Conditional orders in Dark Pools

Discussion in 'Order Execution' started by qlai, May 31, 2019.

  1. qlai

    qlai

    Conditional Orders: The Great Liquidity Aggregator
    [​IMG]
    Campbell Peters

    TABB Group

    Follow | Profile | More

    [​IMG] Share
    As the US equities marketplace has grown more and more fragmented, conditional orders quickly are becoming the de facto method for buy-side traders to aggregate liquidity in a decentralized marketplace. Despite the widespread adoption of conditional orders, however, the complexities around evolving use cases, as wells as the fact that each ATS offers its own flavor of the order type, can be confusing. TABB Group analyst Campbell Peters examines the various conditional order functionality across dark venues, the automation of the conditional order workflow, and the growth of conditional liquidity-seeking algos beyond blocks.

    [​IMG]
    A conditional order represents an interest to trade in multiple trading locations, without committing an order to any single one. When the other side of a trade is matched against a conditional order, the trading venue — before executing the order — notifies trading parties to re-affirm their desire to trade. This workflow allows traders to represent their orders across multiple venues without the risk of double execution. As the US equities marketplace has grown more and more fragmented, it has become difficult for traditional buy-side traders to divine on which of the 33 dark pools the other side of a trade will appear. To ensure that they do not miss a trade, buy-side firms increasingly have been turning to conditional orders to aggregate dark liquidity for trading large blocks of shares.

    And these orders are becoming even more popular. TABB Group’s soon-to-be-released 2019 US Institutional Equity Trading study found that close to 75% of the buy-side firms interviewed employ conditional orders to source liquidity, with more than 30% planning to increase their use over the next year.

    Conditional orders, originally offered only by trading platforms that facilitate trading in large blocks, are now offered by a wide range of dark pools. And though similar functionality exists in other asset classes – such as “last-look” in FX trading and request-for-quote (RFQ) in fixed income – the conditional workflow is unique to equities markets due to their multilateral nature.

    Despite the widespread adoption of conditional orders in US equities trading, many investors know little about them. That’s not surprising considering the complexities around evolving use cases for conditionals, as wells as the fact that each ATS offers its own flavor of the conditional order type. Despite the variation in conditional offerings, however, the underlying workflows that define conditional orders hold true across every venue, making them accessible to liquidity-aggregating algorithms.

    Defining conditional orders

    When defining a conditional order, it helps to separate the functionality provided by conditional orders from the rules applied by each trading venue. While each ATS offers different conditional functionality, two characteristics persist across all dark equity trading venues:

    1. Conditional orders are non-binding, as opposed to firm, which means that when an order matches with a contra-side order, traders are not required to execute the trade until they have reconfirmed trading intent. For this reason, a firm order is another term for non-conditional.
    2. When two orders match, a negotiation period is triggered. During this predefined time window, both counterparties must decide either to firm up or to decline the negotiation. To firm up is to opt to execute a trade with a specified quantity, while to decline is to walk away from the trade.
    The underlying concept of conditional orders has existed, as one ATS operator explained in a recent TABB Group interview, as long as the US equities market has existed. Prior to the birth of electronic equities trading, a trader could call up his broker, saying, for instance, “I’m selling 100,000 shares of stock XYZ. If you find someone willing to buy, I’ve got more to sell.” The broker would call back: “I found a buyer, how much more can you sell?” At this point the broker would mediate the negotiation between the two traders based on the quantity both traders were willing to trade.

    Today, this conditional order process has been fully automated either through a pop-up notification (human-directed) or electronic messaging (algo-driven), and increasingly the need for a telephone call has been replaced by conditional orders.

    Variants of conditional orders across venues

    While conditional orders are always non-firm and need to be re-confirmed, the way they are implemented in various dark pools/ATSs can vary. Exhibit 1, below, lists the venues that support conditional orders and illustrates the differences between the functionality provided by each ATS. The information on ATS features and functionality is taken from Form ATSs and other public disclosures.

    [​IMG]

    Algo-driven vs. human-directed

    Conditional orders vary in functionality and usage across the various ATSs, but they’re intended to serve one common purpose: Allow traders to search for hidden liquidity for trading large blocks of shares in multiple places simultaneously, without risk of being overfilled. Due to the fragmentation of US over-the-counter equities markets, traders often must search 10 or more dark venues to identify a counterparty when trying to trade a large block of stock. Conditional orders enable this search to be done very quickly — negotiation periods usually are 0.5 second to 2 seconds — and allow the workflow to be handled in either an automated or human-directed way.

    [​IMG]Many venues that support conditional orders allow human-directed conditional orders. This is usually achieved through blotter integration with an existing OMS and/or via an OMS-lite that is specific to the venue. Once a match is found, a trader receives a pop-up notification on his front end indicating that the negotiation period has commenced. At this point, the trader can elect to firm up or decline.

    When searching for dark liquidity for trading blocks of shares, human traders will submit nearly identical conditional orders to multiple dark pools simultaneously. When they receive a pop-up notification indicating a match, they will cancel their other conditional orders before firming up in the negotiation. Traders can cancel other orders without paying a penalty, preventing an order from being filled in multiple locations.

    All venues that support conditional orders allow the same workflow to be achieved via an algorithm. In this scenario, the entire process described above is replicated by a liquidity-seeking execution algorithm, a standard piece of most broker offerings. In the algorithmic workflow, there is no human trader watching for pop-up notifications. Negotiation periods and responses are fully automated, allowing traders to utilize the conditional functionality in a manner that is much less time-consuming and further mitigates risk of overfilling.

    Beyond block trading

    Initially developed exclusively for trading large blocks of shares, conditional orders have pivoted to be the de facto method for aggregating liquidity in a decentralized marketplace. Traditional buy-side traders continue to submit human-directed conditional orders for trading large blocks, and sophisticated portfolio managers have modernized this practice by implementing liquidity-seeking algorithms for moving considerable size.

    Retail execution platforms and systematic internalisers demonstrate additional use-cases for algorithmic liquidity aggregation, as execution algorithm providers described in recent interviews with TABB Group. The trend of marketplace fragmentation is not going away, as SI volume has grown to nearly 10% of the total off-book volume in Europe since MiFID II was implemented. In March, Deutsche Bank announced that it planned to launch a single-dealer platform in the third quarter of 2019. This would be the American equivalent of a systematic internaliser and would add to the fragmentation of the US equity market, furthering the need for liquidity aggregation.

    All these tailwinds point toward even wider adoption of conditional orders. Due to their success with buy-side block trading, many market participants are exploring novel use cases for conditional workflow. By successfully addressing the needs of the buy side and beyond, conditional orders have demonstrated their value in the current state of liquidity fragmentation and will become ubiquitous on institutional equity trading floors.
     
    Love2Trade$ likes this.
  2. 无论谁拥有丰富的暗池经验,我总觉得我使用的暗池不容易填充。