Modeling lapse rates using economic variables. A case study for a life insurance company operating in Kenya

Abstract

Lapse rates are an important consideration in the pricing and implementation of Life Insurance Policies. Lapses are defined the premature withdrawal of policyholders from policies that they have previously taken up. The occurrence of a lapse is typified by the cessation of premium payments by some policyholder and in some cases a benefit payment may be made. It is therefore important to estimate lapse probabilities in the pricing of products since these have a direct impact on the duration of premium payments. Additionally it is important for the policy provider to have some idea the sort of lapse rates to expect where policies have been taken up in order to carry out reasonable Asset Liability Management. This project reviews past research that has been carried out on the estimation of future lapse rates and the factors that affect them. An attempt is also made to estimate lapses based on economic variables for a Life Insurance Company operating in Kenya. Amongst the key factors in the estimation of lapses is the type of insurance policy! The model whose construction is attempted draws largely from Changki Kim (2005) 'Modeling lapses using economic variables' where the monthly lapse rate is constructed as a response variable in a generalized linear model. The predictor variables used are Market rates, Unemployment rates, Economic growth rates, and Financial Crises.