Accounting for transaction size and market conditions in FX TCA

Accounting for transaction size and market conditions in FX TCA

The evolution of Transaction Cost Analysis away from a ‘box ticking’ exercise towards making TCA relevant to execution continues apace.  New Change FX (NCFX) is at the forefront of delivering API-driven analysis that can be used in real time to re-evaluate the prioritisation of liquidity providers (LPs) within a smart order router.

At present, the smart order router tends to be rather basic, taking account of little more than price and fill rates, when of course far more is needed to correctly prioritise LPs.  The issue for TCA providers is that post-trade processing means that their analysis is not relevant to the live execution, and in any case, the analysis is not usually in a machine ready format that can be routed directly into the execution logic.

By introducing live APIs that run analysis in real time, NCFX are able to deliver TCA results in machine ready format that can be used in a logical process to reprioritise the LPs.  This in turn guarantees best execution by using all possible measurement parameters.

A recent Aite report[1] on transaction cost analysis (TCA) identifies that one of the most significant issues facing the FX TCA industry is the lack of data for a given size of transaction.  “…when it comes to sizable pricing and transaction data for analytics, FX…impose(s) the greatest challenge for the buy-side to source, consolidate, and normalize data.”

NCFX has been researching the issue around size and market conditions for around two years now and is pleased to have created measures that normalise results for both size of transaction and market conditions.  These are available via API so that they can be used within the live execution process, delivering results that can affect the choices being made about liquidity selection and enable users to accurately compare transaction costs across trades of differing size executed in differing market conditions.

The Unit Cost of Volatility

This API scales effective spread by units of volatility. Effective spread is calculated as the difference between the dealt rate and the mid-rate when the order was externalised to the market. Volatility is converted into a USD value that factors the amount of time taken to complete the trade and the amount traded.

By normalising costs with realized volatility users can compare costs across currencies, times, periods of different volatility and size of trade, allowing comparison of costs using the same metric; costs incurred as a ratio to the market risk prevailing during the life of the trade.

UCV re-expresses costs on a risk-adjusted basis.

USD Cost of Volatility

This API calculates the inventory risk in USD terms of an open currency position based on the prevailing market conditions (volatility) through time. Using the period given by the start time and end time of an order, the model multiplies the amount traded by the volatility and converts into USD dollars.

By normalising costs with volatility users can compare costs across currencies, times, periods of different volatility and size of trade, allowing comparison of costs using the same metric; costs incurred as a ratio to the market risk prevailing during the life of the trade.

Accessing the NCFX APIs

The APIs are available through the NCFX Portal – please contact info@newchangefx.com for details and access.

(https://portal.newchangefx.com).

[1] Multi Asset Class TCA in 2020: Opportunities Abound Amid a Pandemic. Yue Malan & Audrey Blater.  9th December 2020.