Survival Analysis of a Household Portfolio of Insurance Policies: How Much Time Do You Have to Stop Total Customer Defection?

Abstract

Customer-side influences on insurance have been relatively ignored in the literature. Using the household as the unit of analysis, this article focuses on the behavior of households having multiple policies of different types with the same insurance company, and who cancel their first policy. How long after the household's cancellation of the first policy does the insurer have to retain the customer and avoid customer defection on all policies to the competition? And, what customer characteristics are associated with customer loyalty? Using logistic regression and survival analysis techniques, an assessment is made of the probability of total customer withdrawal, and the length of time between first cancellation and subsequent customer withdrawal. Using a European database spanning 54 months of household multiple policyholder behavior, the results show that cancellation of one policy is a very strong indicator that other household policies will be canceled. Further, the insurer can have time to react to retain the customer after the first cancellation, however, this time is significantly dependent on the method used to contact the company, household demographics, and the nature of the household's insurance policy portfolio. Surprisingly, core customers having three or more policies in addition to the canceled policy are more vulnerable to total defection on all policies than noncore customers. Further, the potential customer repelling effects of premium increases seem to wear out after 12 months. Strategic implications of the results are presented. Copyright (c) The Journal of Risk and Insurance, 2008.

The Need to Monitor Customer Loyalty and Business Risk in the European Insurance Industry

“…15 Kaas et al (2004). 16 Guillen et al (2003Guillen et al ( , 2007; Brockett et al (2008); Perez-Marin (2006).…”

“…We believe that customer loyalty and business risk monitoring play a fundamental role in providing information about the changes in the quality of the portfolio, and 14 Brockett et al (2008). 15 Kaas et al (2004).…”

Multivariate Frequency-Severity Regression Models in Insurance

“…For example, a study of term and whole life insurance ownership is in [6]. As an example in customer retention, both [7,8] advocate for putting the customer at the center of the analysis, meaning that we need to think about the several products that a customer owns simultaneously.…”

Predicting Probability of Customer Churn in Insurance

“…Different researchers argue that retaining a customer instead of getting a new one is a profitable strategy for the company, because the cost of finding a new customer is larger than the cost of keeping an existing one [1][2][3]. In fact, long-term customers tend to be more engaged with the company [4] and take less of the company's time than attracting new customers [5,6].…”