Electronic trading sees slow but meaningful growth

When Toronto-Dominion Bank decided to merge its voice and automated trading businesses, it was an acknowledgement that the muni market will eventually become increasingly tech-enabled and automated.

“There are parts of the market that are going to be fully automated and algorithmic,” said Matt Schrager, managing director and co-head of TD Securities Automated Trading, noting the firm understands on a fundamental level the growing acceptance of electronification.

TD became one of the largest automated trading players in the muni market after acquiring quantitative fixed-income trading firm Headlands Tech Global Markets, where Schrager served as the firm’s co-CEO and head of research, in 2021.

“Just having access to the data and the technology those guys have is tremendous,” said Rick Fogliano, TD’s managing director and head of municipal products.

The voice and automated trading businesses had already been working together on several different projects, so merging the two makes a lot of sense, given the way the market’s heading toward automation, he said of the integration, set to be complete by the end of October.

The merger, though, does not eliminate TD’s voice trading business but rather enhances it.

“There will be part of the market that’s purely algorithmic … and then there’s still going to be an important part of the market handled by human traders, who will increasingly be more integrated with the algo and all the modeling,” Schrager said.

While TD’s case is unique, the muni market has seen growth in electronic trading over the past several years, as adoption continues to address the market’s growing need for greater efficiency, transparency, accessibility and liquidity.

Electronic trading has seen slow but meaningful growth.

In the first quarter of this year, 18.6% of new-issue volume in the market was traded electronically, up from 17.3% last year, according to data from Coalition Greenwich.

The growth occurs as total market volume also increased, so on an absolute basis the growth of e-trading is even higher, said Kevin McPartland, head of market structure and technology research at the firm.

The muni market is a large and complex market, especially given the number of CUSIPs, issuers, unique offerings and bid wanteds, said Susan Joyce, head of municipals trading and fixed-income market structure at AllianceBernstein.

If “you have a number of strategies that you’re trying to invest for your clients and you need something very specific, the ability to use electronic trading to go out and source exactly what you need, rather than trying to shoehorn something else into a portfolio, is fantastic,” she said.

Additionally, “the ability to quickly receive liquidity when you need it for cash raises and other reasons you might need to sell, and then to quickly turn around and reinvest that capital and not be exposed to basis risk is huge,” Joyce said.

All of this benefits the market and improves overall execution quality on transactions in both smaller and larger sizes, she said.

Some of the smaller sizes are odd lot trading, which accounts for 96% of all trades, Schrager said.

Automation has made the market more liquid, making it easier to trade odd lots.

“Liquidity in the odd-lot muni market has improved significantly,” said Josh Rosenblum, head of municipal trading strategies at Brownstone. “What once took hours now happens in seconds, as a growing pool of automated liquidity providers, like Brownstone, instantly price inquiries.”

Brownstone has been “deeply focused on building out technology and workflows to give our counterparties the ability to easily access more robust liquidity in a systematic way,” he said.

“You can’t have humans look at all those orders; you can’t hire new people. You need to increase the automation to keep up with where the market’s going,” Schrager said.

“Automated trading proves useful throughout various trading strategies and life cycles, particularly in high-touch, lower-impact trades where capital-committing-traders are able to focus on high-leverage situations where human intervention and judgment are critical,” said John Cahalane, managing director and head of Tradeweb Direct, Tradeweb’s retail-focused platform.
The platform “brings institutional-quality electronic trading to the wealth management and retail brokerage communities,” the firm said. It provides “seamless access” to a range of fixed income products, including munis, through “an intuitive and powerful electronic interface,” the firm said.

The demand for electronification is there, best seen by the adoption of automated trading from separately managed accounts, market participants said.

“The business model of SMAs and the business model of algorithmic trading from a dealer kind of grew up together,” Schrager said.

“You can’t have an SMA if there’s not liquidity in small sizes — SMAs trade small sizes all day long. But, you can’t have a really big algorithmic dealer if there aren’t trades to be done in those sizes,” he said.

As SMAs grow, the average trade size gets smaller because “the position sizes in SMAs are smaller, [and] then there’s a need for a bid on that. So you start to see the algos develop to provide that bid. And then there’s more need for selling, and it’s a positive feedback loop that grows the electronic trading network,” AllianceBernstein’s Joyce said.

