Just in the second quarter, they traded over $42 billion in volume. So just a big move in a number of players that use both direct credit but also the ETFs as well.

We have a question from Patrick O'Shaughnessy with Raymond James.

Can you speak to the market dynamics in emerging markets right now? The industry-wide volumes that you guys track are not up to the extent that U.S. high-yield and U.S. high-grade are and your market share gains have -- while positive, have not been quite as strong as in some of those other products. So what are you seeing right now in EM?

Yes. I think parts of the EM market have been a risk-off environment for investors over the last couple of quarters and with some very difficult market conditions in places like Argentina that are important markets for us in EM. So there is greater caution among investors about adding exposure to emerging markets right now because of some of the economic difficulties and high debt levels in some of those markets. So it's not growing the same way to develop credit is right now. I still think it's likely to normalize and get back on track. But as a product area, market volumes have been more muted in EM than what we see in developed market corporate credit.

Okay. And then can you talk about the market opportunity that you see for your newly launched dealer-direct tool? And should we think of that tool is comparable to Trumid's Attributed Trading protocol? Or is it a different approach?

It's a great question. So we're pretty excited about the dealer feedback on the dealer-direct streaming tool. What's great about it is it allows dealers to customize streaming liquidity direct to a disclosed client. So it's a private way to stream liquidity to select clients. So in that regard, it's very similar to the Trumid's solution but it's not identical. The feedback has been positive. Obviously, it leverages APIs for the dealers. So it's a fairly streamlined setup for them and it allows the dealers to protect their data as well. The data that they show the clients is only for the client size. We don't aggregate that data into any of our data feeds. So it is truly a private market, a streamed private market for select clients.

And I'm showing no other questions in the queue. I'd like to turn it back to Mr. Rick McVey for any closing remarks.

Thank you for joining us this quarter. And please stay safe and stay healthy and enjoy the summer months, and we look forward to catching up with you next quarter.

Ladies and gentlemen, this concludes today's conference call. Thank you for participating. You may now disconnect. Everyone, have a great day.