HOW TO MEASURE MORE SUSTAINABLE HOUSEHOLD FOOD CONSUMPTION USING SUPERMARKET DATA?
Luca Panzone and
No 149628, 2013 Annual Meeting, August 4-6, 2013, Washington, D.C. from Agricultural and Applied Economics Association
Behavioural change by households is increasingly anticipated to make an important contribution to the reduction of greenhouse gasses and other emissions. Global studies have shown that housing, mobility, food, and, with increasing income, manufactured products are important sectors to target from a consumption perspective (Hertwich and Peters, 2009). As a result, monitoring of the environmental impacts of consumption at the household level is necessary to evaluate current performance and to support the understanding of how initiatives for change can be implemented. In Europe food is at the core of the discussion of sustainable consumption being responsible for 20-30% of the overall environmental externalities of household consumption (Tukker et al., 2010). Existing sustainability assessment tools vary in scope and methodological approach but are commonly characterised by a production orientation (Ness et al., 2007; Singh et al., 2009). Well established examples are life cycle assessment, material flow analysis and environmentally extended input-output models. In contrast, consumption-based emission accounting is a recent development with the objective to juxtapose consumption and production emissions in order to demonstrate the effects of emissions associated with international trade, especially for GHG (e.g., Kerkhof et al., 2009; Kastner et al., 2011). Thus the environmental literature has yet to propose a tool to measure environmentally sustainable food and non-food consumption at the household level, surprising as this may appear at the first instance. Against this background we develop an indicator of sustainable household food consumption the Environmentally Sensitive Shopper Index (ESSI). Our index is based on revealed consumer preferences and uses supermarket data. The main concept behind the construction of the indicator is that food consumption spans over many different individual products but that its environmental impact is limited to a number of key “dirty” categories (Kramer et al., 1999; Kim and Neff, 2009). Hence, there is no need for an exhaustive view of total food consumption as long as we observe expenditures for these key items. Yet, our index does not impose an upper limit on the number of categories included and ideally would include all products in the supermarket. A second concept is that sustainable consumption is an ethical practice of consuming differently to reduce environmental impact (Evans, 2011). To formalize these concepts in micro economic terms, imagine a retailer with J food categories including C clean categories, and D dirty categories (C+D ≤ J). We define a category as “dirty” if its consumption is relatively carbon intensive, and “clean” otherwise. In each category, this consumer makes two consecutive choices: whether to purchase a product within a specific food category jÎ J and the amount to spend on food in category j. Next, food consumption has to be related to environmental damage. We explore two approaches. First, we assume that sustainability is linear with expenditures by category: consuming more of a “dirty” good means more environmental damage; while consuming more of a “clean” good means less environmental impact. Our second approach allows for non-proportionality between sustainability and food consumption: a consumer is considered more sustainable whenever her levels of consumption are above a threshold for “clean” categories, or below a threshold for “dirty” categories. These thresholds are defined as consuming more than the mode of the observed expenditure shares for the “clean” categories and less than the mode for “dirty” foods. Consequently, the thresholds define sustainability in relative terms as determined by the most frequently observed consumption behaviour (the mode) in the data. We compare different options for the aggregation of the food categories for both the linear and the binary environmental damage approach: averages, arithmetic mean and geometric means. We illustrate the ESSI with a pilot application for food purchases in the UK focusing on carbon emissions. The selection of “clean” and “dirty” food product categories was based on Defra’s (2010) ‘Food 2030’ and Sustain’s (2007) ‘Eat Well and Save the Planet’ reports, which include a set of consumer-oriented guidelines on more sustainable food consumption. While these reports differ in details the consensus is that more sustainable consumption means: eating less meat, particularly red meat, and more F&V; selecting local and seasonal products; drink tap rather than bottled water; and reduce car travel for the purpose of food shopping. The pilot application uses supermarket scanner data for food purchases in Tesco supermarkets. The pilot is based on the weekly expenditures of Tesco’s 16.5 millions UK cardholders over the period June 2009-May 2011. The application is illustrated with examples of how the ESSI can be used to identify environmentally critical periods during the calendar year.
Keywords: Environmental Economics and Policy; Food Consumption/Nutrition/Food Safety (search for similar items in EconPapers)
New Economics Papers: this item is included in nep-env
References: Add references at CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
http://ageconsearch.umn.edu/record/149628/files/AA ... %20POSTER%201907.pdf (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:ags:aaea13:149628
Access Statistics for this paper
More papers in 2013 Annual Meeting, August 4-6, 2013, Washington, D.C. from Agricultural and Applied Economics Association Contact information at EDIRC.
Bibliographic data for series maintained by AgEcon Search ().