ECO-EFFICIENCY AND INDUSTRIAL ECOLOGY Toward a Methods Framework for Eco-efficiency Analysis? Helge Brattebø E co-efficiency (E/E) is a concept that has gained considerable attention and widespread use during the past ten years. Basi- cally, E/E relates to strategies and measurements of combined environmental and economic per- formance of product and production systems. Be- cause most of us recognize that the assessment of environmental—and economic—impacts of Many assessment and analysis methods are avail- able from which to choose in E/E, and each of them represents a different an- alytical tradition and in- strumental procedure. An E/E methods framework is clearly needed ... products and services need to be founded on a life-cycle ori- entation, we also realize that such strategies and measure- ments will have to include life- cycle performance of products and services, including the dis- tribution and use phases, as well as the end-of-life disposal and recycling phases. The challenge of applying E/E in practice, and in research, thus turns out to be the same as the challenges of life-cycle as- sessment (LCA) and life-cycle costing (LCC): how to set system boundaries, temporal and spa- tial scale phenomena, and functional unit def- initions. We could, of course, apply E/E analy- sis at the local facility scale, or at the corporate scale including only those production units di- rectly controlled by a firm. This is often done in business environmental reports today, leaving out the consumption and end-of-life impacts, fo- cusing strictly on production impacts. c  2005 by the Massachusetts Institute of Technology and Yale University Volume 9, Number 4 But do we agree that this is E/E in terms of holistic principles? I guess not. Moreover, if E/E is applied to the whole life cycle of a product, but it excludes much of the impact on the background system, or environmental impact categories that may be difficult to evaluate, perhaps due to in- sufficient information, is this in line with holistic principles? Again, I guess not. On the other hand, users of E/E claim that the anal- ysis should be designed in such a way, including the choices of impact categories and indica- tors, that the analysis is felt to be meaningful, easy to work with, and appropriate to the user and decision-maker. Hence we clearly have many tradeoffs in- volved, related to the analytic method itself, the boundary set- ting, and the choice of indi- cators, as well as the aggre- gation of environmental and economic information, with respect to the nu- merator and denominator of the E/E fraction and the E/E index itself. One step back from the common E/E defini- tion (E/E = units of value generation per unit of environmental influence), we can think of differ- ent approaches to measuring sustainability perfor- mance. First of all, E/E does not cover the social dimension of sustainability, and this is of course a methodological weakness in itself, because so- cial aspects then need to be studied in parallel to the environmental and economic aspects that are part of the E/E framework. Macroscale economic analysis often makes use of indices such as re- source productivity (the share of primary resource inputs that are converted into useful product http://mitpress.mit.edu/jie Journal of Industrial Ecology 9