Category: Prudential Supervision

Targeted Financial Conditions Indices (TFCI): why average financial conditions miss what matters most

Miguel Herculano, Santiago Montoya-Blandon and Jorge Pinheiro

Financial conditions indices are widely used by central banks and policymakers, including in the UK, to summarise and monitor in real-time the state of financial conditions in an economy. But most indices focus on what happens to average conditions, rather than the risks policymakers are often most concerned about – such as sharp downturns in economic activity or surges in inflation. We develop novel targeted financial conditions indices (TFCIs), using US data, that focus directly on measuring and forecasting these risks. Instead of asking which financial variables move together on average, the approach identifies which ones matter for specific outcomes. We show that different risks are linked to different financial factors – and that focusing on these can improve the forecasting performance of these key macroeconomic targets.

Continue reading “Targeted Financial Conditions Indices (TFCI): why average financial conditions miss what matters most” →

Leveraging language models for prudential supervision

Adam Muhtar and Dragos Gorduza

Imagine a world where machines can assist humans in navigating across complex financial rules. What was once far-fetched is rapidly becoming reality, particularly with the emergence of a class of deep learning models based on the Transformer architecture (Vaswani et al (2017)), representing a whole new paradigm to language modelling in recent times. These models form the bedrock of revolutionary technologies like large language models (LLMs), opening up new ways for regulators, such as the Bank of England, to analyse text data for prudential supervision and regulation.

Continue reading “Leveraging language models for prudential supervision” →