The opinions expressed here are those of the authors. They do not necessarily reflect the views or positions of UK Finance or its members.

This blog is a collaboration between Nikolas Kontogiannis, Polly Wong and Paolo Vareschi from Zanders.

For risk managers tasked with quantifying the impact of geopolitical risk, these headlines trigger a major analytical challenge. Geopolitical events are discrete and inherently unpredictable. Translating them into quantitative risk measures requires clear conceptual framing and methodological choices. 

Against this backdrop, we share practical insights on measuring and modelling geopolitical risk, drawing on a growing body of empirical work.

I would assess its impact if I knew what it is

Quantifying any risk driver starts with defining it. Unlike standard macroeconomic measures (e.g., GDP), geopolitical risk has no universally accepted construction method.

A widely cited definition from Caldara and Iacoviello (2022), referenced in public speeches by the Bank of England and the European Central Bank, defines geopolitical risk as:

“…the threat, realization, and escalation of adverse events associated with wars, terrorism, and any tensions among states and political actors that affect the peaceful course of international relations.”

This definition is useful precisely because it highlights the concept’s breadth. “Geopolitics” spans a wide range of events. That heterogeneity is not just semantic; it directly shapes how geopolitical risk should be measured.

(An inventory of) Geopolitical risk indicators

Recent institutional work has advanced the field precisely on this point. In January 2026, the European Central Bank and the European Systemic Risk Board published a joint report documenting a comprehensive inventory of geopolitical risk and uncertainty indicators, one of the most systematic attempts to catalogue available geopolitical risk measures within a financial risk analysis framework. 

The inventory spans a wide range of indicators, from news-based measures such as the Caldara–Iacoviello Geopolitical Risk (GPR) index to indicators of trade policy uncertainty and supply chain pressure, grouped across thematic dimensions. 

The practical implication is clear: rather than relying on a single catch‑all proxy, risk managers can explore targeted indicators that better capture the shock type of interest. 

A pragmatic modelling workflow

Historical analysis should guide their choice. Examining the behaviour of indicators around past episodes can reveal which measures respond most strongly, with what timing, and how persistent the signal is. Beyond thematic relevance, indicator selection also requires assessing technical time‑series properties that affect empirical performance. From there, a pragmatic modelling framework emerges.

A selected indicator can be incorporated as a shock proxy in time-series models used to estimate impulse response functions (IRFs), tracing the dynamic impact of shocks on macro-financial variables.

Two empirical approaches stand out: structural VARs (SVARs) and Local Projections (LPs). LPs are often more robust to model misspecification and can provide more reliable inference for impulse responses; an appealing feature for geopolitical risk analysis, where structural uncertainty is substantial.

From modelling to risk management

Within this framework, estimated IRFs provide empirically grounded, conditional response paths that describe how the macro-financial system has typically behaved following a well-defined geopolitical shock. 

This makes the outputs directly useful for stress testing and scenario design. IRFs can serve as empirically disciplined anchors for scenario severity calibration, overlays, post-model adjustments, and cross-variable consistency checks.

A final, critical step is portfolio tailoring. The same event can play out very differently depending on where exposures sit, how supply chains are configured, and what businesses do. Some will be directly hit; others will be hit only indirectly. Converting macro-level responses into risk insights requires mapping transmission channels to actual exposures and distinguishing direct impacts from second‑order effects.