In accordance with Fair Value Accounting, a leading German investment bank required a Loan Pricing Service that could be easily extended and integrated with existing platforms.
Background
The Credit Asset Management arm of a leading German investment bank employed the services of Figment Software to design and implement a Loan Pricing Service in accordance with current Fair Value Accounting regulations.
In 2006 the Financial Accounting Standard Board (FASB) issued statement Financial Accounting Standard No. 157 – Fair Value Measurements. FAS 157 provides guidance on how to perform Fair Value estimations of assets and liabilities for financial reporting purposes. The Fair Value of an asset is defined as the price achievable in an orderly transaction between two willing participants. The estimated value of an asset or liability must however be based on real-world market data, rather than inconsistent proprietary financial models.
In 2009 the scope of Fair Value Accounting will be extended further to non-financial assets and liabilities.
Brief
Figment were requested to design and implement a Fair Value pricing component that would calculate the price of Loans using a Binomial Option Pricing Model (BOPM). Using an Option pricing model to price Loan products would give the bank a significant competitive advantage within the arena of Credit Asset Management.
The brief specified that the solution must integrate seamlessly with existing infrastructure, and must also be:
- Accessible from Java and other platforms.
- Robust and self-contained.
- Fast, efficient and reliable.
- Correct and testable, with support for future enhancements in a verifiable manner.
Project
Figment worked in synergy with members of the German-based analytic and quant teams to ensure the delivery of a correct solution that met both functional and non-functional (e.g. performance) requirements. Using an iterative (Agile) development methodology, Figment delivered functionality early in the project lifecycle, permitting early verification by the end user.
Solution
Over a period of 9 months, Figment delivered a cutting-edge Pricing Service consisting of the following:
- An extensible Loan Pricing Service (SOA), accessible as a web service and Java RMI service, including an internet-based console.
- A regression test suite to verify correctness of pricing logic at any given time.
- Integration between existing Mark-to-Market batch processes and the Pricing Service.
- A Microsoft Excel add-in to access the Pricing Service.
Benefits
The delivered solution provided a number of benefits, including:
- Advanced and competitive Loan pricing methodology.
- More accurate Fair Value Loan valuation based on real-world market data.
- Event-driven and on-demand re-pricing.
- Product-specific pricing methodologies and flexibility for future product support.
- Pricing functionality accessible by any web-service enabled technology across the bank, permitting easy integration into Front Office tools (e.g. Microsoft Excel).
- Extensible and future-proof architecture supporting multiple pricing models as plugins.
Technologies
- Loan Pricing Service exposed as a Java RMI service and language agnostic web service. Technologies: C++, Java SE 6 (incl. JAXWS, JAXB), XML, XStream, JUnit, Spring.
- Internet-accessible web console to Pricing Service. Choice of technologies dictated by existing infrastructure. Technologies: Java SE, Java EE (JSP, Servlets).
- Pricing logic regression test system and extensible framework. Technologies: Java SE, XML, XStream, JUnit.
- Integration of Pricing Service with existing Mark-to-Market batch. Choice of technologies dictated by existing infrastructure. Technologies: Java SE, Java EE (EJB3), Spring.
- Integration of Pricing Service with Microsoft Excel, accessing Pricing Service as a web service. Technologies: Microsoft Visual Studio Tools for Office (VSTO), C#, .NET 3.5.
