Although anticipated by some, the mid-2013 sell-off that upended bond markets proved stunning in its ferocity. In the corporate-bond market, investors sold for any number of reasons. But a big contributor to the swiftness and severity of the sell-off was liquidity’s fragile state, resulting from the market’s structural problems.
The industry now finds itself in a challenging transition, according to Corporate Bond E-Trading: Same Game, New Playing Field (PDF–1.5MB), a recent report from McKinsey and Greenwich Associates.1 Its old model—dealers carried inventory and took risks—is giving way to a novel capital-light paradigm that requires new skills, as well as changes in the technological infrastructure, not least for electronic trading (e-trading).
To date, e-trading has not conquered trading in corporate bonds the way it has other asset classes, most notably cash equities. Indeed, our research found, market participants believe that the day it will do so is a long way off—and may never arrive. Nevertheless, e-trading will play an important role in helping market participants to trade more efficiently and cut costs. Dealers and investors must prepare now for the next step in e-trading’s evolution.
Liquidity’s ebbs and flows
While a semblance of calm appears to have returned to the corporate-bond market, its challenges remain. Regulatory reform has handcuffed dealers, some say, by raising capital requirements; dealers have responded by cutting net inventory. The convergence of buy-side investment strategies and the proliferation of buy-and-hold investors has promoted a “one-way” market, in which it has become increasingly difficult to find the other side of a trade.
Yet perhaps one of the biggest problems is the sheer volume of corporate bonds outstanding. Historically low interest rates have fueled an issuance boom, which is good for the debt capital markets (DCM) business and for “on the run” trading. But that same issuance boom, according to the Securities Industry and Financial Markets Association (SIFMA), also left $9 trillion in US corporate debt outstanding by the end of 2012—a record high, and 2.2 times larger than its level in 2002. A tremendous and potentially unsustainable amount of paper is in investors’ hands.
Ultimately, these problems find their expression in liquidity. Many of the industry’s participants expected it to dry up during the 18 months preceding the recent sell-off, as new regulations were phased in and other problems came to a head. Sure enough, net dealer inventories plummeted (Exhibit 1).
While the amount of corporate bonds outstanding continued to rise after the financial crisis, net dealer inventory fell. However, turnover dropped far less dramatically.
Yet turnover didn’t drop nearly as much. Indeed, through midspring 2013, bid–ask spreads in US investment-grade corporate bonds were not dramatically higher than they had been prior to the global financial crisis (Exhibit 2).
Bid–ask spreads through mid-spring 2013 were not dramatically higher than they were prior to the global financial crisis.
During the recent sell-off, bid–ask spreads in US investment-grade corporate bonds widened—not a huge surprise given the magnitude of the bond-market correction. Although it’s not yet clear how the midyear tumult will ultimately play out, the liquidity skeptics may have the last word. The issuance boom and follow-on trading in these issues have probably masked the market’s liquidity problems. As a case in point, look no further than the record $49 billion bond offering of Verizon Communications in September 2013. That was the highlight of a record month for US corporate-bond issuance, as many companies appeared to bring forward longstanding issuance plans before interest rates spiked again, while yield-hungry corporate-bond investors found interest rates attractive enough to pile back into the market. One day after issuance, the new Verizon bonds accounted for more than 25 percent of the overall trading volume in US investment-grade corporate bonds, according to MarketAxess Research.
The market is currently sending mixed signals about liquidity. August 2013 saw a 14 percent year-on-year drop in average daily trading volumes of US investment-grade and high-yield corporate bonds, despite a 10 percent rise in outstanding corporate debt. But September 2013 showed a 15 percent year-on-year increase in average daily trading volumes, powered by the big jump in issuance. It’s possible that August trading was subdued in anticipation of the new issues to come in September.
Will liquidity get better or worse? The answer matters a lot to broker–dealers, in both the primary and the secondary market. In 2012, for instance, sales and trading in cash credit accounted for $4.5 billion of the $17 billion that the ten largest investment banks generated through overall credit sales and trading, according to Coalition, a market-analytics provider. For issuers too, the underlying problems could present formidable challenges, since the market is a critical source of capital for businesses and a vital cog in the global financial system. Companies depend on a robust primary market, which in turn depends on a smoothly functioning secondary one. If liquidity deteriorates, the ability of companies to raise debt capital could be adversely affected.
The other big source of corporate funding—bank lending—is also unsteady. Recent McKinsey Global Institute research suggests that with cross-border lending flows slowing dramatically, national markets in Europe and other regions may become more balkanized.2 With banks in many countries unable or unwilling to lend, companies will naturally turn to corporate-bond issuance, just when liquidity could be drying up.
The shift to e-trading
If liquidity gets worse, that may accelerate the slow but steady migration toward e-trading. At present, most transactions are done over the phone, with dealers acting as counterparties to investors (that is, the market is quote driven). Many dealers and others, hoping to capture more trading activity in this capital-light world, have introduced new e-trading platforms. However, success has been elusive, and some broker–dealers have redoubled their e-trading efforts in the wake of the recent sell-off.
