The UK's central bank has recently started to explore the use of social media chatter as a more timely indicator of mass market sentiment, establishing a Big Data lab staffed by a new analytics team to construct algorithms and models for extracting and analysing data gleaned from unconventional online sources.

In the lead up to the Scottish independence referendum, the team was tasked with building an experimental system to collect and analyse tweets in real-time with the primary question: Could Twitter help predict a bank run?

At the time, concerns were raised about the potential impact a ‘yes’ result might have on the UK financial system. Indeed, the Governor highlighted that “uncertainty about the currency arrangements could raise financial stability issues.”

Using social media as a unique lens through which to observe developments in real-time, the team searched for tweets including terms or phrases that may have indicated depositors preparing to withdraw money from Scottish financial institutions. The aim was to monitor the volume of traffic matching these search terms, and provide an early warning system in the event of a spike.

The team was alerted to a sudden spike in activity on 15th September as detailed in the chart below, when the conjunction of the terms RBS and 'run' started trending.

"On closer examination, it transpired that we were looking at a series of tweets and retweets involving the Minnesota Vikings," explains the team in a post published on the BoE's unofficial Bank Underground staff blog. "This had been captured because they combined the term “run” and the abbreviation “RBs”. But in this context, the reference was to Running Backs and not the Royal Bank of Scotland! To avoid this pitfall, the search terms were subtly changed to avoid that particular pattern of upper and lower case characters."

In the end, Scotland voted to stay in the Union and the nightmare scenario feared by the central bank failed to materialise. Nonetheless, the exercise was viewed as valuable to the Bank in building capabilities and knowledge to serve as a foundation for future projects.

"In particular, the project presented a number of IT challenges," explain the authors. "Our search terms needed to be securely transmitted to the data provider; if it was revealed that we were interested in particular firms this could cause the very run we were concerned about. We also needed to stream, store and query these data in real-time, and achieved this using novel technologies for the Bank, opening up new avenues for analysis and research."