Destroying useful stuff that still works is never a smart idea. It takes energy to make a car, lots of energy. In fact, making a car takes almost as much energy as the car will consume in gas over its entire life. Therefore, you can annualize that initial energy investment into a figure that is approximately equal to the amount of gas the car consumes each year.
Conclusion: For each year that you squander that initial manufacturing energy investment (destroy the car while it still works), you might as well have turned the car on, left it in park, and gotten zero mpg until burning through 586 gallons of gas.
Here is the analysis: First, the average fuel economy of the entire US fleet of passenger vehicles is 20 mpg according to the Environmental Protection Agency (EPA). The average miles driven per year is 12,000, according to the EPA. According to the Secretary of Transportation (in a letter to U.S. Senators), the average fuel economy for the "clunkers" in this program was 15.8 mpg. The average fuel economy of the replacement cars was 25.4 mpg. With that information, we see that the average US car uses 600 gallons of gas per year (GPY), the average "clunker" used 779 GPY and the average new car purchased as part of the "clunkers" program uses 472 GPY -- for a savings of 307 GPY for every "clunker" replaced with a new car.
Each gallon of gas contains approximately 10.5 million joules (megajoules) of energy, or 10.5 MJ. Therefore, the new cars save 3,223 MJ/year as compared to their inefficient but loyal clunkers that now await an execution by pouring a glue into the engine block. (Am I the only one who develops a deep, irrational, emotional connection with his cars?)
Manufacturing costs: It takes energy to make a car. When you add it all up, it takes about 80,000 MJ to manufacture an average car. This is based on an average weight of 1,000 KG, and the energy required to make the primary materials as a share of total weight. The breakdown is as follows, according to the UN Scientific and Cultural Organization:
- Steel makes up 50% of the weight of the car; it takes 15,000 MJ to make 500 KG of steel.
- Plastic is light weight, but a HUGE amount of energy is required to make an average of 70 KG per car, worth 11,200 MJ.
- Glass: recycled versus new makes a big difference, but it averages out to 854 MJ.
- Iron: as with glass, recycled versus from ore makes a difference, but it approximates 3000 MJ
- Aluminum: doing a weighted average based on recycling rates for the difference between smelting from bauxite (which takes enormous amounts of energy) and recycling (which takes much less) takes 30,000 MJ
- TOTAL MATERIALS ENERGY COST: 60,054 MJ
Then you have to cast and assemble all of those materials using approximately 20,000 MJ for a grand total of 80,054 MJ of energy.
According to DOT, the average car lasts 13 years. This annualizes to 6,158 MJ which is the energy equivalent of 586 gallons of gas/year. This means that every year of a car's usable life that you prematurely snuff out wastes the energy equivalent of 586 gallons of gas.
Back to some common sense: You can only drive one car at a time. By participating in the program, a participant does save an average of 307 gallons per year with the new car, but he also squanders 586 gallons per year of annualized capital energy investment by destroying his old car. The only argument that the program could make would be that it shifted consumer behavior into buying a more fuel efficient car and that over the car's lifetime the total annual fuel savings will exceed the initial manufacturing investment loss. That is a point that is impossible to prove or disprove. However, if the end goal was to shift consumer purchases to more fuel efficient cars, incentives that didn't involve destroying capital goods would have been better for both the environment and the economy.
As for the economy, measured by the immediate impact on the auto industry and buyers who participated, the program was a huge success. However measured on its longterm effect on the entire economy, the opportunity cost of labor/continued misallocation of human capital was the number one cost. Men employed manufacturing cars that in the absence of artificial government demand would be unwanted could be doing something else that the economy actually did want. Furthermore, with 10-16% unemployment (depending on whether you ignore "discouraged" workers), this program clearly helps only those who have jobs, and specifically only those who have jobs that pay higher than average -- since the average income of a new car buyer is about 20% higher than the average income of the average American. Finally, with September's auto sales on pace to be the worst of the year, down 23% year over year and 41% month over month, it is apparent that the program was nothing more than a sugar fix for the auto industry and offered no longterm help. Even the short lived increase in sales is debatable on an annualized basis as there is evidence that many September sales were pulled into August.