November 27, 2012 - SWIFT, the financial messaging provider for more than 10,000 banking organisations, securities institutions and corporate customers in 212 countries and territories, has released a white paper on the operational challenge of sanctions compliance.

Financial institutions are increasingly seen as the front line in the fight against money laundering and terrorist financing. As a result, regulators worldwide have intensified their focus on banks' sanctions compliance processes.

This paper demonstrates the scale of the operational challenge faced by the industry. SWIFT data reveals the growth in number of transactions worldwide. By correlating that with the rising number of watchlists, we estimate that the cost of sanctions compliance at financial institutions will double every four years. With banks under pressure to reduce costs as well as to manage their exposure to risk, having efficient systems and processes in place has never been more crucial.

More importantly, the paper outlines the steps banks can take to improve their sanctions compliance process and outlines the three most important characteristics of a successful compliance infrastructure: the ability to regularly test system performance, progressively improve system effectiveness and to encode group compliance policy as part of tests.

Nicolas Stuckens, Manager, AML & Sanctions Initiatives, SWIFT, said: "Compliance with international sanctions is becoming increasingly complex. Sanctions lists evolve daily and the number of transactions that need to be screened is rising rapidly. Banks are expected to keep up and tune their sanctions filter in line with their risk appetite, so it has never been more important that systems and processes are effective and efficient. The aim of this paper is to outline the challenges faced by banks but also to give them an idea of the framework for a successful sanctions compliance infrastructure."

Sanctions lists themselves are not particularly large, amounting to about 40,000 distinct names and synonyms. However, when you apply possible fuzzy matching to searches (to account for potential misspellings, phonetic similarities etc) this generates an equivalent of 4,000,000,000,000 possible names that filters must screen.