An E-Trading Treasury Market 'Flash Crash'? Not So Fast

By Anthony J. Perrotta, Jr., TABB Group

Originally published on TABB Forum 

Volatile swings in equity markets have been blamed on high-frequency trading strategies. On Oct. 15, the US Treasury market woke up and realized it was sleeping with the same enemy. But directing blame at technology and innovation merely diverts attention from the real problem – a structural imbalance in fixed income markets.

Franklin D. Roosevelt sat before Congress on a crisp December morning in 1941 and declared, “Yesterday, Dec. 7, 1941 – a date which will live in infamy – the United States of America was suddenly and deliberately attacked”; you know the rest. By invoking FDR’s words, I am not suggesting the assault on the US Treasury market on Oct. 15, 2014, was akin to the attack on Pearl Harbor. Instead, I am submitting for consideration the notion that the day should serve as a wakeup call for market participants and regulators.

The OTC fixed income markets are changing due to a variety of catalysts. Regulatory reform is forcing large banks to reassess their business models. Inflated government budgets, slow economic growth, and low inflation are making debt more attractive. Accommodative monetary policies are artificially keeping interest rates in check, fueling an unabated pipeline of debt issuance. Asset owners are increasingly growing larger and more concentrated. Investment philosophies are converging, becoming more symmetrical. It seems while the size of the market is growing, the universe of players that operate within that market is getting less diverse.

Combine the regulatory changes with the current market structure and you get a precarious and volatile cocktail. High-frequency trading, automation, and the electronification of execution are all convenient excuses – after the fact – for increased volatility in markets, but have almost nothing to do with the cause. That honor goes to structural imbalance – i.e., the declining ratio of liquidity provided to liquidity demanded, which has evolved since 2004. Nowhere is this more obvious than in corporate bonds, where the impact of dealer liquidity has declined by a factor of three over the past 10 years (see our report, “Corporate Bond Conundrum,” September 2014). Now we’re getting a taste of how this might be affecting the US Treasury market.

In October, the stage was set overnight in Europe, when weak economic data renewed concerns for a deteriorating global economy and Ebola fears permeated the market. Fund redemptions at PIMCO due to the announced departure of bond guru Bill Gross complicated the matter (i.e., the resultant proceeds had not yet been reinvested). The market was already biased toward a short-sided position, so the market was technically vulnerable.

The UST10Y was trading at 2.20% at 4:00am EST. When US retail sales for September came in weaker than expected, buyers emerged and the yield on the UST10Y fell steadily to 2.00% over the next hour, before suddenly dropping as low as 1.86% in a matter of minutes. Not long after, sellers instantly drove yields back to 2.0% before taking the remainder of the day to methodically work their way back to 2.15%. The Financial Times reported the move as a seven standard deviation break from the intraday norm. That certainly qualifies as an aberration (i.e., that type of move should only happen once every 1.6 billion years); but we are inclined to believe this is the beginning of a period in which situations such as this may become more prevalent.

The market is clamoring for a return of volatility, just not this kind of volatility. In the wake of such a drama-filled day, investors, dealers, and regulators are all seeking to identify the culprit responsible for this type of outlier trading session. In the court of public opinion, electronic trading appears to be the fall guy, but we think the focus may be directed entirely in the wrong place.

The true perpetrator is market structure. Under the current, principal-based risk model, liquidity providers – traditionally large banks with significant amounts of capital – provide liquidity on-demand (a.k.a. “immediacy”) to investors. As the amount of capital these banks have at their disposal and committed to market-making declines due to regulations imposed by the Dodd-Frank Act and Basel III Accords – including the Volcker Rule and the liquidity coverage ratio (LCR) – the likelihood of volatility increasing is greater. When viewed in conjunction with the fact that the US treasury market is now $12.2 trillion and the Fed has reduced the outstanding “float” of USTs through quantitative easing, the story gets more compelling. Factor in the notion that asset managers are more concentrated and larger, and the amount of on-demand liquidity requested can sometimes overwhelm the liquidity providing universe.

The development of electronic trading isn’t the cause of volatility, but it certainly can and will contribute to accelerated price movements when a dearth of liquidity exists. The fixed income markets are in the nascent stages of automation; not every market participant is operating with the same level of resources or technology. In fact, the disparity between market participants is probably akin to watching Ferraris zip past bicyclists on the way to the same destination. Undoubtedly, this contributed to the abnormal volatility we experience on that “day of infamy” in October.

The UST market was short from a position perspective, economic data was weak, people were afraid of Ebola, PIMCO redemptions had to be reinvested, Europe was receding, the float of USTs was precariously tight, and people were fleeing all at once to the only safe harbor – the US Treasury market.

Who dare step in front of that freight train barreling down the track? Typically, opportunity would present itself to a group of primary dealers that would continually short bonds into the updraft, knowing the move would eventually correct. Having capital and servicing clients in times of stress was once a profitable luxury; that aspect of the business is fading fast, and along with it, the willingness of dealers to commit to principal-based market-making … even in the homogenous and commoditized US Treasury market.