According to Financial Times News, bond markets have accelerated dramatically with US Treasuries now trading in milliseconds and corporate credit blocks being handled by algorithms. Despite this speed revolution, electronic trading volumes have plateaued at about 50% for corporate bonds and 60% for Treasuries in recent years. The percentage of US bonds that don’t trade on a given day has dropped from nearly a third in 2015 to just 10% by 2025. Meanwhile, block trades over $5 million conducted via voice brokers are bigger than ever. The time for price shocks to work through credit markets has shrunk from over 10 days in 2002 to just two days today, yet human-to-human trading persists alongside the algorithms.
Sponsored content — provided for informational and promotional purposes.
The Speed Paradox
Here’s the fascinating thing about all this automation: it actually created more demand for human trading. When you make everything lightning fast, you quickly discover what algorithms can’t handle. Credit markets remain incredibly idiosyncratic—every bond has its own story, its own quirks, its own risk profile. When markets get volatile or you’re dealing with complex, bespoke trades, traders want to hear a voice on the other end. They want someone who understands the context, who can explain why a price makes sense, who can build trust.
And the algorithms? They’ve basically become the perfect wingman for human traders. By handling all the small, routine trades quickly and efficiently, they’ve freed up capacity for dealers to focus on the big, complicated blocks. The balance sheets haven’t ballooned, but the market moves like they have. So now you’ve got this weird symbiosis where machines handle the high-volume grunt work while humans tackle the high-value complexity.
The Liquidity Illusion
Now, all this speed comes with some serious trade-offs. When algorithms dominate, they tend to compress bid/offer spreads and flatten volatility—Barclays analysis shows portfolio trading activity cuts price volatility by 5 to 7%. That sounds great until you realize it means bonds stop being priced on their individual merits and start moving with the herd.
Basically, you get this illusion of perfect liquidity. The market feels smooth and efficient, but mispricings become harder to spot. Fast liquidity isn’t necessarily better liquidity—it’s just faster. And we’re already seeing the consequences in equity markets, where some investors are quietly moving toward bilateral chats and off-venue trades. Sound familiar?
The Human Edge
So why can’t algorithms completely take over? Credit markets are fundamentally different from equities. Every corporate bond has unique covenants, maturity dates, credit stories, and legal structures. An algorithm can crunch numbers, but it can’t understand nuance or build relationships. When you’re moving $20 million of a company’s debt, you want assurance, not just execution.
Look at what happens during market stress—the 2020 COVID crash, the regional banking turmoil. That’s when voice trading really shines. Humans can make judgment calls, provide context, and offer liquidity when algorithms might retreat. The phone call isn’t nostalgia—it’s risk management.
Where This Is Heading
The most interesting development might be what’s happening in reverse. While everyone was watching bonds become more like equities, equities have started becoming more like bonds. That “bondification” of equities—with more bilateral trading and human interaction—suggests we’ve reached peak algorithm in some markets.
The future isn’t voice versus machine. It’s voice with machine. Execution can’t be purely equity-style when the underlying market isn’t equity-like. And in a bond market moving at machine speed, human judgment still provides the edge that algorithms can’t replicate. The phones aren’t going away anytime soon—they’re just getting smarter about when to use them.
