人壽保險公司與退休基金之違約風險、負債評價與資產配置

Abstract

人壽保險公司與退休基金之違約風險、負債評價與資產配置 基於保險契約及退休金計畫隱含不同存續期間之結構選擇權，成為近期財務評價上 的重要議題，本研究因此探討人壽保險公司與退休基金之違約風險、負債評價與資產配 置之間關係，如何有效進行負債評價及制定監管機制以達到保險市場(與退休基金)財務 穩健之目的。就保險人而言，不僅涉及現有營業之保險人，同時對於即將進入保險市場 之潛在競爭者而言，都是相當重要之議題。主管機關於保險公司面臨財務清償問題時， 循序採取行政輔導、監管與接管機制，監督保險人對於被保險人保證給付義務之履行， 已成為現階段最重要之保險監理議題。 本文延續Grosen 與Jogensen(2002)及Chen 與Suchanecki(2007)之研究架構，研究保 險人於市場狀態改變(regime switch)於公司財務槓桿、政府監理寬容措施與保證給付義務 之相互關係過程， 即市場狀態改變可表示為半平賭表示式(semimartingale representation)之連續時間同質馬可夫過程(見Monter, 2007, 2008)財務槓桿、政府監理 寬容措施與保證給付義務之相互關係，並分別就現行保險法149 條之架構與美國破產保 護法第7 章與第11 章之接管程序，分別引用連續即離散模式之巴黎式選擇權模型計算 保證給付義務，分析監理寬容對於保險人負債價值之影響。保險業退場機制之啟動標準 直接影響被保險人之保證給付，本研究建立財務模型評估監理寬容政策之效益，並呈現 不同監理介入模式(相對監控模式與絕對監控模式)對於保證給付義務之影響，本研究運 用Laplace 轉換與數值反轉換(numerical inversion)方法，精確評估於風險投資行改變下公 司之保證給付。 並後續延伸至保險基金及退休基金評價及資產配置，利用隨機擴散方程式描述資產 與負債之連續動態價值，以多期基金規劃的觀點，依保險機構及退休基金之背景風險， 尋求最適動態資產配置策略降低違約風險，預期依研究結果，擬定較具效益之管理決策。

Solvency risk, liability valuation and asset allocation for life insurance company and pension fund Abstract This study investigates solvency risk, liability valuation and asset allocation for life insurance company and pension fund. Addressing the solvency and capital standards is critical not only for the existing competitors, but also for potential new entrants to life insurance market, enforcing them to be competent based on a prudent regulation framework. How to reduce the bankruptcy cost through early regulatory intervention becomes an important issue in public policy making. Following the framework in Grosen and Jögensen (2002), Chen and Suchanecki (2007) study the bankruptcy cost through the embedded Parisian barrier option based on the US Code with Chapter 7 and Chapter 11 bankruptcy procedures. They have explicitly shown the bankruptcy costs under default and liquidation events. In reality, the standards in triggering the government intervention to control the bankruptcy cost become a practical concern. In this study, relevant and absolute intervention criterions are employed to determine a proper benchmark when the regulatory forbearance is shown. This research is structured as follows：Stage 1 the insurance and pension market and dynamic model review. Stage 2, we discuss supervision intervention criteria: the relative deficiency criterion and absolute deficiency criterion for life insurance. The same issues for pension management will be further developed to incorporate the regime switch, i.e., homogeneous continuous‐time Markov chain admits semimartingale representation, consideration in modeling the insurer’s investment behaviors. Stage 4, we give some numerical results. In the last stage (Stage 5) concludes the findings from this study. There are many reasons to consider regime‐switching models instead of a geometric Brownian motion with constant parameters: (Monter, 2007, 2008) (1) It is well known that the drift and volatility of the assets are not constant over time; (2) when to evaluate investments for long term periods, it is crucial to take into account the macroeconomic structural changes that may affect the firm’s value; (3) empirical evidence shows that financial markets exhibit volatility clustering; (4) regime‐switching models are easier to understand and interpret than stochastic volatility models.