Price Changes and Customer Loyalty & Defection in Consumer Insurance: A Test of ‘Prospect Theory’ - And Why Regression Modeling Should Use Multiple Datasets

20 Pages Posted: 4 Apr 2012

See all articles by John Dawes

John Dawes

University of South Australia - Ehrenberg-Bass Institute

Date Written: April 4, 2012

Abstract

This paper examines the relationship between price changes and customer defection levels in a ‘subscription’ type market, namely car insurance. Two regression models are constructed to estimate this relationship, one model for younger customers and another for older customers. The regression models fit well to the defection rates associated with different levels of price changes. The analysis also shows that the impact of price decreases on defection rates is less than the impact of price increases, consistent with ‘loss aversion’. The paper notes that models of this type should offer true predictive ability and therefore tests the ability of the model to predict defection rates for new data. That is, rather than fitting new models to the new data, the analysis fits the original models to new data. The models performed comparatively poorly when fit to new data, particularly for price increases. The paper concludes that multiple sets of data are needed to develop and validate predictive models.

Keywords: Customer loyalty, customer defection, pricing, insurance, loss aversion

JEL Classification: M31

Suggested Citation

Dawes, John, Price Changes and Customer Loyalty & Defection in Consumer Insurance: A Test of ‘Prospect Theory’ - And Why Regression Modeling Should Use Multiple Datasets (April 4, 2012). Available at SSRN: https://ssrn.com/abstract=2034074 or http://dx.doi.org/10.2139/ssrn.2034074

John Dawes (Contact Author)

University of South Australia - Ehrenberg-Bass Institute ( email )

GPO Box 2471
Adelaide, 5001
Australia

HOME PAGE: http://www.johndawes.info

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
226
Abstract Views
1,378
rank
149,800
PlumX Metrics