The statstics I'd really be intersted in (and these would be impossible to measure precisely, buT well-thought-out estimates woul be helpful.) would be the tota

The statstics I'd really be intersted in (and these would be impossible to measure precisely, buT well-thought-out estimates woul be helpful.) would be the total estimated energy usage of new vehicles purchased under the program, compared to the energy usage of the clunkers traded in, and with that net result compared to the energy used in manufacturing all those new cars.