Now I'm actually reading the article, and there are some interesting little
facts along the way.
There's also a discussion on rec.bicycles.misc here:
http://groups.google.com/group/rec.bicycles.misc/browse_frm/thread/a6a739a9a02b0b59/37366830108b9b3a?hl=en#37366830108b9b3a
Some of those facts:
Average automobile in US uses 3.8 MJ/km. This is substantially higher than
the average for the OECD countries of 2.5 MJ/km. So, roughly speaking, if we
drove the smaller, more efficient vehicles common elsewhere in the developed
world we could drive just as far just as often and still use one-third less
gasoline.
A person on a bicycle uses 17kJ/km directly, or 77kJ/km in terms of food
energy (since the ratio of work done to food energy consumed is about 22%).
The energy required to produce, process, and transform the food is 5.75
times greater than the energy in the food, so the total energy cost is about
0.44MJ/km, or about 1/6 to 1/9th that of the automobile. There's a lot of
difference between 17kj and 440kj, which might explain some of the
discrepancies in some of the discussions on this newsgroup over time over
the food energy cost of cycling.
Several dozen studies show exercise expenditure of 4.2 mJ/wk (about 1000
kcal, or what we usually call 1000 calories) is associated with a reduction
in all-cause mortality.
The fatality rate (from accidents) is assumed to be 12 times as high per
kilometer for cycling as for driving. Fatalities, of course, lower future
energy consumption.
Increasing logevity does not change end-of-life healthcare requirements, but
extends the healthy years.
Ulrich's conclusion is that bicycling that replaces automobile use (even if
you still own the automobile) lowers total per capita energy consumption
from all sources by -.005 (i.e. half of 1%), even allowing for the increased
energy used by living longer. 0.5% is well within the error range of his
analysis, so the effect could be zero.
However, if your biking kilometers don't replace automotive use, you don't
save energy, but still live longer. Therefore your energy use increases by
.037 (3.7%) This makes sense -- many of us have noted that driving to a
ride uses up energy rather than saves it.
Electric scooters or electric bicycles have similar energy patterns and
fatality patterns as a bicycle (they just use electricity instead of food),
but don't increase longevity. Therefore, they have a more positive impact on
energy use over a lifetime.
As others have noted, Ulrich assumes the cyclist still owns the same number
of vehicles so he misses the energy benefits of not manufacturing the car.
Are there a lot of such people with "one less car" in their household (who
aren't posting on this newsgroup)? Is there an estimate somewhere of how
many? If there were such an estimate, this could be factored in.
All in all, the study documents its assumptions and is presented in a clear
manner and is worth a look.
Earl wondered who paid for the research. The paper doesn't say, but Ulrich
does have another bicycle paper in his resume, and this usually doesn't
indicate hostility:
Taylor Randall and Karl Ulrich, "Product Variety, Supply Chain Structure,
and Firm Performance: Analysis of the U.S. Bicycle Industry," Management
Science, Vol. 47, No. 12, December 2001, p. 1588-1604.
There's also this chapter:
Karl Ulrich, Taylor Randall, Marshall Fisher, and David Reibstein, "Managing
Product Variety: A Study of the Bicycle Industry," in Managing Product
Variety, Tech-Hua Ho and Chris Tang (editors), Kluwer Academic Publishers,
1998.
There's a bit of small-world theory involved here, since I just presented a
paper extending one of Reibstein's other models at an INFORMS conference,
(and Management Science is an INFORMS journal).
