A temporal analysis of behavioural brand loyalty among urban Chinese consumers

Abstract

Abstract Chinese marketing is evolving, and this is having a dramatic influence on the incomes and spending patterns of consumers. Set against these longer-run trends, the contribution of this paper is to examine the year-on-year persistence of brand purchasing and behavioural brand loyalty in packaged goods markets in China. Revealed preference data for two product categories (toothpaste and soy sauce) in two contrasting cities (Shanghai and Xi'an) are analysed, with particular attention being paid to whether brand purchasing and behavioural brand loyalty are invariant over a five-year period. Conceptually, the NBD-Dirichlet model of brand choice and purchase incidence is employed. Analysis shows that patterns found in the Chinese data are quite similar to those reported previously for Western markets, with invariant results over time – notwithstanding supply-side and demand-side changes over the study period. In particular, the double jeopardy effect and multi-brand loyalty are consistently observed in the toothpaste and soy-sauce markets in both cities over the full five-year period. Results distinct to the Chinese market are also identified: exclusive brand loyalty appears to be declining in favour of multi-brand loyalty, and there are between-city differences in the magnitude of this decline. Keywords: brand purchasingbehavioural loyaltydeclining loyaltyurban Chinese consumerstemporal invariance analysislongitudinal studyNBD-Dirichlet model Acknowledgements We are indebted to CVSC-TNS Research Co. (CTR) for supplying consumer panel data, especially to Dr Jiang Tao, Leo Lui, and Peter Lee for their assistance. The project was funded with a grant from the Australian Research Council (ARC Grant DP0452489). Helpful advice from the editor and two anonymous reviewers is acknowledged. Notes 1 GDP growth is predicted to slow from roughly 9% in 2008 to as low as 7% in 2009 following weakening external demand and falling domestic investment. This is a significant downturn, but China's growth rate continues to exceed most other nations, including the US, UK, and Japan. 2 Brand penetration is defined as the proportion of consumers in a market who have bought the brand of interest at least once within a specified time period (such as a week, a quarter, or a year). At the individual level, frequency of purchase is the number of times a consumer has made a purchase of a brand within a specified period. Usually this is expressed as an average purchasing rate across all those who have bought the brand at least once within the specified period. 3 Behavioural brand loyalty is defined as the biased purchasing behaviour towards one or more alternative brands from a set of such brands, over some specified period of time (such as a week, a quarter, or a year). An indicator of behavioural brand loyalty is the proportion of category needs met by the brand of interest, averaged across all buyers of the brand. This is the so-called share of category requirements (SCR) measure. Another indicator is the proportion of buyers of a brand who only buy that brand – these are described as 100% loyal or exclusively loyal. 4 Exceptions include small-scale studies in a Western context by Stern, Rungie, & Laurent, Citation2003 and Habel, Rungie, & Laurent, Citation2005 to see if generalisations and patterns persist over successive years. In a non-Western context, Keng, Uncles, Ehrenberg, & Barnard, Citation1988 conducted a systematic study of the buyer behaviour of Japanese consumers, but they did not consider the persistence of patterns over time. 5 No such studies were reported in reviews of Chinese consumer research conducted by Ouyang, Zhou, and Zhou, Citation2000; Sin and Ho, Citation2001; and Yim, Citation2004. A more recent study by Yang et al. (Citation2005) examined the double-jeopardy phenomenon in the detergent and shampoo markets. These researchers mainly focused on the mediating effect of penetration between advertising and brand loyalty, rather than patterns of buyer behaviour, and they did not undertake any longitudinal (within-year or across-year) analysis. 6 The implications of double jeopardy for the management of loyalty programs have been investigated in several Western nations (e.g. Meyer-Waarden & Benavent's Citation2006 study of French programs). Our supposition is that similar implications would carry across to China, but this is a subject that calls for further research.