Numeracy vs feel-good

Most people would agree that increasing energy efficiency is a sensible thing to do, both as a cost-saving measure, to conserve limited fossil fuels and to lower climate impacts. But being willing to save energy does not mean one is efficient in doing so: a new study shows that people are bad at estimating how large energy savings are (or, as The Register put it more forcefully, "People have NO BLOODY IDEA about saving energy"). People tended to think that curtailment (e.g. turning off lights, driving less) was more effective than efficiency improvements (e.g. installing better light bulbs or appliances). They tended to overestimate the benefits of small savings like removing cellphone chargers and underestimate the benefits of large savings such as reducing heating. The study authors somewhat predictably concluded that well-designed efforts to improve public understanding of energy savings would be useful. But would they?

I have in the past blogged about the importance of having a correct sense of proportion. In many domains one thing is not just a bit better than another, but enormously better. Or worse. Substituting fly ash for clinker in cement is much more greenhouse-gas saving than energy-saving light-bulbs, but unless you look at the data for comparison the intuitive guess is likely that they are about the same. The respondents of the study did indeed tend to underestimate the range of energy use between different devices and activities, and this likely contributes to thinking that light-bulbs are about as important as washer settings. The study actually found that numeracy was slightly more important than holding green values in having an accurate view of energy savings (but both did improve accuracy).

We tend to focus on obvious, simple things we can do. Turning off the light when leaving the room is easier and more obvious than getting energy saving bulbs, despite the later actually saving more energy. Hence it seems to be a more plausible action both to do and in terms of saving energy. Anybody can change light-bulbs and satisfy their environmental conscience, despite this having a microscopic effect in the overall scheme of things - changing an industrial process slightly can out-save practically anything one can do privately. But people are interested in (and asked about) what they can do, not what matters in the large. This feeds the myth that 'every little helps'.

Participants who claimed to be more green and do more energy savings were less accurate than those who didn't. There is likely a strong social signalling effect here: many visible green behaviours are likely motivated to some extent by their ability to signal socially popular positions - and that makes their actual effectiveness less important.

The ethical importance of saving energy can be debated but the lessons from this kind of study likely apply across domains. In complex domains we should expect people to underestimate the range of importance different actions have, be biased towards the obvious things one can directly do rather than the strategically important ones, and if there are social rewards for looking like one is concerned then visible actions will be taken with less concern for their actual benefit.

If we really care about achieving something, we should try to research it thoroughly so we can identify the most efficient way of achieving it. Given that different means often differ significantly it has a big effect on achieving the good. Also, many 'facts' people use to motivate action turn out to be erroneous on further investigation: if not checked, we might bark up the wrong tree. Checking facts might be epistemically virtuous but it is also practically useful: we are more likely to do the right thing.

If one cares about something but does not try to learn how to achieve it best, then one is probably more interested in feeling like one cares or look like one cares. This suggests that many current programs of making people "climate conscious" might backfire: people might care more about looking climate conscious than actually doing anything about the climate. It might be better to train numeracy, something that could improve thinking in a wide range of issues.