A psychological carbon tax

December 12, 2011

There’s plenty of understanding that efficiency measures can sometimes (often) fail if they fall victim to Jevon’s Paradox. The common response is that we need to couple efficiency measures with a carbon tax or similar policy that give people both the means and the incentive to decrease energy consumption.

But let’s look at the underlying goal of a carbon tax: to change behavior. How does it change behavior? By penalizing undesirable behavior (financially). The downside to a carbon tax is that it requires overcoming many political hurdles that don’t seem likely to be overcome anytime soon.

This past week, climate negotiators came to a new agreement in Durban, South Africa that depending upon how you look at it was either a great success or just another example of how much trouble we’re in. Given the extremely low expectations going in, it’s a success just that all major nations were brought in on the agreement. But from what I can tell, the agreement is a vague one and will not produce anything close to a cap at 450 ppm CO2 (which by Hansen’s take is still too high). At best it provides a toe in the door for a real agreement in a few years.

We can let this process play out on the global stage, but I’m thinking we should look for alternatives.

If our goal is simply to penalize undesirable behavior, we can use other means. Plenty of work in social psychology and behavioral economics (neither of which I have any detailed knowledge of, so insert appropriate caveats here) has shown that people don’t simply do a spreadsheet in their heads and make the most efficient, rational choice. Thus if we were to leverage this fact—that psychological factors play a large role in decision-making—then we might be able to create a psychological carbon tax: a tool or set of tools that requires no top-down energy policy shifts and yet creates negative incentives for energy use.

Let’s begin with oil. In the U.S. we use about 3 gallons of oil per day per capita, yet most of that oil consumption is entirely invisible. Oil is not on the brain on a daily basis for most folks. In its common conception, oil is something you get changed every several thousand miles in your car, or is something you pick up at the grocery to cook dinner—both minimal uses (and in the latter case, not even the right kind of oil).

So let’s say that only some small fraction of the population (say 10% or so) is aware of petroleum use on a daily basis. Suppose that we could make an additional 10% of the population aware of oil use, and among that 10% suppose that the awareness were to make half of them decrease their oil use by half. That gives us the following:

312M x 0.1 x 0.5 x 0.5 x 3 = 23.4M gallons saved per day.

Clearly there are some unanswered questions this argument raises:

  1. Why would awareness of oil use cause a decrease in oil use?
  2. How can such a large numbers of people be made aware of their oil use?
  3. Is that much of an impact anyway?

Let’s take these in turn. First, why would awareness cause a decrease in oil use? It seems there are a few reasons why it might: guilt (among those who are in-name environmentally aware but put it out of mind), outrage (towards either those who supply the energy, towards those leaders who never take action on energy, etc.), empathy and sympathy (with/for those humans and non-humans affected by climate change and energy problems), envy or aspiration (by learning that someone else has something better than you), and reason (among those few who like to respond scientifically to informed argumentation).

Next, how can we make people aware of their oil use? It seems we have to target the points of origin: the places where people buy their oil. In the U.S. this is primarily at gas stations. Fortunately, there are far fewer gas stations than people (about 160k) so providing awareness only requires covering these locations. And this brings us to the first simple idea of a psychological carbon tax: the lowly sticker.

Imagine a sticker on a gas pump that said something like this: “69% of this comes from dictators” or “1 gallon = 600 hours hard labor”. A sticker isn’t a place for long-form reasoned analysis, but if it’s short and catchy, it might just cause a flicker of awareness during what is usually a fairly routine and mechanical procedure (filling a car’s tank with gasoline). There are different strategies we might employ with these stickers, since different people in different moods with different personalities will respond well to some and not to others. Some stickers might be designed to attract attention, while others might be designed to blend in with the numerous other stickers already on gas pumps. Some might be designed to evoke guilt while others might employ reverse psychology. (Imagine a sticker purporting to be from BP’s or Exxon’s CEO saying that they had such a profitable year last year, they decided to “give something back,” where the something is insignificant or trivial; given that these companies do “give back” via sham PR campaigns, finding an example here probably isn’t hard.)

The key in all of this is interposing something psychologically negative in the process of consuming (really, purchasing) oil. Since first hatching this notion, I’ve put together several designs for gas pump stickers, though I’m not quite happy with them yet. I’m planning on putting up a number of sticker designs at some point and getting them produced to try rolling this experiment out. More on that in a future post.

In general, this approach might work for some things and not for others. Adam pointed out to me that gas pumps may be among the better targets:

The really nice things (strategically speaking) about gas pumps is that they’re stationary, each one serves multiple consumers, and each consumer has to interact with them pretty closely. Obviously people can still operate a gas pump on auto-pilot, but they require a decent amount of attention as opposed to, say, meat.

Which makes (that is, meat does) an instructive contrast. Meat is found in lots of small packages, these packages turn over regularly, only one person can take home any given package, and it’s easy enough to just pick up a package and throw it in the shopping cart without looking too hard. So the parallel strategy for meat—putting stickers on it to say “this pound of beef generated 100 lbs of CO2” or whatever—would require a zillion stickers plus absolutely tenacious labor.

Obviously there are more intelligent ways to apply the strategy to meat; I’m just drawing the dumb analogy to map out two poles in the space of targets. And what I’m thinking of now is how to place other major targets (utilities for buildings (res. and commercial), air travel, industrial ag) in that space, so as to figure out whether the strategy can apply to them, and if so, how it has to be modified as the target varies along the relevant dimensions (mobile vs stationary, one-one vs one-many consumers, etc).

So we’re left with the question of whether the number I calculated above is significant. In that back of the envelope calculation, we’d save 2.5% of the nation’s consumption. Suppose we could increase awareness by 10% per year. That’d ensure a declining slope of consumption to perhaps match the rate of global oil production decline we’re likely to see in a couple of years.

Now there are a couple of minor wrap-up questions to consider. Aren’t most stickers made of plastic, so might this be a bad thing? Yes, they are made from oil (or natural gas) in general, but in the scheme of things consume an insignificant amount (stickers are a one-time cost each and would therefore use far less than even small-scale daily uses of plastic—just consider that 500 million disposable plastic straws are used every day in the United States). And finally, if the stickers become popular enough, won’t the purveyors of businesses such as gas stations catch on and just remove them? Maybe, but that for once would be a good problem to have.

Barath Raghavan

Barath Raghavan is a computer scientist who writes about the intersection of energy, environmental, and technological issues.

Tags: Consumption & Demand, Culture & Behavior, Education, Energy Policy, Fossil Fuels, Oil