A recent study concluded that where you execute large trades can have a significant impact on performance. Volatility was shown to be, on average, 250% higher for other brokers relative to Liquidnet in the one minute prior to a block execution. Significantly increased volatility prior to a block execution is an indicator of information leakage, which can have an adverse effect on execution quality.
For the Asia-Pacific region block trades keep booming as regional share prices climb. Australia is not only one of the region’s major markets but it is also very block-centric with 14.1% of total trading volumes being of block size. The Australian market defines blocks on a three tiered system based on notional value: Tier 1: $1 million+, Tier 2: $500,000+, and Tier 3: $200,000+.
The upstairs market in Australia is busier than ever with the rise of specialist block desks. These desks leverage sales traders to find buyer and sellers, and then facilitate the transactions, often bypassing buy-side trading desks and going directly to Portfolio Managers. Not only does this increase the risk of information leakage, it is inefficient and contributes to a fragmented block market. Our study examined the potential for improved performance leveraging the efficiency and anonymity of a buy-side focused, electronic dark pool.
An increase in stock activity following a large block trade is generally expected as information contained in the block is absorbed by the market. If, however, there is significantly increased volatility prior to the block trade, this suggests that the market was aware of outsized liquidity before it actually executed.
We set out to determine the extent to which information leakage occurs across the block market for Australian equities. Our study of Australian block executions across multiple brokers concluded that there are varying degrees of information leakage across the market, with block executions on Liquidnet exhibiting the least volatility immediately prior to the block hitting the tape.
We investigated price behaviour surrounding Liquidnet block executions and contrasted that with behaviour surrounding blocks executed by other brokers.
The dataset comprised 31,423 block trades (> AUD$200,000 with appropriate condition code) executed over the period 1 Jan 2015 to 30 June 2015 and spread over 424 distinct symbols.
The data was filtered to exclude the following:
- Irrelevant condition codes1
- Non-crossed blocks
- Opening and closing prints
- Executions not clearly attributed to a venue or broker
- Brokers not considered directly competing in block trading
We then assessed the average incremental volatility in the minutes preceding and following blocks executed by Liquidnet and competing brokers.
Immediately apparent in the study was the lower incremental volatility observed for Liquidnet block executions in the period immediately preceding the time of execution. While an increased volatility following a block execution is generally expected as the market absorbs information about the block, increased volatility preceding a block execution suggests information leakage. It is reasonable to correlate the magnitude of volatility with the level of information leakage.
Our findings suggest that leakage, implied by incremental volatility, is on average 250% greater in the minute leading up to the block execution for other brokers versus Liquidnet.
1 Relevant condition codes: S1XT, S2XT, S3XT, NXXT, S, B, CXXT, CPCXXT
Data as of Jan 1, 2015 to Jun 30, 2015. Source: ASX and Chi-X.
© 2015 Liquidnet Australia Pty Ltd. Liquidnet Australia Pty Ltd. is registered with the Australian Securities and Investment Commission as an Australian Financial Services Licensee, AFSL number 312525, and is registered with the New Zealand Financial Markets Authority as a Financial Service Provider, FSP number FSP3781.