Almost always, but the reason is more subtle than you think
In two previous posts, I’ve attempted to establish that additionality is neither some strange concept relevant only to carbon offsets nor an awkward patch used to fix a defect in the design of carbon markets.
Rather, the concept of additionality is applicable to any incentive system, whether subsidy, tax, or whatever. The real question is what degree of additionality is actually necessary or desirable in any given system. Put another way, when should we care enough about additionality to incur the costs of measuring and enforcing it?
Those costs can be quite high, and the benefits sometimes uncertain. Let’s return to one of my previous examples: the grocery store owner who offers coupons to lure new customers, even though most coupons will probably fall into the hands of old, non-additional customers. In this case, additionality is difficult to enforce, and the benefits of enforcement are low (because coupon programs don’t cost much to run). High cost, low benefit: additionality isn’t a concern.
Now let’s examine the carbon offset market. Here, the cost of measuring additionality is high, but the need is even higher. There are at least two reasons for this. The first is the obvious one: carbon offsets can be used to satisfy emission limits under a cap-and-trade program. Non-additional offsets undermine the cap. Good offset projects help to reduce the strain of carbon caps on the economy by lowering the cost of reductions. But too many bad offset projects threaten the whole system by allowing emissions to keep growing.
The second reason is more subtle but equally important: a large proportion of potential projects are non-additional. For example, a recent McKinsey analysis suggested that 40 percent of the greenhouse gas reductions available to us are revenue-positive. And, of course, a lot of possible projects are just flatly unhelpful — I routinely field questions from well-meaning people wondering if they can get offset revenue for leaving some tract of land undeveloped. If we don’t test carefully, non-additional projects will soak up all the available offset financing, undermining a carbon cap and stranding the projects we want to support.
It’s these two facts together that make additionality a critical component of any offset market. If the pool of possible non-additional projects were fairly small, we might not want to incur the cost of measuring additionality. Unfortunately, that pool is large, so we need to be discriminating.
Sean is right to raise questions about incentives in the carbon market, but abandoning additionality offers no solution. In my final post on this topic, I’ll talk about the specific issue of Sean’s cogen projects, and why I think additionality isn’t really the problem here at all.