Lies, Damned Lies...

I was surfing on the interwebs the other night, when I came across one of those crazy reports that you often do. it claimed that big fat ugly SUVs cost if anything less energy then nice, fluffy lovely environment patting hybrids.

Well, naturally, I was curious at how such claims could possibly be justified, so I downloaded and read the 500 page report. And if I'm going to waste so much of my time on something like that, you can be sure that I'll waste your time as well.

So how does this all work? Well, the study is all about the energy cost of vehicles. That is, the total amount of energy involved with creating, running, maintaining and tidying up after the thing is done with. So the numbers, which appear large, apply to the total energy cost. So this is not just the price paid by the consumer, but also the energy involved in the car plant workers getting to work and so on. All sounds good so far.

The first confusing part of this report is that the price is stated in terms of price per mile travelled, based on the expectation of the number of miles travelled. But this varies a lot, and includes some highly dubious assumptions (apparently big fat ugly tanks will get driven three times as many miles as nice fluffy hybrids). So what I did was to pick out, from the hundreds of vehicles and thousands of figures, just a few examples to illustrate how these numbers work.

Here are the totals, stated in terms of total energy cost for my four chosen vehicles: Toyota Corolla, Honda Civic (both "cheap", fuel efficient regular cars), Toyota Prius and Hummer (spit) H3, the latest iteration of the coward's car.

Total energy cost:
Corolla: $123K
Prius: $354K
H3: $403K
Civic: $430K

So straight away these figures seem surprising. How come the Corolla and Civic, two vehicles perceived as quite similar by the public, differ so much in terms of their energy cost? How is the Corolla so cheap? How can a tiny little Civic be so much more costly than the big, fugly H3?

Well, let's split these figures. The 400 page report goes on for pages and pages, breaking down the cost into about 15 components, which makes it a little confusing to the casual reader. But with detailed study, we see that all the careful detailing (down to the energy cost involved in replacing tires) is somewhat irrelevant, because these energy costs are dominated by the cost of disposal of the vehicles:

Disposal Energy Costs
Corolla: $110K
Prius: $326
H3: $363
Civic: $399

(now, the figures in the report don't quite add up properly, but they do more or less work out).

These numbers are big. Huge. They completely dominate the further calculations, and really minimize any other costs. And where do these numbers come from? Why is the energy cost of disposing a Civic 25 times more than the cost of buying a car? There's no explanation. The only break down we have is into recyclable, reusable, and non-recyclable components: $123K recyclable, $254K non-recyclable, and $22K reusable. So now we know why scrap dealers look so cheap and shabby: every time they scrap a little Honda Civic, it costs them quarter of a million dollars in energy.

Are these numbers accurate? Hard to say. What does it mean to say that it costs about four hundred thousand dollars of energy to dispose of a car? I have no idea. But it means that I find it hard to understand, and hard to take seriously. Suppose we focus on the energy cost of these vehicles in terms of their lifetime costs, and put aside for the time being these disposal costs. This gives us the following costs:

Total lifetime energy costs:
Corolla: $30K
Civic: $65K
Prius: $88K
H3: $74K
(by the way, I did warn you that the numbers don't quite add up)

These figures seem a lot more reasonable -- at least they are within constant factors of the cost to the consumer of these vehicles. They still seem a little suspect. The development costs for the Prius are pegged at $30K, the costs for the H3 only $8K -- because, so many more H3s have so far been sold that Priuses. So this cost, while legitimate, does rather seem warped by the number of vehicles sold. The "cost" of the H3 is so low in part because it is basically a Chevy truck body with a big fugly tank piec of plastic stuck on top. The maintenance and manufacture energy costs of the Prius are twice those of the H3 -- possibly due to the complex nature of the vehicle? -- but adding a total of $23K to the lifetime energy cost. Still not clear why a nice little Civic costs twice as much in energy to create and run as its Corolla cousin.

So is this study valid? I dunno, I'm not an expert. But it's clearly a very complex issue. I'd like to see a much clearer explanation of why it is quite so expensive to dispose of vehicles -- and who ends up paying for all this energy? If this is the cost, surely someone is paying? Who? Or are these "projected costs" that it would take to completely and cleanly dispose of the vehicles?

Meantime, expect to see plenty more idiot media outlets taking the press release from the libertarian "reason foundation" glibly repeating "maybe a Hummer is better than a hybrid" soundbite, followed by kneejerk "but hybrids are much more fuel efficient than SUVs" responses, without anyone taking the time and effort to really understand where these numbers come from, or what they denote.