5 ms·
Offsets make sense if those selling them can prove the emissions are physically avoided. For example, if I place a CCS system on my cement plant, for each year
by manholio 5y ago
Offsets make sense if those selling them can prove the emissions are physically avoided. For example, if I place a CCS system on my cement plant, for each year I use it I have avoided those emissions, so I can sell an offset certificate to someone who can't do the same, perhaps in some other industry where it's much harder to do CCS, say airflight.
This withstands the "what if everybody is doing it?" test, because if everybody tries to offset in this manner the demand for certificates would push their price to infinity, so everybody is forced by the market to stop emitting instead of purchasing offsets.
The whole rainforest affair sounds like a complete scam. I haven't the faintest idea how rainforest protection, even if it's real, can offset carbon emissions by some other party. It sounds like an ideological connection, because both are "green" things to do, albeit unrelated.
- snewman 5y agoThe problem is that in this scenario, the airline is still emitting CO2, so global emissions never go to zero. If you want to say that the airline is not at fault, because they purchased an offset, then the cement plant operator is at fault: they never offset their own emissions, they only offset the emissions of the airline.
- manholio 5y ago> the airline is still emitting CO2, so global emissions never go to zero. Sure, but global emissions decrease, and the market will allocate effort to the industries where it's cheapest to reduce emissions. So you achieve the maximum reduction with the minimum cost, as opposed to a prohibitive "net zero" goal, which is neither achievable nor a worthwhile target.
- snewman 5y ago> So you achieve the maximum reduction with the minimum cost Offsets do allow for a certain amount of optimization, which is a good thing. However, cap-and-trade, or a carbon tax, are better ways of reducing emissions while allowing for optimization. And unlike offsets, these other approaches can actually take us to net zero. It's also worth noting that in practice, regardless of the policy frameworks we apply, we will not get to net zero unless we start climbing the learning curve for every GHG-emitting sector of the economy (except for those where the best long-term solution is hard offsets via sequestration). A flaw in all of the models that allow optimization via trading of credits (offsets, cap-and-trade) is that they allow us to defer those critical early steps. If all the cement plants buy offsets for the next 10 years, then we don't start learning how to decarbonize cement. > a prohibitive "net zero" goal, which is neither achievable nor a worthwhile target. Citation needed. The mainstream consensus – as embodied in, for example, the IPCC reports – is that it is extremely important that we get to net zero greenhouse emissions. In fact, many scenarios envision that we can+should go to net negative emissions for a while, to draw down some of the excess carbon in the atmosphere. And with the ferocious progress in solar, wind, batteries, and an exploding plethora of other fields, net zero is starting to look very feasible. It's too bad we didn't start sooner, but we will get there.
- arbirk 5y agoBut taxing emmissions and lowering other taxes / adding incentives has the same effect with much fewer drawbacks.
- manholio 5y agoThey are essentially the same approach, emissions taxes require a governmental mandate while offsets work voluntarily, where for various reasons political action is inefficient, for example in the US.