Data and technology are pushing the boundaries of primary and secondary market infrastructure. Danielle Myles examines how banks have found ways to remain valuable to their clients.

Corporate bonds embedded

Corporate bond markets are undergoing a period of experimentation. The most obvious evidence is the explosion in the number of electronic trading (e-trading) platforms. There are estimated to be up to 90 worldwide, which offer the buy side and sell side an alternative to picking up the phone when they need to find a counterparty to their trade.

While equities, foreign exchange and even government bonds are predominantly traded on screen, the corporate bond market is one of the last bastions of voice trading. There are good reasons for this, stemming mainly from corporate bonds’ lack of standardisation and therefore illiquidity. Infinite combinations of maturities, coupons and ratings mean compatible buyers and sellers would struggle to find each other on big, stock-like exchanges.

Do not open?

The limited number of potential counterparties also makes investors reluctant to advertise their purchase or sale interest (their axe) on open exchanges, because such information leakage could cause prices to move against them.

But while the corporate bond market may be slow to relinquish voice trading, it is far from old-fashioned. The growing number of e-trading venues is just one (and certainly not the best) example of how participants are testing ways to create a more liquid, more efficient and less capital-intensive industry.

Changes in market infrastructure are widely attributed to the batch of post-crisis reforms, including the US’s Volcker Rule and Basel III capital charges, which limit banks’ ability to act as market makers – historically, the major sources of bond liquidity. But the revolution occurring in corporate bonds is not caused solely by reaction to regulatory tightening but also because participants are embracing technology, which means they are not limiting themselves to the secondary market.

Electrifying the primary market

By the end of 2016, European investment-grade corporate bond issues could be sold electronically through an online portal called Investor Access. This will streamline the way bookbuilding is done today – via a series of emails, phone calls and online messaging – into a single point of access for communications between investors and the bank syndicate.

Bond buyers can still contact a bank’s sales team for more information about a deal, but Investor Access will electronify the rest of the process (that is, convert it to a digital form). This includes the deal’s announcement; communication of terms and conditions (T&Cs); the placing, upsizing and downsizing of orders; pricing guidance; and allocation.

Investor Access will be operated by capital markets software firm Ipreo, which hosts an electronic interbank platform that helps reconcile syndicates’ bookbuilding processes. Work began in 2014, when a handful of banks asked Ipreo about the possibility of launching a similar system at the investor-bank level.

Ready for take-off

Fast-forward to today, and 11 banks are sponsoring the initiative, including HSBC, Société Générale Corporate & Investment Banking (SG CIB), BNP Paribas and MUFG, and some 50 buy-side firms have agreed to participate.   

The pilot phase has started. Permission-entitled T&Cs are being communicated via the portal, and orders and allocation processes are in the final stages of testing. Herb Werth, Ipreo’s global head of buy-side product strategy and marketing, expects a live deal to be executed by the end of 2016.

More investors and banks are expected to sign up, and for good reason, as Investor Access will cut the time spent on administration. For buy-side traders, knowing where they stand on a deal at any point in time will help with their juggling act. “There’s a lot of inefficiency today,” says Mr Werth. “Not only does all information need to funnel through the trader who needs to make sense out of it, but at any given time they may be dealing with multiple banks, a number of different active deals, and each of their portfolio managers.

“Things are pretty busy for traders, and they need to be careful not to miss anything. The feedback we’ve received so far is that this solution helps to consolidate the information flow and reduce the overall risk in the process for them.” 

It is a similar story for banks. Eric Cherpion, SG CIB’s global head of debt capital markets syndicate, says: “For the salesforce, it completely spins off the low added value of the process by automating the order collecting process. It means they can spend more time explaining the credit, sector and relative value around the issue, and so on. For the syndicate desk, it minimises operational risk and the risk of duplication of orders by electronifying delivery. It’s all about enhancing efficiency.”

Once rolled out in Europe, Ipreo will see if there is an appetite for Investor Access in other regions starting with Asia, as well as other fixed-income securities.

E-trading 2.0

For secondary market participants, Investor Access may be a flashback to the 1990s when the first e-trading platforms came about. They have had a bumpy growth trajectory, but volumes started to surge in 2012. According to a Bank for International Settlements (BIS) report, between 2012 and 2015 the proportion of trading done electronically grew from about 15% to 40%.

