Models are indispensable tools in insurance risk management, enabling firms to quantify exposures, inform strategic decisions and meet regulatory expectations. Yet, models by their nature are simplifications of reality and inevitably contain limitations. Some intentional to maintain practicality, others unintentional and requiring mitigation. These limitations are often addressed through mechanisms commonly referred to as “overlays”, “manuals” or “out-of-model-adjustments”.
However, the broad use of these terms can lead to ambiguity in distinguishing between practical implementation choices and genuine model risk concerns. For example, some adjustments merely reflect the implementation of a valid method in an alternative system, while others represent corrective interventions to address deficiencies in the model.
To facilitate greater transparency and rigour, the PRA has introduced Model Limitation Adjustments (‘MLA’) within the Solvency UK framework for Internal Model firms. This article examines what should be identified as an MLA and how it fits into the wider Internal Model Framework.
