Over at the Stash, Noam has a thoughtful post on whether it's okay to "offset" carbon-dioxide pollution—say, by buying credits to plant trees when you take an airplane trip:
I guess it comes down to how you feel about pollution--or, at least, gratuitous pollution (that is, pollution with no obvious economic or utilitarian benefit--like buying an SUV just because you like the look of it, not because you need one to navigate tough terrain). If it's immoral to pollute gratuitously, then buying an offset doesn't somehow make it better, any more than buying a papal indulgence or an infidelity offset wipes away a sin. But if the only problem with polluting gratuitously is that it results in pollution, then Frank's right and the offset conceit is fine. I guess I'm inclined toward the former view. It doesn't surprise me that an economist would tend toward the latter.
I think I'd side with Robert Frank here. The act of burning carbon, even for gratuitous expenditures like SUVs, has positives (creating economic growth) and negatives (greenhouse gases that are wreaking havoc on the global climate). The problem here is that people who burn carbon don't pay any price at all for those negatives, so the best thing to do would be to slap a price on that externality.
But it's difficult to calculate what that externality is, precisely, so the next best thing is to use a cap-and-trade system to limit overall carbon concentrations in the atmosphere below what the scientific consensus deems a safe level (roughly, 450 parts per million), and then let the market decide the best price. Under that system, offsets that actually work are quite acceptable—if you emit X tons of carbon but "pay" for it by purchasing offsets and preventing X tons of carbon from being emitted, then your activity is technically carbon neutral. And if you use offsets to prevent 2X tons from being emitted, you've reduced overall emissions.
The main dilemma is that these offsets are tricky to verify. Say you drive your car cross-country, creating carbon emissions, and then pay to have a few trees planted somewhere, which will remove carbon from the air. But how do you know that those trees wouldn't have been planted anyway? And how do you know that the trees will actually stick around for as long as the carbon you've emitted lasts in the atmosphere—hundreds of years? Monitoring can be done, but it's difficult. (Here's a report that ranks private offset providers, and delves into why some offsets are shadier than others.)
One big criticism of the Waxman-Markey climate bill is that it allows polluters to buy, all told, some two billion tons worth of offsets instead of reducing their emissions the old-fashioned way. Joe Romm has been interviewing a number of experts about this, and he now thinks the offsets in Waxman-Markey are nothing to fear—in part because he figures that the main task of curbing emissions by improving efficiency and swapping out fossil fuels with cleaner energy sources will actually prove cheaper than buying offsets. Read his whole post. But on the original question, I'd say the practical aspect of offsets are much more worrisome than the moral aspects.