European buy-side traders are increasingly voicing frustration about feeling pushed to trade at low participation rates via strategies like volume weighted average price (VWAP) and percentage of volume (POV), whilst also keen to utilise dark block liquidity.
Traders and investment firms use these strategies for a variety of reasons including the lack of clarity on price direction, as well as fear of market impact and being adversely affected by the price of a dark block.
On top of this, increased volatility seen since the start of the pandemic has led some to ‘trade the average’ in anticipation of being caught on the wrong side of price moves, which can increase the opportunity cost and timing risk through longer order durations. Increasing order durations might be ok when there is no alpha present in the order, such as for a CRB desk with a neutral alpha unwind, however, for most buy-side traders this is not the case and managing for momentum is crucial. Using low participation rate strategies while utilising dark block liquidity therefore is challenging but is there a way to do it effectively?
The use of low participation rate strategies such as VWAP and POV in the market is significant. The recent algorithmic survey compiled by The TRADE in spring 2021 showed 59% of respondents cited using VWAP and 57% using POV1. Other industry reports have shown that approximately 50% of algo strategy usage in the market today is also through these strategies2.
As well as contending with increased volatility, traders’ ability to forecast appropriate price levels at which to potentially change trading strategy has been partly hindered by changes in market microstructure. Overall market volumes have reduced but, more importantly, where specific volumes are found in the market has changed. For example, composite lit continuous volumes are down from 73% of total market volumes in July 2016 to 58% in April 20213, a 20% reduction in the liquidity that VWAP and POV algos typically interact with when creating their trading schedule and reacting to being ahead or behind of that schedule. In addition, the closing auction has become an even larger part of any VWAP calculation, making the proportion of any order traded during the continuous session even lower. In contrast, overall dark trading volumes have returned to pre-MiFID II levels4 and dark liquidity has become blockier in nature, with ~40% of dark principal now above LIS5.
Tapping into dark liquidity helps traders reduce their trading costs, as cited in a recent FCA paper looking at the effect of banning dark pools. The paper showed better performance vs. the arrival price (implementation shortfall) for orders with a larger proportion of executions in venues with lower pre-trade transparency levels such as in dark pools6. This is also evident in the savings Liquidnet clients have made through trading blocks, as seen in our ‘Value of the Block’ analysis. Clients saved on average 19.7bps of market impact and spread costs on executions in Stoxx 600 names during 2020. The timing risk saved through block executions in Stoxx 600 names was also significant, saving 1.4 days (743 minutes) of exposure in the market vs. trading at 10% POV7.
With the benefits of dark block liquidity evident, and the need to trade via low participation rate strategies also clear, traders have increasingly asked Liquidnet to develop solutions that help them to reconcile these two positions: trading over longer periods of time and at lower participation rates, whilst smartly accessing dark block liquidity. Liquidnet Smart Blocks is a global proprietary solution that aims to achieve this by dynamically stepping in and out of the block market, at opportunistic price levels that the algo has determined are appropriate levels to trade, while trading along a VWAP or POV strategy as chosen by the trader.
The opportunistic price levels determined by the algo are calculated using pre-defined rules related to choices made when deciding the benchmark and the degree of movement required, for example 10% of the stock volatility, on-side versus the stocks correlated basket. This negates the need to have a clear price-level in mind pre-trade or a clear view on price direction, something traders often are unclear on at the start of an order.
In the first example shown (Image 1), depicted as a buy order for Just Eat Takeaway (TKWY NA), the benchmark was chosen as the correlated basket signal (which is our proprietary signal from Parameta Solutions, designed to highlight when a stock has dislocated from its peers) and the required offset was 15% of the stock volatility, so in this instance 25bps of movement. As the stock price moves throughout the day, there are periods of time when the price is favourable compared to the threshold-adjusted correlated basket, activating the Smart Block signal and the search for block executions in dark venues for the order. As the stock rallies relative to the basket, the signal deactivates and the POV order continues as its pre-set rate of execution.
The second chart (Image 2) is a representation of a real VWAP order by a Liquidnet client using Smart Blocks, where the trader used a 10% volatility offset vs. the interval VWAP. As the price became favourable, Liquidnet Smart Blocks became active and the algo achieved a block execution using a conditional order type in an external MTF, at favourable price levels, outperforming the interval VWAP by 21.5bps.
If the challenges outlined here seem familiar to you as a buy-side trader, dynamic POV or VWAP combined with the functionality of Liquidnet Smart Blocks could help you to realise the benefits of dark block liquidity, while trading at low participation rates via a longer time horizon strategy. Smart Blocks is in the process of being made available globally, so please speak to your Liquidnet representative to explore the functionality in more detail.
1 The TRADE 2021 Algorithmic Trading Survey
2 Virtu European Market Landscape February 2021
3 Big XYT – _Liquidity Trends since July 2016 – _Lit Exchange/MTF April 21 vs. July 16.
4 Big XYT – _Liquidity Trends since July 2016 – _Dark MTF April 21 vs. July 16.
5 Bloomberg, LIS % of Dark Market, Q1 2021
6 FCA Occasional Paper 60: Banning Dark Pools
7 Liquidnet Analysis – LIS Executions in Stoxx 600 names via Liquidnet, January to December 2020