The market is dominated by three players: MarketAxess, Bloomberg and Tradeweb. While they are headquartered in the US, research by Greenwich Associates shows the percentage of European investment-grade bonds traded electronically is double that of the US.

Electronic trading in corporate bonds

The first generation of e-trading platforms is, by and large, an electronification of the traditional voice model. They are dealer-to-client and rely on the request-for-quote (RFQ) protocol. With RFQ, a client asks multiple banks to provide a quote for the bonds they want to buy or sell, and they execute with the one who makes the best price.

Performing this process electronically improves efficiencies but does not, in itself, improve liquidity. However, over the past three years an array of new protocols has emerged, signalling the arrival of the second generation of e-trading.

All-to-all platforms

One innovation is the growth of all-to-all platforms that do not distinguish between counterparty types. Trades can be between two dealers, two investors or a dealer and an investor. The advantages to the buy side are twofold: they can trade with a dealer who they do not have a relationship with, or directly with another investor. At a time when dealer liquidity is hard to come by, this is a valuable proposition. All-to-all platforms unlock different sources of liquidity by connecting counterparties that may not have found each other on dealer-to-client platforms.

There is disagreement over the extent to which buy-side can be price makers. Fiduciary obligations prohibit pension funds and insurers from taking unnecessary risks, while hedge funds (many of which are staffed by former traders) have more scope to use their bond holdings as a source of liquidity.

“It’s a pretty big conceptual change as a means of trading. Compared to what they have been used to for the past 30 years, relying on relationship dealers to make bids and offers, this is asking them to act very differently,” says Scott Eaton, chief operating officer of MarketAxess Europe, which launched its all-to-all protocol, Open Trading, in the US in 2012 and in Europe in 2015.

Nonetheless, interest in all-to-all is growing and about half the liquidity provided via Open Trading is from buy-side clients. These protocols will not, however, eliminate dealers’ role as intermediaries. Andy Hill, senior director at the International Capital Market Association, says: “I think it’s naïve, or wildly optimistic, to think we can move to a pure client-to-client or all-to-all system. There is still a role for banks to play – they will still be working orders using some degree of inventory and their own axes.”

Axe-driven innovations

Another innovation is the rise of alternatives to RFQ, which accounts for 95% of fixed income e-trading, according to BIS research. Despite doing little to enhance liquidity, RFQ has survived the transition from voice to e-trading because it prevents information leakage, a primary concern of investors. But some platforms have found a way around this; by connecting parties based on axe data rather than bids and offers.

“That is the paradigm shift that’s starting to happen: the move from a quote-driven market to an axe-driven market,” says Mr Hill. “Some of these platforms don’t even facilitate the execution. Rather they put people with matching interests together, following which they execute off screen or on another platform.”

One example is Algomi’s Honeycomb Network, which allows buy-side firms to access information about dealers’ current and past axes, helping them identify potential counterparties without risking a price move against them. Another is the dark pools launched by Liquidnet in September 2015 and MarketAxess in June 2015. The latter’s Private Axes protocol allows clients to post axes privately in their own system, and alerts other clients if it matches their recorded preferences. It is anonymous (there is no public broadcasting) and trades are negotiated and executed bilaterally.

It is hoped that advancements such as these will allow big or particularly illiquid trades to occur electronically, but today European investment-grade bond tickets of €5m or more are still generally done via voice RFQ.

Dealer adaptations

Banks culling their bond-trading desks have made headlines in recent years, but this should not be misconstrued. Traditional dealers have not been sidelined by stricter regulations and more e-trading platforms (the biggest of which they have joined). Instead, they have responded by reducing bond inventories, offloading clients whose trading does not warrant their on-boarding and maintenance costs, and limiting the types of credits in which they make markets.

Some have also set up their own e-trading systems. Contrary to some press reports, these are not intended to compete with the incumbent platforms, but to integrate their clients into investment-grade bonds’ growing and diversifying liquidity web. UBS’s all-to-all liquidity network, Bond Port, is widely regarded as one of the most successful. As of June 2016 it had handled about 1300 trades each day and volume in the second quarter of 2016 was $14.1bn – more than double the volume for the second quarter of 2015. That success is in no small part attributable to its somewhat unique set-up.

