Interesting analysis, Tim. One potentially meaningful adjustment you might make is avoiding treating "small," "economy," "fuel-efficient," and "inexpensive" cars as if they're all the same thing. The bottom part of your post does this better than the first part.

A quick way to look at it: it probably costs between 18,000 and 22,000 to build most cars. (That's a SWAG, not based on a published source.) Car prices aren't just the materials and assembly -- they're retirements, healthcare, management, overhead and inventory, advertising, and dealers -- all of which are very stable per-unit costs, no matter what the unit is. (More or less.) So an inexpensive car has a *very* thin margin (though I'm not sure they're all automatically sold at a loss). And inexpensive cars *tend to* be smaller and thus get better mileage, so they boost CAFE averages a bit. BUT, there are different CAFE averages required for cars than for trucks, so Toyota doesn't really need to sell all the Priuses in the world to offset their "gas guzzling" big trucks as far as CAFE averages go.

So it's not really that makers are taking a loss on small cars in order to boost their CAFE; it's that small cars cost so much to produce now, from all the associated overhead, that the price that recoups the cost of a small car is more or less the highest price the market will bear to begin with. The only reason SUV's cost as much as they do (says the former owner of ultracheap 70's Blazers and TrailDusters) is because people demonstrate themselves willing to pay it. The automakers will always charge as much for a car as they can until the high price leads to less profit (through fewer units sold) than a lower price would. With small cars the market won't pay more than $20,000 for, automakers do what they always did: make a little, break even, lose a little. With SUVs they can sell for $45,000 that cost maybe $5,000 more in parts and assembly than the small car, they'll laugh all the way to the bank. That's profit-taking for you.

I don't think that really undermines any of Tim's analysis -- I'm just saying, it doesn't probably have quite so much to do with CAFE as simply the realities of production economics for cars. Might be wrong, though.

Cheers --
Doug

2007 HCH: lovin' every minute of it