Oases Forming On The Liquidity Landscape
James Cooper, Head of Execution at Troy Asset Management argues that institutional dealers are refusing to concede defeat in their struggle to cope with a challenging liquidity environment.
Barely a day passes without new commentary on the scarcity of liquidity in financial markets. Broker dealers in particular have been keen to point out how difficult it is nowadays for buy-side firms to implement trades quickly, efficiently and with minimal impact. This is especially the case for fixed income trades but the observation is made about equity markets too.
Recent flash crashes (such as August 24, 2015) have been seen as symptoms of the liquidity crisis and have shaken investors’ and regulators’ faith in the well -functioning market objective.
Ample liquidity and an ability to execute trades effectively are important for the broader economy as a robust market place facilitates capital formation by corporates both large and small. But the question remains, is there really a liquidity crisis or is there, as many observers have suggested, merely a problem with “the plumbing”? Is trader behaviour and market structure merely being slow to catch up with a changing regulatory environment? Are there grounds to believe the industry will adapt to solve the problem?
The reasons behind the perceived liquidity problem are well publicised but are worth quickly revisiting.
Volumes have declined since 2009 for a number of reasons. Firstly, the G20 rules on use of capital by investment banks have had a significant impact on volumes through different channels; the ability of fixed income broker dealers to hold inventory and offer acceptable spreads to their institutional clients has significantly diminished. The same rules though have impacted equity traders too and market-making desks are being much stricter about how and to whom they offer their balance sheet. Scarcity of bank capital is felt further down the chain too, as hedge funds – traditionally a good source of trading flow – find it harder to tap into those constrained prime broking balance sheets.
Secondly, the Volcker rule has decimated proprietary trading. There are virtually no “internal hedge funds” left among the London-based banks and these reliable commission payers have left a significant hole in trading activity compared to the pre-crisis and pre-Volcker years.
Thirdly, hedge funds have, on average, lengthened their average holding period. Ritesh Shah of Citadel (Asset Management) had this to say in a recent interview.
“More broadly, given the higher transaction costs resulting from these shifts, we have moved towards more fundamental-based, longer-duration investing as opposed to model-based, shorter-term investing; our strategies as well as our personnel reflect that new reality”.i
The move towards a buy and hold mentality by hedge funds like Citadel are generally leading to less activity than previously.
Finally, the activity of retail investors has slowed to a trickle in recent years. These participants used to demonstrate reasonable turnover rates but have now mainly entrusted their investment decisions to the lower turnover mutual funds.
But liquidity is not merely about volumes. Information leakage has been a large and ever growing problem. The recent advances in HFT techniques have meant that in 2016 an institutional equity order in the US must run at 3% of volume or less for it not to be detected by the more predatory traders. This is a seismic change on 2007, when institutional dealers were routinely passing orders with instructions to participate at 33% of volume.
Impact costs have become almost impossible to control in the current environment because of the febricity of the Continuous Limit Order Book (CLOB); whilst spreads may be narrow, displayed order book liquidity is truly a mirage: cancellation rates have been a hot topic in recent years, but for as long broker algos are designed to deliver price improvement and dodge negative selection, and for as long as electronic market makers cancel at the first sign of momentum, the CLOB will remain a dangerous place from which to draw water.
It was hoped that dark pools would help with information leakage, but institutional investors have been badly let down. Trade sizes have plummeted and the short term reversion is little different from the CLOB reversion.
The final piece of the liquidity, or impact cost, equation is volatility. Here it is reasonable to expect volatility in its various forms (but best measured by the VIX) to start rising after recently hitting all-time lows. Heightened volatility widens spreads and increases total costs of execution.
So given the decline in volumes, the lack of order book robustness and a likely increase in volatility, it is difficult to see how the institutional dealer can adequately and safely quench his thirst.
Well in fact there are significant grounds for optimism on all these fronts.
Volumes on the uptick
The chart above shows how volumes as a percentage of total free float have actually been improving (modestly but steadily) for more than two years. The data above is for the FTSE but the pattern is almost identical for the S&P and Eurostoxx 50.
It is difficult to isolate exactly the reason for the modest inflexion in volumes but there are three possible explanations. Firstly, High Frequency Trading (HFT) firms are becoming significantly better capitalized. Companies like Jump Trading and Virtu have joined Citadel as robust, well-funded institutions in their own right. Secondly, the recent good health of CTA (trend following) businesses and other systematic strategies has meant greater assets under management in that segment and greater resulting trading volumes (especially visible in the first two months of 2016). Thirdly there is an inkling that institutional dealers are daring to come to the water’s edge more frequently.
There are two reasons why institutional dealers might be trading more. Firstly, they are benefitting from the impressive development of the investment banks’ Centralized Risk Books (CRB’s) that have been honed over the past five years. These have been particularly helpful for executing transition or programme trades but they also provide a useful source of liquidity when the bank’s risk is later unwound into the market. Secondly, institutional dealers are finding alternatives to the two main sources for European trading in the last 25 years, which have been the CLOB and the single-stock market maker.
We have already looked at the failings of the CLOB but it is worth remembering that the construct only came into existence to the UK with Sets in 1997 and arrived in the US five years later and so does not need to be a permanent fixture. As for the single-stock market maker, the value of this liquidity source has diminished as he or she finds it almost impossible to unwind his risk without creating impact and holding up the client from continuing with their business.
i Top of Mind Interview, August 2, 2015 – Goldman Sachs Global Macro Research