Sean recently wrote a provocative post on why “additionality” — one of the bedrock principles of carbon markets as presently designed — is an expensive waste of time. This is a rich topic, and my perspective as a carbon offset retailer differs from his as an energy producer. It’s worth spending a few posts exploring why.

When we ask whether a greenhouse-gas reduction is “additional,” we’re asking if it would have happened in the absence of whatever incentive we’ve applied. According to the logic of additionality, if the reduction would have happened anyway, then we’ve wasted our money. Additionality is a fairly horrible piece of jargon, and I’ve only ever heard it used in the context of carbon policy. But in understanding why the concept matters, it’s useful to step back and consider how it applies in other contexts.

In fact, the question of additionality arises every time we try to use an incentive to achieve a certain outcome. Say you own a grocery market, and you run a coupon in the Sunday circular to entice shoppers to your store. Lots of people redeem the coupon in the following week. Was the promotion a success? Well, it depends. Presumably some of the people who redeemed the coupon would have shopped at your store anyway. You wasted your money on those folks. But some of the people who redeemed the coupon are new customers, “additional” to your normal store traffic.

To know whether your promotion was effective, you’d have to measure these two groups to see whether the new shoppers compensate for the free riders. Needless to say, such measurements are difficult and expensive in real life, so most retailers don’t bother. They use some rules of thumb to limit their risk and assume that coupons are on balance a boost to business.

Similar logic applies to just about any arena in which economic incentives are applied. The California Clean Car Discount Act aimed to stimulate the purchase of fuel-efficient cars by refunding money to people who buy hybrids. Of course, people were already buying plenty of hybrids in the absence of this incentive, so some significant portion of the refunds would necessarily be “nonadditional,” wasted on people who would have bought hybrids anyway.

Which isn’t to say such refunds are a bad idea, any more than grocery store coupons are a bad idea. Some degree of nonadditionality is inherent in any incentive system. That’s the rub. The degree and costs of nonadditionality matter a lot in designing an incentive program, a point to which I’ll return in a future post.