“Bond Port hasn’t been developed to be just another platform, it isn’t about inventing trading protocols or getting people onto a specific screen. It’s more an attempt to efficiently link our clients together and allow them to provide or take liquidity from/to each other,” says Mark Russell, UBS’s global head of credit execution services. “Bond Port has all of the components we think are needed for success. Yes, we have the liquidity pool, but it also includes a component that provides smart order routing and aggregation, and a 22-hour voice desk to offer support when needed.”

Liquidity comes from three sources: buy-side clients (which represent the biggest number of liquidity providers); other e-trading platforms from which liquidity is aggregated into Bond Port; and independent liquidity providers and other bank dealers (predominantly regional players that want to leverage UBS’s global client base).

Crucially, this excludes the balance sheet of UBS, which simply facilitates the trades. “UBS is sitting in the middle as a matched principal agent and allows clients to interact with the network from their preferred systems,” says Mr Russell. “We bring them together and make it easy to trade with many potential sources of liquidity, but clients choose how to interact with the network.” It means they can interact with Bond Port via their existing order management systems.

HSBC uses its e-trading system, Credit Place, to execute for clients in the independent e-trading venues. It is based on RFQ but by early 2017 will move to firm pricing (similar to equity’s old ‘indications of interest’ model) for investment-grade bonds.

Unlike UBS, HSBC is not leaning towards an agency or broker-style of bond trading. “We are a principal market-maker and intend to remain so. That is our business model,” says Mehmet Mazi, HSBC’s global head of traded credit.

He does not believe capital requirements are the crux of dealers’ problems. “It’s a price discovery issue rather than a liquidity problem from a market-maker point of view. Lack of execution venues make instant hedging impossible, therefore fair pricing and risk management is key,” he says. “People talk about the lack of balance sheet dedication, but balance sheet is available for most of what the market is trading day to day. Appetite isn’t being fully utilised as there isn’t enough comfort in the price discovery.”

HSBC is among the banks automating the price-making process. Real-time data and analytics are making calculations quicker, accurate and more consistent, expanding the number of bonds in which it can provide liquidity.

A data-driven future

In 10 years’ time, more of the primary and secondary markets will be electronic, but traditional bank roles will not become extinct. Sales must still explain the merits of a new issue, and investors will pick up the phone to make block or illiquid trades. Even MarketAxess’s Mr Eaton, who expects “much, much more to trade electronically” in the future, says there will always be a role for dealers. “They help with price formation and liquidity provision – particularly during times of market stress,” he adds.

The number of e-trading platforms will shrink to about five, according to market concensus. Some that are active today will fail, such as Bondcube did after just three months, or be absorbed by bigger players, as happened with MarketAxess’s integration of BlackRock’s Aladdin network. The survivors will combine all-to-all connectivity with a critical mass of investors and dealers, and will allow clients to choose between RFQ and dark pool-style protocols for any given trade.

The other key ingredient is data capabilities. “Analytics that support trading based on who is holding or has an interest in which bonds and which counterparties are most likely to be on the other side of the trade will be crucial,” says Mr Hill.

To maximise liquidity, platforms must provide this information to the buy side in particular if they are to make prices. Some describe this as bond trading’s move from being ‘electronic’ to ‘smart electronic’. Indeed, Mr Eaton attributes much of Open Trading’s success to the integration of market data, including from its post-trade subsidiary Trax. The US trade reporting system Trace, and Europe’s incoming Markets in Financial Instruments Directive II, mean data will become more widely available, placing a premium on platforms that can make it meaningful to clients.

But there are still many unknowns, including the evolution of the bond product itself. For instance, some believe investment-grade bonds’ inherent illiquidity means secondary market exposure will be gained via exchange-traded funds.

Such uncertainties make some people wary of predictions. “I wouldn’t bet on any possible future market structure,” says Mr Russell. “We remain as flexible as we can and work closely with our clients.”

PLEASE ENTER YOUR DETAILS TO WATCH THIS VIDEO

All fields are mandatory

The Banker is a service from the Financial Times. The Financial Times Ltd takes your privacy seriously.

Choose how you want us to contact you.

Invites and Offers from The Banker

Receive exclusive personalised event invitations, carefully curated offers and promotions from The Banker



For more information about how we use your data, please refer to our privacy and cookie policies.

Terms and conditions

Join our community

The Banker on Twitter