It wouldn't be my first choice, since people are still driving afterwards, but it's looking to be one of the simplest, fastest, and most voter-friendly carbon reduction schemes ever tried on a mass scale.

I guess this depends on whether you look at the carbon necessary to produce a new car as a sunk cost or not. I heard a bit on NPR the other day suggesting that the amount of time it would take for fuel efficiency gains to make up for production carbon cost would in the majority cases exceed the lifetime of the car, making the program a net CO2 emitter.

It seems to me that those cars would have been produced regardless (car inventories were already high) but I'm not sure whether that's the way to look at it.