The Shape of Things to Come: Probabilistic Emissions Forecasting for Climate Risk

Most transition risk analysis assumes that corporate emissions remain unchanged into the future.

It is simple. It is clean. It is also systematically distorting.

Emissions rarely remain stationary. Some companies have been reducing emissions for years. Others have been steadily increasing them. Extending flat emissions into the future ignores observable behaviour that contains predictive information. The result is a structural bias in forward-looking climate risk metrics: exposure is overstated for companies already trending downward and understated for those on an upward path.

For financial institutions assessing portfolio-level exposure, that distortion matters.

This research introduces a Monte-Carlo Momentum (MCM) framework that replaces static baselines with probabilistic emissions trajectories grounded in recent corporate behaviour. Using annual Scope-level emissions data from 2020 onwards, and requiring at least three valid reported years per entity-scope, the model estimates both momentum and historical volatility. Each entity-scope is then simulated with 1,000 Monte Carlo paths, generating a distribution of plausible futures rather than a single deterministic projection.

The result is a forward-looking, uncertainty-aware baseline that reflects how emissions have actually behaved.

Forecasts are filtered to exclude implausible outcomes and highly uncertain cases. Where emissions histories are too volatile or too thinly disclosed to support meaningful projection, forecasting is avoided rather than presented with misleading precision. Where behaviour is stable, the model produces a median trajectory and an explicit uncertainty range that can be carried through to scenario analysis.

These probabilistic emissions forecasts feed directly into Emmi’s scenario-based transition risk framework. Instead of applying climate pathways to a flat line, risk metrics are recalculated against a dynamic baseline. Transition Value at Risk (TVaR), temperature alignment, and emissions reduction requirements become sensitive to observable business dynamics. A company with a persistent downward trend may exhibit lower projected liability relative to a static assumption. An upward trajectory may imply higher exposure and more urgent transition requirements.

Comparing the dynamic baseline with a constant-emissions baseline makes the distortion visible. It shows how much of a company’s future risk is driven by its recent emissions behaviour, and where static assumptions misstate portfolio exposure.

This is not a prediction in the strict sense, nor a substitute for explicit decarbonisation commitments. It is a behavioural baseline: a view of what the future might look like if recent patterns continue. When combined with climate scenarios and Science-Based Targets, it provides a clearer lens on whether companies are drifting toward or away from alignment, and where transition risk may be concentrated.

Corporate emissions change over time. Forward-looking climate analysis needs a baseline that recognises this.

Download the full white paper to examine the methodology, validation and portfolio-level implications in detail.