Unlocking Corporate Bond Liquidity With AI

AI can help improve liquidity in the bond market.


Shutterstock

Artificial intelligence could be on the verge of unlocking previously unattainable levels of liquidity in the corporate bond market.

The U.S. corporate market has more than $9.8 trillion in bonds outstanding, but less than $38 billion trades in the secondary market each day. That discrepancy is growing as a surge in new issuance far outpaces growth in secondary trading.

The lack of a central limit order book or real-time trade reporting in the OTC market makes price discovery difficult and sometimes impossible. On average, 80% of corporate bond-trading activity comes from just 20% of the outstanding debt issues. Many bonds trade only rarely, or not at all. In fact, most investment professionals have a large percentage of their portfolios in “trapped bonds,” meaning they have no ability to find natural buyers or sellers for the securities at market-clearing levels.

The combination of illiquidity and poor price discovery often results in significant market impact and poor pricing for firms looking to buy or sell bonds. These conditions also consume time and effort: Portfolio managers spend significant time creating trades they would like to do, only to find insufficient liquidity to get the trades done at their target prices.

Historically, “backstop” liquidity provided by broker-dealers mitigated the worst of these effects for market participants. However, a decade ago, regulators attempting to shore up the banking system after [WP1] the global financial crisis [WP2] hiked banking industry capital reserve requirements. That move prompted banks to shed the corporate bond inventories they used to make markets and provide liquidity on a marketwide basis.

Liquidity levels never fully recovered, and market participants are still feeling the effects today. Although the growth of electronic trading to about 40% of corporate bond market trading volume has helped create liquidity for smaller trades, market participants struggle to complete bigger trades. In many cases, executing a large corporate bond trade means seeking out liquidity from multiple counterparties, risking information leakage.

Creating liquidity with next-gen tech

These challenges have created a huge incentive for investors to seek out new sources of liquidity, and for brokers and vendors to invest in innovative technologies that can make the market more efficient. These investments are starting to pay off.

New technology platforms are able to monitor and report corporate bond liquidity conditions in real time. By using AI to process input from both the buy side and the sell side, they can tell a portfolio manager how likely it is that they will be able to complete a given trade and send set alerts to notify users when there is potential liquidity to trade. This gives market participants the ability to assess liquidity without tipping their hand by asking multiple dealers.

The platforms also have the ability to tell market participants which dealer is most likely to get the trade done, execute the trade fastest and complete the transaction with the least market impact or most price improvement. They do so by analyzing a dealers past performance on similar trades and assessing the strength of the dealer’s network of investors. By looking at historical trading patterns across a dealer’s client network, these AI applications can project the level of interest in a specific trading opportunity. Routinely, these models are able to identify one or more clients in a dealer network with significant interest in a potential trade.

Bearing the benefit

These new technologies allow asset managers to initiate a trade with the knowledge that there is sufficient liquidity at their expected price and ask the best-equipped dealer to work their order—based upon that dealer’s current and past performance. In turn, dealers can leverage the same AI technology to identify interest and liquidity within their own networks, and efficiently connect the most probable clients to optimize a successful transaction.

A corporate bond market running on such advanced systems will be much more transparent. New trading protocols allow each of a dealer’s clients to express their interest in a specific trade with a firm bid/offer for a desired size. Each bid/offer can be seen by the initiator of the trade, the hosting dealer and any of the other clients of that dealer who have submitted a bid/offer of their own. In this way, every participant benefits from price discovery. Each participant—including the originator—can improve their price based upon the interest they are seeing to arrive at the best market-clearing price.

These platforms are already starting to help market participants find liquidity and complete trades in a challenging market environment. Over the longer term, AI-powered trading systems have the potential to transform corporate bond trading into a market with almost unimaginable levels of transparency and liquidity.