Analyse this: why the buy side is hungry for FX trade data

The use of third-party transaction cost analysis (TCA) has become part and parcel of the way buy-side firms assess liquidity providers and the effectiveness of their execution algos.

On the post-trade side, this focuses on the cost of execution and drivers such as information leakage from the LPs. On the pre-trade side, it’s more concerned with forecasting the expected performance of the algo versus execution, and how this could be influenced by factors such as currency pair, trading venue and speed.

It is increasingly an expectation of buy-side clients that their sell-side counterparties should plug all this data into third-party TCA providers like BestX and Tradefeedr. As a result, a lot of buy-side trading decisions are being shaped by the data they get from these vendors.

Communal transaction cost analysis (TCA) data is often a good starting point for buy-side firms looking to assess their counterparties and follow what their peers are doing. For those firms that are new to using algos and have fairly complex execution protocols, using aggregated data from third-party vendors could be the best way to gain initial insights on their flow.

But data from third-party TCA providers may not provide the entire picture. There are other data elements which go into shaping the total cost of the trade that may not be available through the vendors, such as client-specific last look hold times or rejection rates.

Vendor-driven TCA data may not meet all individual execution goals. For a more complex buy-side firm, a collection of normalised execution data may not provide the insights it needs.

Furthermore, not all clients may be happy to have their data collected and published in a communal pool. This issue was raised at the FX Markets Europe event this month, where one bank participant said it has non-disclosure agreements with some clients that do not want their trade data sent to these vendors because of its commercial value.

Taking a page out of their equities playbook, an emerging trend at some of the more sophisticated buy-side firms is the hiring of in-house quants to assess the quality and overall cost of their bank execution algos. They then use this analysis to supplement the information they get from analysing the communal TCA data, which together helps shape execution policy around how they trade.

There is where other technology solutions, such as algo wheels and liquidity provision analysis, could become more widely used by the buy side.

Banks also see this as an opportunity. Many hope that providing complimentary TCA and pre-trade analytics will entice buy-side firms to execute on their single-dealer platforms.

Dealers say this is particularly relevant for more complex derivatives instruments and “lumpier” trades. But how independent and useful this data could be for buy-side best execution mandates is up for debate.

As the buy side becomes more advanced in collecting and understanding TCA, some believe firms will use a combination of data from third-party vendors, banks and internal quants. But to retain full control and ownership of their execution information and flows, there may be an emphasis to keep this in house.