One More Awareness Gap? The Behaviour–Impact Gap Problem

Abstract

Preceding research has made hardly any attempt to measure the ecological impacts of pro-environmental behaviour in an objective way. Those impacts were rather supposed or calculated. The research described herein scrutinized the ecological impact reductions achieved through pro-environmental behaviour and raised the question how much of a reduction in carbon footprint can be achieved through voluntary action without actually affecting the socio-economic determinants of life. A survey was carried out in order to measure the difference between the ecological footprint of “green” and “brown” consumers. No significant difference was found between the ecological footprints of the two groups—suggesting that individual pro-environmental attitudes and behaviour do not always reduce the environmental impacts of consumption. This finding resulted in the formulation of a new proposition called the BIG (behaviour–impact gap) problem, which is an interesting addition to research in the field of environmental awareness gaps.

Appendix

Calculating the Ecological Footprint per Spending Unit from a Top–Down Approach

A combination of carbon footprint data and symmetric input–output tables (SIOT) are used in order to calculate the total carbon footprint for consumption categories. Ecological footprint per spending unit is then derived from an environmentally extended input–output analysis.

Bicknell et al. (1998) were the first to introduce generalized input–output analysis, incorporating it into the methodology of ecological footprint calculations. The method is based on combining macro-economic input–output tables with land multipliers representing the environmental input in the system. The suggested environmentally extended input–output analysis proved to be extremely useful as it was able to capture direct as well as induced environmental impacts throughout the supply chain, thus showing the environmental consequences of consumption. The issue of responsibility of consumption activities for pollution could be raised and a meaningful answer could be provided.

Lenzen and Murray (2001) revised the Ecological Footprint method. They made modifications to the original concept in order to make it suitable for input–output analysis. Wackernagel and Rees (1996) also combined Ecological Footprint accounting with a monetary input–output analysis. Tukker and Jansen (2006) produced a review of studies focusing on extended environmental impacts, which used the input–output analytical methodology. They concluded that the major part of environmental impacts is associated with the following consumption categories: housing, transport, and food. Kerkhof et al. (2009) came up with a novel approach in their analysis of household expenditures in the Netherlands; they were the first to make a consistent analysis of expenditure elasticity in four impact categories. This analysis allows the examination of the elasticities by regression analyses and a comparison at the product level. The study deals with household expenditure, a main driver of energy requirements and environmental impact (Lenzen et al. 2004, Pachauri 2004). Using the top–down technique of input–output analyses, the first step utilized by Kerkhof et al. (2009) was quantification of the impact intensities of household goods and services. Following this, results were derived from a combination of household consumption data. Indirect impact intensities were calculated by means of environmentally extended input–output analysis. The study analysed annual expenditure on six aggregate product groups by equivalent expenditure deciles, showing that high-expenditure households spend more money on “development, leisure, and traffic” and on “house,” but with growing expenditure the demand for food decreases relatively. It is thus important to examine how consumption patterns change according to level of expenditure.

In a recent study by Druckman and Jackson (2009), the authors raised the question how much CO₂ is attributable to different kinds of needs and desires and to what extent is decoupling occurring between household expenditure and CO₂ emissions. The study aimed to define the carbon footprint of different segments of UK society compared to each other.

EUROSTAT (2008, pp. 481–489) and OECD have taken up the methodology and EUROSTAT currently publishes emissions of greenhouse gases and air pollutants induced by final use of products as a result of environmentally extended input–output analysis. The data provided by EUROSTAT refer, however, to the European Union as a whole only; data for individual countries are not provided. Thus the author followed the methodology suggested by the literature and EUROSTAT and calculated the ecological footprint/spending and carbon footprint/spending ratio for Hungary.

The methodology is based on combining symmetric input–output tables taken from EUROSTAT accounts and ecological footprint data. The ecological footprint data for Hungary were drawn from the country report of the Global Footprint Network. Their ecological footprint calculation has the advantage of taking into account the embedded carbon in imported goods, thus identified carbon by import rather than production of goods.

This way a hybrid economic-environmental national account was produced that serves as a basis for defining the Leontief inverse and the ecological footprint requirement of consumption. Leontief inverse captures induced impacts throughout the supply chain and permits the calculation of total requirements embracing both the direct and indirect input requirements of consumption.

The environmentally extended hybrid input–output tables are of the following form:

Table 1 Structure of environmentally extended input–output tables

A(x) is a 59 × 59 matrix, called a transaction matrix.

Calculation followed the following steps:

• 1. Production of an environmentally extended input–output table that complements the industry transaction matrix A with industry aggregate ecological footprint data. This step requires disaggregating all ecological footprint components, including crop footprint, grazing footprint, fishing footprint, carbon footprint, and built-up area footprint into sectoral data. E(x) is produced as a result of this step. EF(c _h ) includes the direct emissions of households mainly from heating and mobility activities, and built-up land.

• 2. Calculation of the physical coefficient vector (EF(x)/X).

• 3. Calculation of the Leontief inverse matrix, (I−A)⁻¹, using an industry-by-industry symmetric input–output matrix from the EUROSTAT database. I is the identity matrix.

• 4. Calculation of the total intensity vector as the product of the Leontief inverse matrix and the physical coefficient vector. EF(x)/X × (I−A)⁻¹

• 5. Calculation of total EF requirements of households by multiplying the diagonal intensity vector and the final demand by household vector. EF(x)/X × (I−A)⁻¹ × C. The multipliers for primary inputs are multiplied with a matrix of final demand by category to assess the direct and indirect primary input requirements (this case ecological footprint) for the various categories of final demand.

• 6. Reallocation of final demand into consumption categories using national statistics, COICOP tables, and CPA classification of Eurostat. COICOP is the classification system of United Nations, short for Classification of Individual Consumption Purposes that classifies consumption categories into final purposes. Thus it gives a much more aggregate and useful functional categorization of consumption activities than the 59 sector or product group matrix of NACE and CPA used by EUROSTAT. NACE refers to the categorization of sectors while CPA to categorization of products. Both product-by-product and industry-by-industry SIOT (symmetric input–output tables) are available in the EUROSTAT system.

• 7. Adding direct ecological impacts EF(c _h ) of households to the table and reallocate them to final purposes according to COICOP. These include direct greenhouse gas emissions from combustion of fuel for heating or mobility, and built-up land.

• 8. Calculation of the ecological footprint for individual consumers, based on individual spending data derived from the representative survey and the EF per spending data gained from the statistical analysis described above.