A carregar...
Publicação
Metodologia de testes de stress ao setor segurador em base Top-down
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 9.98 MB | Adobe PDF |
Autores
Orientador(es)
Resumo(s)
O presente trabalho surge no âmbito do estágio curricular na ASF com o objetivo
de promover, numa perspetiva de estabilidade financeira, a avaliação da resiliência e a
mensuração do impacto no setor segurador português face a potenciais cenários adversos.
No contexto europeu, existem metodologias estabelecidas em abordagem de Bottom-up
porém a realização desses exercícios é frequentemente um processo moroso e dependente
de cooperação por parte dos operadores do setor. Existe, portanto, uma necessidade de
ferramentas alternativas e mais ágeis que superem essas limitações.
Este problema reveste-se de interesse e complexidade por combinar aspetos de Solvên-
cia II, técnicas atuarias, de modelação estatística e computacional. Por um lado, a avaliação
de riscos exige coerência com os pilares de Solvência II e robustez suficiente para apoiar
decisões de supervisão prudencial; por outro, é necessário lidar com as limitações práticas,
como a indisponibilidade de dados detalhados ou a necessidade de resposta rápida a
cenários de mercado em constante evolução.
A solução proposta consiste no desenvolvimento de uma ferramenta modular de
Stress Test seguindo uma abordagem Top-down, capaz de aplicar choques aos ativos e aos
passivos e reconstruir balanços por entidade a partir de dados de reporte regular, estes
denominados por Quantitative Reporting Templates (QRT).
A arquitetura da ferramenta inclui funcionalidades de análise determinística e de
futura implementação de uma infraestrutura para simulações de Monte Carlo, permitindo
quantificar incertezas associadas aos choques e métricas de risco como Value at Risk e Tail
Value at Risk.
Os resultados obtidos, decorrentes do teste à ferramenta com base num cenário híbrido
inspirado no exercício de Stress Test da EIOPA de 2024 — nomeadamente o cenário de
"Reintensificação ou prolongamento das tensões geopolíticas" — evidenciam que, à data de
referência de 31 de dezembro de 2023, o setor segurador português apresenta um excesso
de ativos sobre passivos positivo, sugerindo resiliência face ao cenário adverso.
The present work arises in the context of the curricular internship at ASF with the objective of promoting, from a financial stability perspective, the assessment of resilience and the measurement of impact on the Portuguese insurance sector under potential adverse scenarios. In the European context, established Bottom-up methodologies already exist; however, the execution of such exercises is often a lengthy process and dependent on cooperation from market operators. There is, therefore, a need for alternative and more agile tools that can overcome these limitations. This problem is of particular interest and complexity as it combines aspects of Solvency II with actuarial, statistical, and computational modeling techniques. On one hand, risk assessment requires consistency with the pillars of Solvency II and sufficient robustness to support prudential supervisory decisions; on the other hand, it is necessary to address practical limitations, such as the unavailability of granular data or the need for timely responses to rapidly evolving market scenarios. The proposed solution consists of the development of a modular Stress Test tool following a Top-down approach, capable of applying shocks to both assets and liabilities and reconstructing balance sheets at the entity level from regular reporting data, namely the Quantitative Reporting Templates (QRT). The tool’s architecture includes functionalities for deterministic analysis and the future implementation of a Monte Carlo simulation infrastructure, enabling the quantification of uncertainties associated with shocks as well as risk measures such as Value at Risk (VaR) and Tail Value at Risk (TVaR). The results obtained — stemming from testing the tool under a hybrid scenario inspired by EIOPA’s 2024 Stress Test exercise, namely the scenario of “Re-intensification or prolongation of geopolitical tensions” — show that, as of the reference date of December 31, 2023, the Portuguese insurance sector presents a positive excess of assets over liabilities, suggesting resilience in the face of the adverse scenario.
The present work arises in the context of the curricular internship at ASF with the objective of promoting, from a financial stability perspective, the assessment of resilience and the measurement of impact on the Portuguese insurance sector under potential adverse scenarios. In the European context, established Bottom-up methodologies already exist; however, the execution of such exercises is often a lengthy process and dependent on cooperation from market operators. There is, therefore, a need for alternative and more agile tools that can overcome these limitations. This problem is of particular interest and complexity as it combines aspects of Solvency II with actuarial, statistical, and computational modeling techniques. On one hand, risk assessment requires consistency with the pillars of Solvency II and sufficient robustness to support prudential supervisory decisions; on the other hand, it is necessary to address practical limitations, such as the unavailability of granular data or the need for timely responses to rapidly evolving market scenarios. The proposed solution consists of the development of a modular Stress Test tool following a Top-down approach, capable of applying shocks to both assets and liabilities and reconstructing balance sheets at the entity level from regular reporting data, namely the Quantitative Reporting Templates (QRT). The tool’s architecture includes functionalities for deterministic analysis and the future implementation of a Monte Carlo simulation infrastructure, enabling the quantification of uncertainties associated with shocks as well as risk measures such as Value at Risk (VaR) and Tail Value at Risk (TVaR). The results obtained — stemming from testing the tool under a hybrid scenario inspired by EIOPA’s 2024 Stress Test exercise, namely the scenario of “Re-intensification or prolongation of geopolitical tensions” — show that, as of the reference date of December 31, 2023, the Portuguese insurance sector presents a positive excess of assets over liabilities, suggesting resilience in the face of the adverse scenario.
Descrição
Palavras-chave
Stress Test Solvência II Top-down Provisões técnicas Simulação de Monte Carlo Supervisão prudencial