Generally, the most efficient e-trading is thought to be done through match-based market models, such as central limit-order books; these are prevalent in the most advanced e-trading markets (for instance, cash equities and futures). The structure of the corporate-bond market is not even remotely conducive to that kind of e-trading, however. Corporate bonds are much more heterogeneous than cash equities. The US stock market, at its 1997 peak, boasted about 8,800 listed companies. That pales in comparison with the US corporate-bond market, which had 37,000 publicly traded, TRACE-eligible3 issues outstanding in 2012. The sheer number of issues greatly reduces the probability of multilateral trade matching.
In 2012, the average US stock traded around 3,800 times a day. But the 13 most liquid US investment-grade and 20 most liquid high-yield corporate-bond issues traded, on average, only about 85 times and 65 times a day, respectively (Exhibit 3). Furthermore, when corporate bonds did trade, they involved a great deal more money than the average stock trade did. Not surprisingly, survey4 participants expect quote-driven e-trading platforms, such as the multidealer request-for-quote (RFQ) platforms operated by Bloomberg and MarketAxess, to dominate e-trading for years to come.
Even the most liquid corporate bonds are far less suitable for match-based e-trading than is the average US stock.
The recent market tumult may spur a reevaluation of e-trading. Indeed, those multidealer platforms recorded their best June ever during the sell-off. Is this the shape of things to come? Perhaps. What we can say for sure is that the inventory-driven business model is giving way to a new paradigm that mandates skills and a technological infrastructure consistent with a capital-light world. To ensure optimal positioning for the range of directions the market might take, dealers and investors would be well advised to take action now—to the extent that they haven’t already done so.
First and foremost, to lay the foundation for significant changes to the business model, dealers must align the organization. They will need to break down the remaining resistance to e-trading and to get people thinking more broadly about how it can be used to compete more successfully in so-called flow products—commoditized financial instruments for which achieving scale is paramount. This effort must start with an unambiguous commitment from top management to the e-trading business. Dealers will need to adopt a governance model that balances the centralization of e-trading activities and the related infrastructure (such as connectivity) with decentralized control over product-specific elements (such as pricing, coverage, and hedging strategies).
Dealers should also develop a centralized pricing and trade-processing infrastructure that serves both traditional and electronic channels, thereby reducing time to market, vendor-specific infrastructure, and costs. Ultimately, this central infrastructure should help dealers to navigate the market landscape efficiently and to maximize opportunities for crossing (or matching) buy and sell orders.
Furthermore, dealing houses need to harness the data opportunity. They should begin with a push to consolidate their myriad data interfaces (for example, those with internal systems, clients, market venues, and data providers) and to store data in a centralized, accessible manner, ideally in a single “golden source.” This approach will help firms to feed pricing engines more quickly and effectively, to develop new trading strategies, and to better inform the sourcing and placement of corporate bonds. Such a tracking system should be integrated with sales tools to make the sales force more effective.
Finally, banks should provide incentives that reward sales reps for shifting uneconomical trades and clients from voice to e-trading channels. Motivating the right sales behavior will free reps to deliver more valuable services, such as exploring the needs of clients, generating trading ideas, providing market commentary (or “color”), and offering counsel on complex trades. However, banks will need to attract, cultivate, and retain a new breed of sales talent, with the skills to meet the demands of a new era far removed from the old one, in which dealer inventory largely drove the sales function.
Buy-siders ought to consider a couple of moves. First, they should reassess their investment decision-making process. Managers must factor liquidity into it more prominently and embed related metrics more rigorously, in both primary and secondary markets. Ultimately, managers need to ensure that they are paid for the liquidity risks underlying the corporate bonds in which they invest. Naturally, firms must equip their portfolio-management teams with the skills and tools necessary to operate in the postcrisis corporate-bond world, in which the ability to price liquidity risk systematically could mean the difference between success and failure.
Second, firms must revamp the trading function to navigate a fragmented market landscape more efficiently and to transact in smaller sizes economically. Several steps ought to be considered. To begin, firms should decide which market venues and dealers to engage with when they execute their trading strategies. They should also define their appetite and approach for providing market liquidity as a price maker and for disclosing their trading interests to a broader swath of market participants.
As for technology, investors need to align the architecture of front-end systems (for order management and execution management) with trading specifications. The infrastructure must ensure that traders can easily shift across liquidity venues, thereby minimizing missed opportunities. The buy side will also need to redefine middle- and back-office workflows and to install the systems needed to process, in a scalable manner, high-frequency trading and the corresponding allocations to the underlying fund vehicles.
Buy-siders with sufficient scale in corporate bonds should consider separating the investment and trading functions (if they haven’t already) to assure sufficient specialization on the trading side. To maximize scale economies and crossing opportunities, asset managers with multiple corporate-bond fund vehicles should consider consolidating the trading desks across them. Larger players might create a specialized trading desk equipped to transact more frequently in smaller round-lot, odd-lot, and micro sizes. Ultimately, firms will need to assemble teams of traders ready for a world with a dwindling number of inventory-driven dealers.