The dealer community and the sell side were early adopters of electronification, but the buyside has become more comfortable with automated trading, market participants said.

AllianceBernstein, for example, uses electronic trading in several different ways: They sell to multiple venues, offer bonds out, bid on bids wanteds and buy offerings from a number of platforms, Joyce said.

The firm focuses on where it feels it has a “pricing edge” and where it can use its automated trading capabilities to engage with the market, she noted.

AllianceBernstein’s electronic trading has grown to 30% of its trading activity year-to-date, up from 20% a few years ago, according to Joyce.

The buyside’s comfort includes the increased adoption of order management systems and execution management systems, “enabling more seamless connectivity to dealers and platform liquidity,” Tradeweb’s Cahalane said.

For example, more than 10 dealers on Tradeweb’s platform use algorithmic trading strategies to quote a larger quantity of line items, he said.

This trend led to a record trade count year-to-date, a 27% increase year-over-year, Cahalane said.

“The rise of all-to-all trading is reshaping the landscape — buy-side firms are increasingly providing liquidity themselves, not just relying on dealers,” Brownstone’s Rosenblum said. “As roles blur between traditional buy and sell-side participants, success increasingly depends on being agnostic to protocols — whether bid-wanted, inventory negotiation, or other mechanisms.”

“More dealers are willing and able to respond automatically to buyside muni quote requests, while the buy side is looking to automation to keep up with increasing market volumes and flat or decreasing headcounts,” McPartland said. “That virtuous cycle keeps the market innovating and moving forward.”

Volatility remains an obstacle for electronic trading, Joyce said, “in periods of heightened volatility, electronic trading drops off somewhat as the movements in the market make it more difficult to use automated processes.”

Anything that causes big intraday movements, whether from macro events or technical factors in the muni market, like supply, can be a headwind, she said.

It requires a lot of investment in the types of risk controls to ensure “you’re not getting run over with how many securities there are in the municipal market and being able to stay on top of intraday movements,” Joyce said.

However, during the massive tariff-induced volatility of April, when a record $32.5 billion of municipal bonds traded on April 9, “electronic trading played a crucial role in enabling the market to support the massive volume spike, with the rising tide lifting all boats,” a report from Coalition Greenwich said.

Eighteen percent of volume in April was traded electronically, flat from April 2024 but down one percentage point from the prior month, the report said.

“There is no doubt that sales traders and financial advisors earned their money in April, helping institutional and retail investors, respectively, navigate the uncertain market,” the report noted. “But without the market’s newfound automation, many buyers and sellers would have been left on hold.”

Overall, “the volatility spikes [will not] scare people away from electronic trading. I think it makes it clear that you need it in order to get the best execution for clients and to get invested in the municipal market,” Joyce said.

Concerns remain that the move toward electronification may lead to the loss of the interpersonal nature of the muni market, market participants said.

Relationships remain important in every institutional market for different reasons, but in the muni market, given its local nature, “the number of CUSIPs and helping clients to understand exactly what the right price is and maybe what bond will meet their needs,” those relationships matter, McPartland said.

It’s understandable people don’t want to lose that human element, but at the same time, more efficiency and automation are ultimately better for everybody’s bottom line, he said.

However, even as the market shifts toward automated trading, the personal touch of munis will never be lost.

For Brownstone, its “formula is simple: relationships plus technology,” Rosenblum said.

The firm leverages “machine learning and design automated workflows to improve liquidity and streamline execution,” he said. Using “AI effectively is now a fundamental expectation of everyone at Brownstone.”

However, “even as technology evolves from competitive advantage to industry standard, relationships remain the foundation of our business and ultimately our core differentiator,” Rosenblum said.

TD, for instance, pictures a future where voice and automated trading work hand-in-hand to complement each other, Schrager said.

“When a voice trader responds to a request for quote, traditionally they do look the bond up on Bloomberg, do some analysis, but mostly it’s in their head. Now we want our traders to be able to do that, but then also have modeling and data analytics that come from our algo modeling and have them right in front of their face at the time that they’re going to make a trade,” he said.

“We believe really strongly in what we call the hybrid model,” Schrager said. “There’s going to be some part of the market that is going to be fully algorithmic, like, basically no humans involved … But the voice trading part of the market is never going to go away; it’s always going to be important.”