Cash for Clunkers was certainly a popular program and was certainly more "stimulating" than the bailouts. The question is, was it really worth it? From an environmental and/or an economic standpoint?

I was always persuaded by the environmental reasons. After all, it got clunkers off the road a few years earlier than they would have been otherwise, and meant less emissions of carbon and other pollutants. Upon further analysis, even leaving out the emissions it takes to destroy the clunker and the very real possibility that an environmental friendly car will be driven more often than a clunker would be, I would conclude that it was not worth it.

Cash for Clunkers is hardly the most cost-effective way to reduce carbon emissions. The average change is from 16 mpg to 27 mpg. Assuming the new car is driven just as much as the clunker (and that's a generous assumption), the average driver drives 12,000 miles a year.

That brings us from 750 gallons of gasoline per year (12,000/16) to 444.44 (12,000/27), or saving about 305 gallons of gasoline. A gallon of gasoline emits about 20 pounds of CO2, so we're releasing about 6,100 pounds less of CO2 for each year the clunker would have been on the road. A ton is 2,000 pounds, so we're talking about 3.06 tons of carbon each year the clunker is off the road. If the clunker is taken off the road 4 years earlier, that's 12.22 tons of CO2 saved. If 5, 15.28. If 3, 9.17

Then we need to take the average rebate, $4,215, but let's do $4,200 for simplicity, to see how much we are spending per ton of carbon. If we are taking the clunker off the road 5 years earlier, saving about 15.28 tons, we are spending $274.87 per ton (4200/15.28). If 4 years earlier, we are spending $343.70 per ton (4200/12.22). If 3, $458.02 per ton (4200/9.17).

To put this into perspective, Waxman-Markey would be $28 per ton according to the CBO. To get that kind of bang for our buck assuming $4,200, the new cars would need to save 150 tons of CO2. In other words, the clunker would need to have been on the road for about 50 more years otherwise. Again, at best. You can access this Excel file to input the numbers yourself

Furthermore, I fail to see why we should be paying people to stop polluting. We don't pay people not to murder or steal, we make it illegal and punish people who do. It's not like education, where positive externalities exist so we subsidize it. People who drive fuel-efficient cars are not causing a positive externality, they just are causing less of a negative externality. Not taking is not the same thing as giving, and not causing a negative externality is not the same thing as causing a positive externality. I've always been a supporter of pigovian taxes and other ways to internalize the costs. We should make people pay the true cost for negative externalities, not pay people not to do them.

But we are told, the environmental benefits are secondary. The real benefits are economical. Not really.

As we know, the Cash for Clunkers program was meant to prop up demand for fuel-efficient cars and give an economic boost, creating more jobs. Similar to the typical economic model: government spending stimulates aggregate demand which gives businesses more money which helps them hire more workers to keep the supply up with the demand. However, we ignore the other half of the equation. Taking $4,200 out of the economy then putting $4,200 in amounts to awash. Take money from Peter and give it to Paul, Paul is now incentivized just as much as Peter is disincentivized. In other words, no net gain. We need to take into account the loss of economic activity from Peter, not just the gain from Paul.

Theoretically, it amounts to zero, but practically, it amounts to less than zero. For one, higher marginal tax rates reduce the incentive to work (marginal analysis). No, people don't refuse a pay raise because of the tax bracket they enter, but everyone has a breaking point to which he or she will no longer work (for example, if a breaking point was 50%, the person would not be willing to work extra if only half [or less] of everything earned gets to be taken home. Time would be better spent on leisure). Apart from that, the government has neither the information nor the incentive to spend money based on economic calculations. No individual, government included, can aggregate or calculate the subjective well-being of everybody. Only individuals can do that for themselves; individuals can spend their own money for their own subjective well-being. A pure cash transfer from Peter to Paul, therefore, is the closest thing you can get to zero, as Paul can now spend the money based purely on economic calculations.

Even if the government was able to aggregate everyone's subjective well-being, government has the incentive to spend money based on political calculations. Sometimes political calculations may coincide with economic calculations, but the key word is coincide. Government has the incentive to do what's necessary to stay in power, and sometimes that may be providing for the general welfare, other times not. Public choice theory explains a lot. Individuals, on the other hand, have every incentive to spend money economically, as mentioned above, and the information to do so. Government has a role in internalizing the externalities, so economic decisions are based on the true costs and benefits.

Subsidies, and anything else that reduces the costs to individuals, turn decisions that previously did not make economic sense to individuals into ones that do. Taxes, and anything else that increases the costs to individuals, turn decisions that previously made economic sense to individuals into ones that don't. The whole premise of externalities is that they are external, so decisions that externalize the costs the most are the most economic for individuals to choose, and pigovian taxes serve as a way to counter that. Beyond that, taxes prevent decisions that would have been beneficial to all parties involved.

Subsidies, on the other hand, beyond for positive externalities, are like steroids. A football player takes steroids, he previously did not make sense for a team to sign at $x, but now does. The team signs the player, and at first everything is great. The player is performing, the team is doing well, but then something happens. The player knows that if he stops, his performance will go down and he may be released or given a pay cut, especially if the team figures out they have been tricked. So the player keeps taking the steroids, but they finally take their toll on the player, he can no longer play, and his career is cut short. All parties involved are now only the worse for the player having taken steroids and the team having signed that player. The player is a mess. The team has made a malinvestment, investing money in that player that would have been better invested on someone who made sense without the steroids, and they are now hurt over the long haul.

Which brings us to Cash for Clunkers. Take the steroids analogy and apply it to a situation where the team has been investing much money into an underperforming player. The team would be better off either reducing that player's pay, or releasing him, and investing the money in another player or players. But instead of either of those, the player is given steroids, and the team and player are due for a bust later. Essentially, that's the situation with Cash for Clunkers. The auto industry over-expanded during the boom, investing more in capital and labor than was economical, and now is experiencing the natural consequences, otherwise known as a loss. We'd be better off if capital and labor moved to more economical uses, but instead, we are propping up demand in that industry, keeping the capital and labor there, or worse yet, causing the industry to buy even more capital and labor. As the Washington Post said, they are due for a hangover. At best, Cash for Clunkers was going to delay the inevitable movement of capital and labor, preventing it temporarily from going to more economical uses. But we now have a worse situation where the auto industry has acquired more capital and labor, so the inevitable movement and loss will now be even worse to the auto industry and the economy as a whole.

If we want to spur investment in the right direction and have it last, we should be internalizing the costs of going in the wrong direction. A carbon tax, more than Cap and Trade and certainly more than Cash for Clunkers, is the answer. No making winners and losers, just taxing the cost at the source, encouraging more efficient and less use of carbon. Make the use of carbon reflect its true cost, and investment in alternative fuels will be able to stand on its own two feet.