Disentangling direct and indirect effects of credence labels

Abstract

Purpose – The purpose of this paper is to disentangle the direct and indirect effects of three credence labels (Australian, animal welfare and grass-fed) on US consumer attitudes toward buying beef steaks. Furthermore, it explores the impact of consumer attribute knowledge, usage frequency, education and opinion strength on the magnitude of direct and indirect effects. Design/methodology/approach – Data are collected through an online experiment with 460 US consumers and analyzed with path modeling. Findings – The Australian label generates a 86 percent negative direct effect vs a 14 percent negative indirect effect on consumer attitudes, which means that US consumers do not make strong inferences to form their attitudes toward buying Australian beef. The animal welfare label generates 50 percent direct and 50 percent indirect effects. The grass-fed label generates only indirect effects (100 percent). The higher consumer education, attribute knowledge, usage frequency, education and opinion strength, the weaker are the indirect effects of credence labels. Research limitations/implications – The study focusses on consumers in one country (USA), one product (beef steak) and one label across three attributes, therefore generalization of results is limited. Practical implications – The study offers a tool to agribusiness managers as well as to policy makers, NGOs and consumer groups to design and assess the effectiveness of communication campaigns attempting to strengthen (or weaken) consumer inferences and attitudes relative to credence labels. Originality/value – Despite the wide literature on consumer inferences based on credence labels, this is the first study that quantitatively disentangles the complex set of inferential effects generated by credence labels and explores common relationships across multiple credence attributes.