Should we care about the path not travelled?

One of the nice things about researching a range of questions relating to energy is that you notice commonalities: instances where the same fundamental question pops up in different contexts. I’ve recently been thinking a lot about how we incentivise use of renewable electricity, and also about how consumers adjust their electricity usage in response to price or other signals (demand side response). In both cases, I’ve observed an active question of whether it is possible and necessary to take into account the counterfactual, ie the path not travelled.

When we (as the government, organisations, or consumers) select renewable electricity projects to support, there is a natural desire to only support those that wouldn’t take place without our support. After all, it feels like a bit of a waste to spend money that isn’t actually going to make a difference. This is seen in, for example:

• the reluctance of some customers to pay more for green electricity coming from hydroelectric power stations that have been around for decades
• the fact that suppliers in some European countries aren’t allowed to market electricity to customers as green, if it has been government subsidised
• the sense by some customers that the green electricity label would be meaningless if it didn’t cost the supplier something.

In the case of demand side response, the supplier wants to encourage customers to reduce their electricity demand at times of high system demand (perhaps by delaying their consumption). But demand reduction can only be calculated if we know the counterfactual, ie what would otherwise have been consumed. After all, the supplier will hardly want to pay the customer to not use electricity, if the customer wasn’t going to use it anyway.

So in both of these cases, we have a natural desire to know what would have happened without our intervention — without our investment in renewable production or without the incentive to reduce demand. Unfortunately, this desire is problematic.

Firstly, a lot of actions result from combinations of incentives. To what extent can we claim that one of those actions is the main relevant factor. A producer might not be able to justify renewable electricity production with either government subsidy or marketing rights on their own, but together they might be enough.

Secondly, it is very difficult to know what would have happened without our intervention. We are forced to make assumptions — perhaps that they will behave like they did in the past. I would be sceptical of anyone’s attempt to tell me what my electricity demand would have been if prices had been different.

Thirdly, by making rewards dependent on the path not taken, we create an incentive to put effort into worsening that counterfactual. Companies may rationally strive to demonstrate they would never invest in renewable electricity without the incentive. Individuals may rationally work to increase the amount of electricity they use, in order to maximise the possible amount of reduction. These efforts are at best unhelpful, and are often counterproductive.

As a result, I prefer mechanisms that are less sensitive to counterfactual. I prefer to reward or punish individuals an groups for what they have done, independent of what I believe they would otherwise have done. I support sustainable renewable electricity production, even where it would have happened without my support. And I believe we should reward demand minimisation rather than demand reduction.

That isn’t to say we should completely ignore what would have happened without the incentive. We are liable to be inefficient in achieving our objectives if we don’t take into account which incentives are actually needed. But I believe it is possible to think about how we might influence outcomes, without linking the actual rewards to the counterfactual.

There is also a question of justice, of fairness. Some consider a reward only fair, if it causes the desired outcome. Causation and justice are rich philosophical topics, and I don’t claim to have good theories of these. However, I would note that there are many situations in which we would even prefer to reward someone where the reward wasn’t the cause of outcome, most notably in sport. I am quite happy to see someone rewarded for not using electricity, even if they had no plan to use it anyway.

To summarise, while we should definitely think about what it takes to incentivise desired outcomes, I believe we should resist the desire to tie the incentive too closely to the path not taken; we are liable to get it wrong, and end up with the wrong outcomes.