美國壽險保單解約率之研究：門檻共整合模型之應用

Voluntary Termination of Life Insurance Policies in the U.S: Threshold Cointegration Approach

傳統人壽保險因具有現金價值之累積，保戶可於契約終止前申請解約，取回保單價值準備金或解約金。然保戶的解約行為可能影響保險公司的現金流量，若保險公司無法明確預估解約率，可能對清償能力造成負面影響。過去文獻對於保戶的解約行為理論有二：急用資金假說與利率假說。自2008年金融海嘯後，傳統模型對於利率環境的改變與解約率無法提出適當解釋。本研究延續過去文獻之內容，考量金融海嘯之後的利率環境，針對壽險解約率的假說，採用非對稱計量模型，探討在利率環境變動下的解約理論。結果顯示，利率與失業率對解約率均有長期影響，同時支持急用資金假說與利率假說。因此，壽險公司在評估未來的解約率與現金流量時，需多方面考量，審慎為之。

Life insurance policyholders usually are granted the option to terminate their policies before they mature. The action is called voluntary termination, or surrender. Since surrender options are exercised at the discretion of the policyholder, these actions usually cannot be fully anticipated by insurance companies in advance. Not able to access correctly may increase the company's insolvency risk. There have been two different hypotheses in the behavior of life insurance policy surrenders: the Emergency Fund Hypothesis and the Interest Rate Hypothesis. During the financial crisis, interest rate dropped dramatically, while there is less change in the voluntary termination rates of life insurance in the U.S. This study extends the data period of previous literature to 2012 and finds out that both the interest rate and unemployment rate have long term impact on policyholders' voluntary termination, implying that both the Emergency Fund Hypothesis and the Interest Rate Hypothesis may provide some explanation for life insurance policy voluntary termination. Furthermore, our TAR model found that the voluntary termination rate responds to a positive discrepancy, such that, when the actual change in the voluntary termination rate is higher than expected, we should observe adjustment toward the opposite direction in the future.