Automobiles: Manufacture vs. Use

Carnegie Mellon University, 1998

This life-cycle inventory of impacts due to the manufacturing and use stages of an automobile was published by Heather L. MacLean and Lester B. Lave of Carnegie Mellon University, in 1998.1 Maclean and Lave used a method of life-cycle assessment (LCA) known as economic input-output (IO) analysis. This method of LCA has the benefit that it allows the researcher to easily trace the environmental impacts of a car purchase not just through the automobile manufacturing industry, but in turn through its various suppliers (of raw materials, parts, chemicals, etc.) The drawback of the method is that it relies on national-average data for most impacts, and cannot provide detail about the reasons for specific impacts. MacLean and Lave analyzed a number of different environmental impacts over the life-cycle of the car. In all cases, they chose not to analyze environmental impacts from the recycling and disposal stage, because they agreed with earlier studies indicating that the environmental impacts of manufacture and use greatly outweighed those of disposal. They based their analysis on a 1990 Ford Taurus, assuming a vehicle lifetime of approximately 14 years and a fuel efficiency of 21.8 mpg.

Figure 1 shows the distribution of energy use over the manufacturing and use stages. The entire manufacturing stage is represented by the slice "Manufacture," which accounts for 10% of the car's total energy impact. The remaining four slices comprise the use stage, 90% of the total energy impact. "Fuel" indicates the energy in the gasoline or diesel fuel used to drive the car. "Fuel cycle" indicates the energy required to extract, refine, and distribute the fuel. "Service" represents the parts and labor required to keep the car working for fourteen years. "Insurance" represents the energy consumed by the offices and services of insurance companies that support car owners.

Figure 1 - Energy consumed over the lifetime of a typical car. The total amount of energy represented by the pie is 1.2 million MJ.
Figure 2 - Toxic releases over the lifetime of a typical car. The total releases represented by the pie are 66.3 kg.


Though energy consumption is a fair approximation of overall environmental impact, the emissions of substances that are toxic to humans don't correlate very well to energy.2 Emissions of toxics tend to correlate more closely to direct human health impacts, rather than environmental impacts. MacLean and Lave report the toxic releases related to car manufacture and use, and these appear in Figure 2. Interestingly, any environmental or health impacts related to toxic releases will be split fairly evenly between manufacture and use, in contrast to energy, which is dominated by the use stage.

This LCA, as well as many others like it, clearly indicate that the bulk of environmental impacts from automobiles occur during the use stage. The implicit message is that if you can replace your car with one that is more energy efficient, chances are high that you truly will be reducing your overall environmental impact. However, if you are a person who considers toxic releases more important than energy use, then it is wiser to hold on to your existing car, in order to avoid promoting the manufacture of a new one.


1 ILEA regularly prints reviews of LCA research published elsewhere. These reviews often leave out many details of the summarized work, and opinions expressed by ILEA may not be the same as those of the original author(s). Please consult the author's original work for a full treatment of their analysis and perspective. The full citation for the work described here is:
Maclean, Heather L. & Lester B. Lave, "A Life-Cycle Model of an Automobile." Environmental Policy Analysis v.3 n.7 (1988), pp. 322A-330A.

2 This is not a conclusion of MacLean & Lave, but rather a rule of thumb used at ILEA based on our experience with life-cycle inventories. The main reason for the lack of correlation is that very small quantities of toxics can have large impacts; an isolated source in one particular sector or stage can strongly influence the overall result.

This summary first printed in the ILEA Leaf, Summer 2002 issue.




Last Modified on Sept. 12, 2003.