Gunnar - Thanks for the comment and the link to your blog. I noticed you also commented on Rob Hopkin's blog, which is here for anyone else who wants to check it out: http://transitionculture.org/2011/04/15/a-chat-with-the-nation-and-just-how-much-energy-is-in-a-litre-of-petrol/
Contrary to your assumption, I do NOT assume that all our food goes to work. I am calculating energy input based on how much energy a human is expending per hour (125 calories per hour). Whether that person is sitting in front of a TV or weeding carrots is irrelevant. You may observe people all around you at any time and it is quite obvious that they put a considerable amount of time into their leisure activities - very often more than they do into their paid work. (!) I use 125 calories per hour as an overall average because I do things all day at the same rate. I don't necessarily burn more calories weeding carrots than I do walking out to the field or doing my data entry at night. Overall, 125 calories an hour for the 3000 hours I put in over a year's time seems reasonable.
Part of the problem comes from the confusion of input and output. When you say that only 20-25% of human input is output, you may be right. However, both Rob and I were comparing inputs only (liter of petrol vs. energy used by a human laborer for 35 days - how much he gets done is irrelevant). When we eat we consume fuel in the form of calories. We use those calories during the day. If I use more than I burn up, I gain weight. If my calorie input is the same as my calorie output, my weight stays the same. What I get for my output is NOT the same as my input. Modern economists regularly dismiss energy inputs as externals and only measure outputs. This allows them the fiction of saying industrial agriculture is more efficient because each farmer feeds more people, even though it takes 7-10 more inputs (in calories, joules, BTU's or whatever measure you want to use) to produce outputs. Even though I feed fewer people than an industrial farmer, I am more efficient because my energy inputs are lower. This is what input/output analysis is all about. We need a measure of sustainability and the ratio of inputs to outputs is the requisite measure. When I say that I put in 1 million calories input to produce 3.5 million calories of output, it is irrelevant what percentage of my calories go to burn fuel in my muscles and how much lifts up the shovel or the hoe. I am still burning up the calories and THAT is what I am measuring. Rob Hopkins was making the same analogy because he was talking about the thermodynamic internal energy (i.e. capacity to do work plus the capacity to release heat) of a fuel versus the thermodynamic internal energy of the fuel to drive the human engine for 35 days.