Artificial intelligence meets business intelligence: the anatomy of inflation

Marcus Buckmann, Galina Potjagailo and Shobhit Rawat

What can artificial intelligence reveal about firms’ pricing decisions from text-based business intelligence? We use a large language model (LLM) to quantify pricing narratives from reports documenting discussions during the Bank of England’s company visits. The LLM is prompted to derive scores for output-price changes and decompose them into determinants, including direct and indirect input costs, labour costs and demand. The resulting measures recover economically meaningful signals from firms’ narratives over time, from weak demand following the global financial crisis (GFC) to broad cost pressures and subsequent effects through wage and price setting during the recent inflation surge. The latest narratives point to marked cost increases following Middle East conflict, but so far less broad and persistent than in 2022.

Continue reading →

Narratives under the microscope: what MPC transcripts can tell us about policy thinking

Dragos Gorduza, Livia Silva Paranhos and Ivan Yotzov

Monetary policy decisions reflect incoming data, forecasts, and policymaker preferences. In this post, we use large language models (LLMs) to analyse Bank of England Monetary Policy Committee (MPC) transcripts from 2015 to 2017. We extract structured measures of individual views on inflation and growth, and the narratives justifying policy decisions. This gives a new perspective on UK monetary policy makers’ reasoning, using recently published transcripts. We show that concerns about economic growth relative to its potential weakened and later strengthened over the sample, while concerns about above-target inflation increased. We link these measures to voting behaviour and show that, over the sample period, both growth and inflation measures predict votes for higher rates, but only growth measures explain votes to cut Bank Rate.

Continue reading →

Tracking property completions in near real time with CHAPS purpose codes

Simran Mehta and Valentina Macchiati

UK property transactions can be identified and observed directly within minutes of payment settlement using CHAPS, the UK’s high-value payment system. Taking advantage of the migration to the ISO 20022 payment standard in June 2023, and the introduction of mandatory purpose codes from May 2025, we have developed a real-time indicator of housing activity. We find that this closely aligns with existing official housing market statistics and therefore can act as a reliable real-time indicator of these statistics, superseding previously developed value-based proxies.

Continue reading →

When the financial system becomes searchable

Andreas Viljoen

Financial crises rarely begin with one self-contained weakness. They emerge when vulnerabilities connect: leverage meets a margin call; a margin call meets an illiquid market; falling prices meet common collateral; and a funding concern becomes a run. Before the event, each link may sit in a different spreadsheet, institution or jurisdiction. Afterwards, the route through them can look obvious. This post explores a possibility raised by advances in artificial intelligence (AI): that the financial system could become searchable, making more of those routes visible beforehand. It outlines two specific scenarios and their implications for financial authorities. First, system-wide testing by authorities should learn to search the way agents will. And consequential agent decisions and actions should be observable in operation, unlike today.

Continue reading →

When the floor rises: how Basel output floor could tilt bank lending

Marzio Bassanin

The 2017 finalisation package of the Basel III reforms to bank capital regulation aims to make capital requirements more robust and consistent across banks. One of its most significant changes is the introduction of the ‘output floor’, which limits how far capital requirements calculated using internal models can fall below those based on standardised approaches. But the output floor is not just about capital levels. We find that it may tilt lending towards corporate loans and some riskier mortgages, and away from the safest mortgages. And if banks’ responses to past reforms are any guide, banks won’t wait until its full implementation in 2030 to start reacting.

Continue reading →

A new framework for monitoring risks in the UK housing market

Tihana Škrinjarić

In my recent paper, I present a new model that helps assess risks in the UK housing market. Unlike traditional approaches that focus on average house price growth, the model estimates a full range of possible outcomes, allowing policymakers to identify potential risk of big house price drops. The analysis also highlights important regional differences: areas with more constrained housing supply tend to be more sensitive to changes in interest rates. Expanding supply can help ease price pressures. These insights can help improve the monitoring of housing market vulnerabilities and support financial stability policy.

Continue reading →

Distributional consequences of borrower-based macroprudential tools

Jagdish Tripathy, Arzu Uluc, José-Luis Peydró and Francesc Rodriguez-Tous

Borrower-based macroprudential measures – such as limits on loan to income (LTI) and loan to value (LTV) ratios – have become a standard feature of the post-crisis regulatory landscape. A growing body of country-specific evidence suggests these measures are effective in moderating the self-reinforcing loop between mortgage credit and house prices, and in reducing default rates and limiting house price volatility during periods of economic stress. Yet their distributional consequences are less well understood. In a new paper, we survey the existing evidence and find that these tools deliver clear financial stability benefits, while also generating distributional effects across borrower groups. Further, we identify areas where future research is needed to provide a comprehensive welfare assessment of these measures.

Continue reading →

Monetary policy transmission: it’s all in the curve

Sofia Carollo and Natalie Burr

While monetary policy sets short-term policy rates, households and firms borrow at varying time horizons. How a policy decision reshapes the whole yield curve therefore matters. We trace the reactions of yields in narrow windows around UK monetary policy announcements across two dimensions: a ‘level’ surprise that shifts the entire curve and a ‘slope’ surprise that changes its steepness. We find that a level surprise affects CPI inflation more than a slope surprise does; this result is difficult to recover from surprises that conflate the two dimensions. So, the policy rate tells only part of the story: two curves considered equivalent from a stance perspective can lead to different inflation outcomes. Policymakers must be attuned to these differing effects.

Continue reading →

Has UK food inflation been under the weather?

India Rimmer, Hannah Copeland and Boromeus Wanengkirtyo

Global extreme weather events may feel far away, but they leave behind a trail of higher prices in our shopping baskets. As outlined in past Monetary Policy Reports, droughts, flooding and heatwaves occurring overseas often impact UK food inflation, which averaged 4.2% in 2025. But how much of the rise in food inflation last year can we blame on the weather? By constructing a new proxy for global weather shocks, we find that they increase UK food prices with a peak impact after one year. In the latest period, our model suggests that weather shocks contributed 0.8 percentage points to food inflation at peak in May 2025. Weather continues to matter for inflation amidst the current El Niño phenomenon.

Continue reading →

Is UK productivity growth low? A historical and cross-country perspective

Sophie Piton and Fabrizio Cadamagnani

A lot has been written about UK productivity and how weak it’s been in recent years. This post assesses UK productivity trends in a historical and cross-country perspective. Productivity growth has been weak across G7 economies over the past two decades, reflecting the end of the information and communications technology (ICT) revolution and the flattening gains from globalisation. The slowdown was particularly large in the UK, mainly because it experienced a larger decline in the share of manufacturing than peers and then because of the impact of Brexit. In recent years, US productivity growth has been accelerating thanks to tech, offering some optimism for the future of UK productivity.

Continue reading →