Innovation in execution: Credit Suisse

Risk Awards 2021: Swiss bank’s AES platform excelled in US Treasury market during March turmoil

Gregory Garofolo, head of rates execution services

The liquidity drought that afflicted the US Treasury market at the height of the Covid-19 market volatility was its worst ever. Off-the-run Treasuries traded like distressed debt and liquidity shrank even in some of the most liquid on-the-run tenors, as buy-side firms rushed for the exits.

In this stressful of environment, where even the most sophisticated electronic market-makers stopped providing prices, one method of execution was a port in the storm for its 25-plus users. Advanced Execution Services – Credit Suisse’s suite of algorithmic trading strategies, tools and analytics – came out shining.

Total volumes going through AES were up 400% in March compared with February, with the average notional trade size a hefty $400 million. Through the platform, Credit Suisse was able to provide its clients with 100% fill rates during the month and executed, at the best bid/offer or better, 78% of clients’ outright duration orders and 92% of multi-leg relative value trades.  

“We were up and stable the entire time through March and April. Clients had confidence that, despite the actual market conditions, the platform itself was stable from a technology perspective,” says Gregory Garofolo, head of rates execution services at the Swiss bank.

AES is a liquidity aggregator, connected to electronic trading venues, primary dealers and high-frequency traders, and integrated with buy-side order management systems. Initially deployed in equities, AES has expanded to cover options, futures, foreign exchange and, most recently, US Treasuries.

Around five years ago, some of Credit Suisse’s biggest clients began asking why they could not trade the bonds on the platform, given the rise of central limit order books and electronic trading in fixed income.

“Large clients were asking, ‘if I can trade Treasury futures with an equities algo, why can’t I trade 10-year cash bonds like that? The liquidity and the profile are all very similar. Why do I have to just go to one dealer every single time when I can just set a trade and work a strategy, if that’s what I want to do?’” Garofolo says.

So in 2017, AES expanded to Treasuries.

The platform’s main attraction for clients has been the ability to anonymise their flow. This way, they can execute passively in the market instead of being a net remover of liquidity via the voice market or the electronic request-for-quote protocol seen on platforms like Bloomberg and Tradeweb.

Clients had confidence that, despite the actual market conditions, the platform itself was stable from a technology perspective

Gregory Garofolo, Credit Suisse

“That’s really helped us become a top-tier counterparty for some of our most important accounts,” says Justin Mitrani, head of electronic macro sales for the Americas at Credit Suisse.

Anonymity enables clients to shield their strategies from market-makers. Garofolo notes that many large clients have said their number one transaction cost relates to information leakage, regardless of other fees or brokerage costs they may pay during trading.

“That’s been at the heart of AES as a franchise – protecting that information across asset classes,” he says.

To source liquidity in Treasuries, AES links up to well-established electronic venues in the market, such as BrokerTec and Nasdaq Fixed Income, as well as newer venues like Fenics UST and MarketAxess-owned LiquidityEdge. AES is also connected to Virtu’s fixed income engine, primary dealers and Credit Suisse’s own internal matching engine.

That breadth of aggregation helped AES outclass traditional forms of fixed income execution during March. The three arms of the bank’s rates business – voice, electronic and AES – had typically managed a third of client flow each. That changed in March when AES Rates grew to execute the notional majority of client related on-the-run trading activity.

Mitrani says AES’s ability to aggregate all forms of liquidity meant the platform continued providing liquidity in Treasuries throughout the crisis.

Justin Mitrani, head of electronic macro sales for the Americas, Credit Suisse

“There was a bit of a pullback, but I think it’s important to note that if we aggregate liquidity, then the singular effect of someone pulling back is less. The way that [buy-side] clients are leveraging the platform to be passive, and in a sense provide liquidity, helped improve the execution outcomes during that period,” he says.

The advantages of AES extend to normal times too. Algorithmic trades are often associated with smaller trade sizes, but AES typically does large clips, with its largest spread trade and outright duration trade standing at $12 billion and $4 billion of notional respectively. Generally, Treasury trades on AES range anywhere from $1 million to $1 billion.

Even with large-size trades, AES users that trade around predictable times – such as the 3pm close, Treasury auctions or Federal Open Market Committee meetings – are achieving better execution compared with trading via request-for-quote or directly with a dealer: in the past 18 months, clients have saved on average three ticks per each $1 million traded by executing via AES. This amounts to roughly $14 million in annual savings on average, according to Credit Suisse.

“Clients give us their benchmarks from the investment management team and we will then do some data analysis around that and then help them work out a strategy that is optimal for their execution profile,” Garofolo says.

That does not mean AES is always the best option. Garofolo says his team does not shy away from telling clients that they should trade via request-for-quote or pick up the phone to the bank’s principal trading desk for certain transactions.

Link-ups

In keeping with the trend for direct connectivity in fixed income markets, AES has also linked up to many of the most commonly used order management systems, including Aladdin, FlexTrade, Bloomberg and Charles River, as well as proprietary systems that some asset managers have built.

Garofolo says this integration of AES into the market structure and its development around clients’ best execution needs has contributed to the product’s success. 

“We aggregate liquidity, it’s anonymous, we have integration into client workflows and we have post-trade analytics. These four key components all really boil down to our clients’ best execution mandate and are the things that you need in order to be able to interact with the client base that we’re trying to service,” he says. AES users include asset managers, hedge funds, broker-dealers and insurers.

In a testament to the platform’s success, some clients have asked Credit Suisse to expand AES to other rates products, such as mortgage-backed securities, Treasury Inflation-Protected Securities and European government bonds.

“Anywhere where this model can work in fixed income, clients are starting to ask us to expand the use case. And we think that’s a good sign,” Garofolo says.