Monday, October 26, 2009

Energy Return On Energy Investment (EROEI)?

According to this New York Times article:
"Last week, about 50 scholars in economics, ecology, engineering and other fields met at the State University of New York's College of Environmental Science and Forestry for their second annual conference on biophysical economics. The new field shares features with ecological economics, a much more established discipline with conferences boasting hundreds of attendees...

...Central to their argument is an understanding that the survival of all living creatures is limited by the concept of energy return on investment (EROI): that any living thing or living societies can survive only so long as they are capable of getting more net energy from any activity than they expend during the performance of that activity.

For instance, if a squirrel burns energy eating nuts, those nuts had better give the squirrel more energy back then it expended, or the squirrel will inevitably die. It is a rule that lies at the core of studying animal and plant behavior, and human society should be looked at no differently, as even technologically complex societies are still governed by EROI...

...Through analyzing historical production data, experts say the petroleum sector's EROI in this country was about 100-to-1 in 1930, meaning one had to burn approximately 1 barrel of oil's worth of energy to get 100 barrels out of the ground. By the 1990s, it is thought, that number slid to less than 36-to-1, and further down to 19-to-1 by 2006..."

I haven't looked up how these scholars calculate EROEI. But it seems to me that an approximate value for EROEI is easily found from Leontief Input-Output tables, as available in, for example, the United States National Income and Product Accounts (NIPA). Consider the reciprocals, expressed as percentages, of the diagonal elements of the Leontief inverse, (I - A)-1. The NIPA express the entries in Leontief matrices in dollar terms, not the physical terms I prefer for theory. But since these percentages are pure numbers, calculated as ratios, this makes no difference.

I gather the EROEI is approximately this reciprocal for the petroleum sector. In the North American Industry Classification System (NAICS), the sector of concern is industry code 211, "Oil and Gas Extraction".

A better approximation would account for the energy embodied in the other inputs in the vertically integrated oil and gas extraction industry, that is, the corresponding column of the Leontief inverse. An even better approximation would include some concern for depreciation of fixed capital.


Alex said...

I've always been dubious about this because fans of EROEI tend to a) assume that it's a more rigourous metric than the ones The Economists use, and b) therefore the energy situation must be much worse.

However, in a traditional return on investment calculation for an energy project, the cost side includes a lot more than just the fuel bill, so such a calculation is intrinsically conservative compared to EROEI.

EROEI doesn't take in account energy quality or renewability or carbon-intensity, either.

Robert Vienneau said...

I'm not happy with even my understanding of how to map actual NIPA data into price theory. But I think one should be able to also do accounting in labor value or in energy. I don't see one set of accounts as more conservative than another.

I agree EROEI, as little as I understand it, doesn't take in account many issues of interest. I think input-output analysis can contribute towards analysis of, for instance, carbon or pollution in general.

BruceMcF said...

I-O tables are certainly used in determining lifecycle energy use, but you first need to determine the direct EROI on the natural resource extraction process itself and then the energy content per dollar of the sales of each resource sector in the I-O table.

IOW, the existence of I-O models is why the focus of the EROI research is on the direct energy consumption of distinct energy resource technologies - that's the missing information.