Skip to content

Content Library

Back to Audio

Audio May 5, 2026

Why IRRBB Models Are Failing and What Banks Must Fix Before Regulators Do It for Them

Monzo’s ALM chief Svetlana Kardan exposes why many banks’ IRRBB, NMD, and structural hedge models are outdated and vulnerable. She explains how non linear deposit behaviour, geopolitical shocks, and two way rate risk demand new modelling, faster governance, and real time ALM agility.

display-sc-embed-code-here

Svetlana Kardan delivers a provocative warning: most banks’ IRRBB and NMD models would not withstand regulatory scrutiny if submitted today. She argues that the industry’s framing of a “post‑hiking cycle” is dangerously outdated. The SONIA curve’s violent bear‑steepening, driven by geopolitical shocks rather than domestic fundamentals, exposes the fragility of assumptions embedded in structural hedges, deposit models, and ALCO playbooks. Parallel shocks no longer capture sequencing risk; term‑premium repricing is now structural; and the divergence between UK and European NIMs is driven not by market structure but by hedging strategy and deposit pass‑through behaviour.

Ultimately, Kardan argues that ALM must shift from static compliance to dynamic balance‑sheet strategy. Banks need recalibrated models, non‑parallel shock frameworks, governance that can respond intra‑quarter, and a recognition that CSRBB, liquidity risk, and credit‑spread volatility are now first‑order risks. In a world where rates can swing 100 bps in three months without a single move in the overnight rate, the institutions that survive will be those that treat ALM as strategic capital allocation — not a reporting function buried in the corner.

Connect Plus

Unlock more with Connect Plus membership

This content is available to Connect Plus members.