Thursday, March 22, 2012

The Lovins Paradox: "this old canard"

Amory Lovins has responded to David Owen's commentary at the New York Times on "Efficiency’s Promise: Too Good to Be True." In his Times comment, Lovins cites his complete response at the Rocky Mountain Institute blog, wherein he asserts: "There is a very large professional literature on energy rebound, refreshed about every decade as someone rediscovers and popularizes this old canard."

Now, "this old canard" -- the Jevons Paradox -- is based explicitly on an even older canard the Sandwichman has dubbed the "Rasbotham rebound", the truism opposite to the so-called lump-of-labor fallacy. Although he may not be aware of it, in trivializing the Jevons Paradox, Lovins is implicitly trivializing the Rasbotham rebound as well. So, in effect (as many economists would claim) he is embracing the lump-of-labor fallacy. You cannot categorically reject both the Jevons Paradox and the lump-of-labor fallacy because the two are diametrically opposing principles. If "A" is absolutely false, then "B" is absolutely true. However is "B" is absolutely false, then "A" is absolutely true.

There is, however, a third possibility, which is that both "A" and "B" are only conditionally true. They are true in some circumstances but false in others. In that case, their relative importance vis a vis each other can only be gauged in context. It is not sufficient to find circumstances where "A" is true and other circumstances where "B" is false. One must examine the relationship between "A" and "B" in a given circumstance. Or -- to bring it back to the language of energy efficiency, energy consumption and employment -- one must look specifically at the energy intensity of employment, not the energy intensity of GDP or the micro-level effects of energy efficient light bulbs on the demand for lighting.

The Lovins Paradox thus can be stated as: even if Amory Lovins is right about the Jevons Paradox (or rebound effect) being an "old canard", the implications for energy consumption are troubling because of the intricate linkage between energy consumption and employment. In other words, dispensing with the rebound still leaves us with what David Owen calls The Conundrum (see video embedded below). The following chart compares the energy intensity of GDP in the U.S. with the energy intensity of employment (energy consumption per worker). The green line shows the index Lovins likes to cite, energy intensity of GDP from 1949 to 2009. The blue line shows energy intensity of employment in the U.S. for the same period. The red line shows the energy intensity of the labor force (because employment data is not available) for the world from 1980 to 2006.

Sources: World Bank, U.S. Bureau of Labor Statistics, U.S. Energy Information Agency

World energy intensity of employment in 2006 was around five percent higher in 2006 than it was in 1980. This is not an improvement, not even a relative improvement.

2 comments:

JW Mason said...

You are definitely right that the two "fallacies" are logically inverse of each other. So you're right, if one is *categorically* true then the other must be categorically false. But as you say, they both may be conditionally true, in which case both could be true or both false in a particular context.

Seems to me another way of framing this is, is the binding constraint on output labor, or natural resources, or neither? If it's labor, then both lump-of-labor and the Jevons paradox are fallacies. This is the mainstream view, and it is the correct view in situations where -- as in our standard macro models -- labor is the binding constraint on output in anything outside the demand-determined short run. This may have been a reasonable description of developed economies in the 20th century.

On the other hand, if it's natural resources, then lump-of-labor and Jevons paradox are not fallacies, they are both correct. This may be true of the world we are heading toward, and was almost certainly true for most countries before the industrial revolution.

Finally, if the binding constraint is something else, then lump-of-labor is true but the Jevons paradox is false. Or so it looks to me.

Sandwichman said...

"Far from dampening global development, lower energy bills accelerate it." -- Amory B. Lovins, "More Profit with Less Carbon," Scientific American, 293, 74 - 83 (2005).

Not that I disagree. But unless one wants to split hairs about the semantics of "global development" and "lower energy bills", Lovins said that energy efficiency ("lower energy bills") accelerates energy consumption ("global development"). If it looks like a duck, swims like a duck, and quacks like a duck... the it is probably "this old canard"!