ORSA Scenario Design and Board Use

By Jonas Osman Abdelghafour.

The Own Risk and Solvency Assessment (ORSA) under Solvency II Article 45 requires each insurer to form its own view of the risks it faces, the capital needed to support them over the business planning horizon, and continuous compliance with regulatory requirements. In practice the quality of an ORSA is determined less by the modelling than by the scenarios chosen and how the board uses the results.

What "own view" means in scenario terms

Standard-formula SCR calculations reflect a calibrated 1-in-200 shock across prescribed risk modules. The ORSA asks a different question: given this undertaking, its business mix, its strategy, and the environment it operates in, what combinations of events would put the business plan or solvency at risk, and what actions are available? The scenarios that answer that question are almost never identical to standard-formula shocks.

Designing a proportionate scenario set

A useful ORSA scenario set typically has three layers. The first is the base plan, run through the capital and liquidity models over the full planning horizon. The second is a small number — three to five — of severe-but-plausible scenarios that flex the risks most material to the strategy. The third is one or two reverse stress tests that identify what would have to happen for the business model to fail.

Each scenario should specify the narrative, the calibrated shocks by risk factor, the assumed correlations, the timing over the projection horizon, and — critically — the management actions available and their assumed effectiveness. A scenario without pre-agreed management actions produces a number, not a decision.

Calibration discipline

Severity should be justified explicitly. Common anchors include historical episodes (1994 rates, 2008 credit, 2020 pandemic dispersion), regulatory reference scenarios (EIOPA insurance stress tests, climate reference scenarios from the NGFS), and portfolio-specific tail analytics. The point is not to hit a particular return period but to make the assumed severity defensible and comparable across cycles.

Reverse stress testing

Reverse stress tests start from failure and work backwards. For an insurer, "failure" is usually defined as breach of the SCR, loss of a rating threshold, or inability to write new business. The exercise identifies the combination of events that would cause it and assesses how remote that combination is. The value is not the probability estimate — which is typically uncertain — but the identification of vulnerabilities the ordinary scenario set may miss.

Board use

An ORSA that changes no decisions is a failed ORSA. The board pack should surface three things clearly: the residual capital and liquidity headroom after each scenario with management actions applied; the specific decisions being asked of the board (risk appetite recalibration, reinsurance changes, product limits, capital actions); and the assumptions on which those decisions depend. See the companion note on actuarial communication for risk committees for how to present uncertainty without either overstating or hiding it.

Integrating with risk appetite and planning

Scenario outputs should be tied back to the risk-appetite statement. If a scenario breaches appetite even after management actions, the response is either to change the strategy, change the appetite, or change the mitigations — not to soften the scenario. The ORSA also feeds business planning: capital allocation, product priorities, and reinsurance structure should reflect what the ORSA revealed.

Common failure modes

Three patterns show up repeatedly. Scenarios calibrated to what the model can produce comfortably, rather than to what the business fears. Management actions credited at full effectiveness in every scenario, including the ones where they are least likely to work. And reverse stress tests written as narrative exercises with no quantitative anchoring.

Limitations

An ORSA is a point-in-time exercise informed by imperfect models. It cannot capture every plausible tail event, and its usefulness declines quickly if it is refreshed only annually in a fast-moving environment. Trigger-based interim updates — after material market moves, portfolio changes, or emerging risk events — are increasingly expected by supervisors.

Conclusion

Scenario design is where the ORSA earns or loses its value. Severe, specific, plausible scenarios paired with pre-agreed management actions turn the ORSA into a decision-making instrument. See also market risk stress testing beyond VaR and climate risk data lineage and proxies.

Written by Jonas Osman Abdelghafour, actuary and financial risk manager. Background and contact details are on the about page.