States of industry
You log on to the website, input your information, and there it is: an instant appraisal of how your company compares with your sector, your industry, Europe-wide. This is the ultimate vision of the MEPI (Measuring Environmental Performance of Industry) project, co-ordinated here in the UK by the Science and Technology Policy Research Unit at Sussex University. Matt MacAllan reports.
Indicators are condensed information. Producing indicators, then, involves gathering that information which is to be condensed. MEPI is aiming to produce three types: physical indicators concerned with mass and energy flows through the manufacturing process (i.e. solid waste per unit of output); economic indicators linking physical data to business performance (i.e. emissions related to value added); and environmental indicators linking environemntal impacts to physical data on inputs, emissions and outputs (i.e. acidification potential per unit of output). Thus, for the first 18 months, MEPI set about first defining, and then collating, the information that forms the raw product of the project, the database.
"We came up with two conclusions," says Berkhout, on the issue of exactly what information should be included. "One is that there are some variables that you want to collect for all sectors, the generic variables to do with energy input, water input, CO2 emissions and so on. But in order to do justice to the technological and other environmental specificities, obviously you need industry specific variables as well."
It¹s a long list. "We had some very tricky discussions, some of which are still raging." Berkhout is philanthropical. "This is a pilot study, of sorts. We don¹t want to believe that all these variables will always be necessary, but we will include them in the first instance and see what makes sense."
450 company years
Each partner then went out to companies in each of the six sectors, gathering information from corporate environmental reports, EMAS and ISO 14001 registration, National Emissions Inventories and company surveys. Inevitably, very few of the variables identified were reported consistently to any extent by companies. At the close of the information gathering stage, however, at the end of December, MEPI had accumulated somewhere in the region of 450 company years of site, business unit and firm-level data, from some 280 companies. "In a way that is the most important effect of a project like this," says Berkhout. "The demonstration effect. It shows that, if you know where to look, and if you care to look, then actually there is a lot of information about companies¹ environmental performance out there. It can be collated and made to say sensible things."
Notably, the project has failed to make any headway whatsoever in the computers sector, due, it is believed, to the sheer depth and breadth of the supply chain, and to the fact that the environmental impacts of the computer are not in assembly. The impacts of the very small segment of the supply chain inhabited by the computer manufacturer, as Berkhout points out, are largely irrelevent: "All the Life Cycle Assessments that have been carried out on computers demonstrate that 80% of the energy consumption that takes place does so while the computer is switched on."
Whilst, at the outset of the project, risk analysis concluded that a potential lack of data posed the greatest threat to MEPI¹s usefulness, the collection and collation of so much data for the remaining five sectors posed an entirely different question: validation, as Berkhout admits: "There is a data quality problem. We have to hope that when a company publishes a corporate environmental report, that it does so honestly. Even so, how is a number actually derived? Is it measured? Is it estimated? And if it is measured, how is it measured? What instrument was used? Has it been calibrated recently? Indeed, what is the number? Over what time scale? "We have to assume that the data that is published in an environmental report, or in an Emissions Register, is good. Our attitude is that we wanted to get our fingers dirty with it, and that is the second validation: when we bring these data together and develop indicators, do reasonable numbers or reasonable trends emerge? And, oddly enough, they do."
A MEPI website, scheduled to be up and running in May this year, will
present the findings of the project in sector report format, with selected
data available in tabulated form. The benchmark report-generating site
it is hoped, be the subject of a follow-on project. Berkhout: "We have
demonstrated that there is data out there, and so there is value in
continuing to collect and to build more sectors, more companies, more
countries. We hope that eventaully this becomes a pan-European benchmarking