Type to SearchView Tags

Application of RPA, AI, ML in Post trade Ecosystem
Muralitharan K Business Manager, HCL Technologies, Malaysia, SDN, BHD. | May 18, 2020
408 Views

Introduction

Most of us are aware of buying and selling securities like equities, fixed income instruments, foreign exchanges and derivatives. These are widely spoken and covered by press in modern world. However very few discuss and illustrate how these trades are settled through post trade functions where the buyer ultimately receives the securities and seller receives cash.

Post trade is a process comprising of services that are performed to the execution of trade and it includes:

  • Clearing
  • Settlement
  • Affirmation
  • Confirmation
  • Allocation
  • Matching
  • Custody and Asset Servicing
  • Safekeeping Assets
  •  Income Collection
  • Corporate Action
  • Proxy voting
  • Tax reclamation
  • Lending and Collateralization
  • Risk Mitigation and
  • Reporting

Limitations in Post Trade

Business, technology, and regulatory changes has reshaped the industry practices, roles of market participants and the market structure. Market participants want themselves to be ahead of their competitors to remain competitive in marketplace.

Buy side and sell side firms had heavily invested in their front office Order Management system (OMS) / Execution Management systems (EMS) and infrastructure to ensure best execution, multi asset coverage, achieve straight through processing, low latency order execution and algorithms for trading. In comparison, most of the bank post trade systems are either working on the legacy systems or older technology monolith requiring significant manual intervention in business operations and process flows. These systems have undergone enhancements over the years beyond the boundaries of design principals, thus introducing inefficiency, operational risk, and increase in operation cost.

The current global pandemic situations of COVID-19 is undoubtedly going to push further pressure on reducing the cost of trade with an aim to boost their margins by identifying and removing the inefficiencies in the trading ecosystem.

Coronavirus

“Coronavirus (COVID-19) could reduce world trade by up to a third”, - World Trade Organization

Application of RPA, AI, ML in Post Trade to overcome inefficiency

Advancement in technologies like Robotic Process Automation (RPA), Machine Learning (ML), Artificial Intelligence (AI) offers various products and services to reduce manual intervention and improving operational inefficiencies in the middle and back office operations. 

A report by Goldman Sachs state that by 2025, AI and ML will optimize cost and improve revenue from US$34 billion to US$43 billion in the finance sector.

RPA, also known as bots can be used to increase efficiency in the following areas.

  • Repetitive work prone to human error.
  • Rule based calculations and workflows.
  • High volume-based transaction.

How RPA can be used in post-trade process?

  • Settlement and fail management
  • Trade Reconciliation – Monitor, reconcile, and Report 
  •  Regulatory and management reporting
  • Transaction, billing, and invoice – monthly and adhoc billing request
  • Communication channels like E-mail/chats

Cognitive technologies are more advanced than bots which analyze huge amount of data and mimic the human judgmental behavior.

Applications of RPA and cognitive technologies can be integrated by investment firms to extend automation, remove  inefficiencies, increase revenue, and customer satisfaction.

How cognitive technologies can be used to improve the clearing and settlement process?

The illustration below shows trade settlement failure in trade back office and how AI and ML can be used to overcome the inefficiency and boost the margins.

Discover Problem

Discover Problem

Using AI tools one can analyze vast amount of historical transaction data including failed post-trade settlements. On trade date or value date -1 the ML program could be run to identify the anomalies or gaps in the current process. This can be built on top of current systems to evaluate these failed trades and fix them in fraction of seconds, which otherwise would take minutes if not hours through trained back office personnel. Through discover mechanism AI tools can be programmed to identify the problem pattern or root cause of the issue with the available historical transaction data.

The current economic situation due to COVID19 pandemic is forcing the investment firms to have a relook at avenues for reduction of cost and operational risk using RPA, AI, and ML in Post Trade Settlement,

Investigate and Predict

AI tools can analyze large amount of data in fraction of seconds. When a trade fails there could be numerous reasons such as incorrect settlement account, incorrect clearing account, incorrect settlement currency, trader error, invalid counterparty code, and wrong value date. AI tools can be used to analyze the trade failures and automated to fix them in fraction of seconds, which would amount to significant cost saving for the investment firms.

“It takes a human five to ten minutes to reconcile a failed trade. A bot can do it in a quarter of a second.” – American Banker

After analyzing the historical data, the AI tools can be programmed or re-programmed through pattern recognition analysis to predict the likelihood of which trade might fail through its algorithms and can redirect the failures in more efficient, accurate and automated way. This can create more vigilant and self-dependent settlement process leading to fewer exceptions and lesser operational cost and greater savings forthe investment firms.

“UBS introduces an automated AI solution for managing clients’ post trade allocation requests. The system reads client emails to determine how they want to allocate large block trades between funds and then executes the trades with no manual intervention”.

Conclusion

The current economic situation due to COVID19 pandemic is forcing investment firms to relook at avenues for cost and operational risk reduction. Application of RPA, AI, and ML in post-trade ecosystem is worth considering for these reduction targets.

According to IDC, due to COVID-19, APEJC (Asia Pacific excluding Japan and China) IT spending in 2020 will plunge to 1.2%, down from its earlier projection of over 5.2% in constant currency terms.

Reference

  1. https://www.broadridge.com/_assets/pdf/broadridge-what-are-the-applications-for-artificial-intelligence-in-securities-finance-andcollateral-management.pdf
  2. https://capitalmarketsblog.accenture.com/improving-trade-settlement-fail-prediction-with-artificial-intelligence-and-machine-learning
  3. https://www.globalcustodian.com/blog/case-reforming-post-trade-tech/
  4. https://cfte.education/2019/07/30/how-ai-used-investment-banks-trade-settlement/
  5. https://www.finextra.com/
  6. https://www.weforum.org/agenda/2020/04/wto-financial-crisis-coronavirus-covid19-recession-trade-global
  7. https://www2.deloitte.com/content/dam/Deloitte/nl/Documents/financial-services/deloitte-nl-fsi-fsi-disruption-asset-servicing-secure-report.pdf
  8. https://www.idc.com/getdoc.jsp?containerId=prAP45596719
  9. https://www.computerweekly.com/news/252480568/Coronavirus-takes-toll-on-IT-spending-in-Asia