Draft February 25, 2008 Environmental Life Style Analysis(ELSA) Timothy Gutowski, Amanda Taplett, Anna Allen, Amy Banzaert, Rob Cirinciore, Christopher Cleaver, Stacy Figueredo, Susan Fredholm, Betar Gallant, Alissa Jones, Jonathan Krones, Barry Kudrowitz, Cynthia Lin, Alfredo Morales, David Quinn, Megan Roberts, Robert Scaringe, Tim Studley, Sittha Sukkasi, Mika Tomczak, Jessica Vechakul, and Malima Wolf. Massachusetts Institute of Technology Dept of Mechanical Engineering Abstract In this study we connect life styles and spending patterns to environmental impacts and economic implications for people living in the United States. The results show that even the most modest life styles(Buddhist monk, homeless etc) have impacts many times larger that the world average. Key Words: Life Cycle Analysis, Input/Output Analysis, Life Styles Introduction The study of environmental impacts associated with household consumption has been an area of considerable attention for the last 30 years. A few examples of this work would include the early papers focused on energy use in the U.S. [Herendeen 1976, 1981], and studies that look at energy consumption in other countries [Lenzen 1998, 2006]. More recent work has looked at the growing importance of imports [Weber 2007a, 2007b], and at the broader range of impacts associated with products [Tukker 2006]. These studies found, among other things, that large amounts of energy are embodied in products and services, that there is a close correlation between energy use and expendable income, that imports and their off-shore energy use are of growing importance, and that housing, transportation and food often dominate environmental impacts. While the current study does not disagree with any of these results, it differs from other studies in several ways. First, it focuses more on individual life styles (in the United States) rather than national statistics. This allows us to look at variations in life style within a certain income bracket, and estimate what potential improvements are realistically possible. Secondly, because this study works from an expenditure basis, it can directly compensate for certain rebound effects which are often ignored when looking at the impacts associated with products.1 And thirdly, this study includes “subsidies” received to support a life style and the impacts associated with those subsidies. One of the largest of these so-called subsidies is from the government. Methodology and Allocation Issues This work was done as part of the Spring 2007 M.I.T. class “Environmentally Benign Design and Manufacturing” i.e. Mechanical Engineering 2.83 (for graduates) and 2.813 (for undergraduates). The students, listed in Table 1, (at the end of the paper) developed the spending profiles for a wide range of life styles for people in the U.S. In many cases the students interviewed the subjects about their life style choices and obtained detailed data. In a few cases, especially for the extremely rich, the life styles were estimated from information available to the public. The estimated annual disposable incomes for these people ranged from negligible to tens of millions of dollars and beyond for the very well to do. In each of the nine expenditure categories listed in Table 1, the graduate students in the class developed detailed spreadsheets that standardized data collection, and in most cases interacted with the Carnegie-Mellon University/ Environmental Input-Output Model (CMU/EIO) to calculate impacts [Hendrickson 2006]. In a few cases, the modules functioned as stand alone models, for example, in such areas as transportation fuels use, and residential fuel use. In general the expenditure categories were the 1 For example, if an individual buys and uses a medium sized hybrid car instead of a conventional engine SUV, she must also account for how she spends the savings (both from the purchase price and the gasoline savings). If for example she flies to Europe for various holidays, her impact may, in fact, be larger than with the SUV and no trips to Europe.Draft February 25, 2008 same as used by the Bureau of Labor Statistics [BLS 1998]. Because of potential over lap in some of these categories, care was taken to avoid double counting. A special feature of this study is that the methodology employed here allocated all identifiable “subsidies” to the recipient, and included the associated impacts in their impact totals. The single largest subsidy was from government services identifiable in the input/output tables. These included components of federal, state and local government which were allocated evenly to all inhabitants in the U.S. resulting in a subsidy of $4391 in 1997. At the same time, services with clearly identifiable recipients such as educational services, Medicare, Medicaid etc. were allocated only to those individuals who received the service. Other sources of subsidies included in this study were from parents (for children), insurance companies (for the severely impaired or disabled), ones own savings (particularly in retirement), and soup kitchens and homeless shelters (for the homeless). Special care was taken in differentiating benefits associated with work (frequent flier airplane trips, generous travel or entertainment budgets etc. which were allocated to work and not to the personal profile) from personal expenditures. This procedure has all of the benefits and liabilities associated with using the input/output approach. For example, we were not able to differentiate between a luxury car and a compact car, except for price and fuel mileage. However, in some categories such as food, there can be considerable differentiation. In each of the nine areas spending options were ranked according to impact intensity and plotted as Pareto plots. This allowed us to return to these life styles and explore variations in spending within a fixed total amount and calculate the impact effects. “Total incomes” were calculated per person according to the general formula: “Total Income” = income – taxes – subsidies paid to dependents + subsidies for this life style. For example, in the case of a family of four with two income earners, the income was pooled, taxes and support for the children was subtracted, and the result shared between the income earners, to this was added any individual subsidies from the government etc. The disposable income, expenditures,