The impact of food marketing on purchase and the moderating role of motivational quality and socio-economic status

The food environment has received extensive interest from researchers and policy makers, as most blame the high availability of affordable sweet/fat food and the omnipresence of food cues for the current obesity pandemic. Although many criticized the adverse impact of marketing, only a few studies actually provided rigorous theoretical and empirical evidence of its impact, or explored the underlying mechanisms. This thesis elaborates on the theory of food as a motivated choice and on individual differences in responsiveness to the environment. We conduct a thorough review of the literature and propose a comprehensive list of hypotheses regarding the impact of food marketing on purchase behaviour under a Compounded Double Jeopardy framework. We further test the proposed hypotheses empirically using a combination of secondary data. First, we hypothesize that low motivational quality foods (low sugar/fat food; LMQF) are at a double disadvantage compared to high motivational quality foods (high sugar/fat food; HMQF). We hypothesize that LMQF are subject to fewer marketing activities and that the behavioural response to marketing activities directed at LMQF is also lower. Second, we theorize that regions of low Socio-Economic Status (SES), compared to regions with high SES, are exposed to more marketing activities of HMQF and are more susceptible to these marketing activities. We used commercial weekly retail and advertising data from the top grocery chains in the Quebec region for 13 food categories during a three-year period to test our hypotheses empirically. The total number of observations is 89,232 (13 categories x 44 stores x 156 weeks). We derived marketing indicators to measure the 4Ps of marketing, including availability (package size, variety and outlets), affordability (price and price promotion), and promotion (advertising, display, and feature). We further combined two additional secondary datasets. First, we used the Canadian Nutrition File to classify foods into LMQF and HMQF. Second, we used Census Track data to obtain our demographic and SES variables. Our results provide direct empirical evidence supporting our hypotheses within the Compounded Double Jeopardy framework. We found that exposure and susceptibility to marketing activities are strongly influenced by motivational quality and SES. LMQF seem to be at a great disadvantage compared to HMQF, as they are available in less variety and are priced higher. Furthermore, our results indicated that the sales of LMQF are less susceptible to all the marketing indicators (availability, promotion, and affordability) except availability – package size. The high prices and lack of availability of LMQF, coupled with their weaker ability to encourage sales through most marketing activities, can be a major cause hindering consumers' ability to engage in healthy eating. A similar pattern also manifests when we look at SES. We found that prices of LMQF are highest for consumers living in low SES regions. At the same time, promotional activities and advertising of HMQF seem to be more prominent than those of LMQF for consumers from low SES regions. Furthermore, we found that low SES consumers, compared to high SES consumers, are more sensitive to these HMQF activities in both the short term and long term. All these disadvantages make it very difficult for consumers from low SES to engage in purchase of LMQF in place of HMQF.In conclusion, we empirically prove that marketing activities are important factors determining food purchase behaviour. Our results shed light on the double disadvantages of LMQF that hinder their purchase and subsequent consumption. We also indicated obstacles that consumers from low SES regions have to overcome to pursue healthier food intake. We further suggest ways by which policy makers and marketers could benefit from our findings and develop strategies for more favourable food environment, especially for vulnerable consumer groups.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.