Governments should more frequently publish CO2 emissions data: Leveraging human psychology to fight climate change

The most recent report by the International Panel on Climate Change (2018) states the danger very clearly: urgent action is required to avoid possible climate disaster. The primary cause of global climate change is greenhouse gas emissions, primarily CO2 emitted into the atmosphere in the burning of fossil fuels. We must drastically cut emissions at national and planet-wide scales in order to reduce the risk of worst-case scenarios to acceptable amounts.

The major problem in bringing about the required large-scale change is not technical or economic, but social. We know roughly what to do, and we can afford to do it. Where we urgently need progress however is in motivating human decisions (at scale) that are effective in decreasing CO2 emissions.

Along these lines, governments and the responsible media should frequently report and heavily publicize CO2 emissions data in much the same way that some countries currently release quarterly unemployment or economic statistics. This is a societal “nudge,” i.e. a measure meant to indirectly influence policy-making and behavioral choices (Thaler & Sunstein, 2008) that has the potential to cause large-scale CO2 emission reductions in the short term.

It may seem counterintuitive to think that simply reporting statistics differently could have much, if any impact. What difference could such a change make? As people who care about this issue often observe, even in countries where there is widespread acknowledgement that human induced climate change is happening, there is often a lack of political will to implement the sometimes costly changes required to combat the problem.

It is not hard to understand why. Imagine a government that raises the price of fuel by implementing a carbon tax (where the level of taxation correlates with the degree of emitted CO2). This is a measure that is widely recognized by government agencies, economists, and behavioral scientists as one that is likely to be effective in reducing emissions (Caron et al., 2018; Barron et al., 2018). The logic is that people will adjust their lifestyle choices and purchasing decisions (e.g. preferring fuel efficient cars over gas guzzlers) in the medium term to adapt to new market prices, a prediction supported by empirical studies (Li, Timmons, & Van Hoffman, 2009). However, raising the price of gas is not generally popular amongst people who drive cars, and there is thus a political cost to pay for implementing effective measures. For some recent evidence, see the French public’s backlash against Macron’s carbon tax (LCI, 2018).

Socially, a mechanism is needed that would improve the image of decision makers as they implement policies that are effective in reducing CO2 emissions (in the carbon tax example, there may be an additional need to fairly redistribute the increased income that the tax generates). Such a reputational mechanism could serve as counterweight to the economic costs required to cut emissions. Regular reporting of emissions data would fill this gap.

This proposal leverages two further powerful psychological principles. The first is that “you are what you measure” (Ariely, 2010).  Organizational psychologists have long argued that if an organization, for example, measures performance on a, b, and c but not x, y, and z, people will optimize their scores on a, b and c but not x, y, and z. If employee bonuses are dependent on measured customer satisfaction, employees will figure out a way to make their customers happier. If a CEO’s performance is judged on stock price, then they will do everything possible to improve stock price. And if forecasters are held accountable for accuracy, they become statistically better at predicting future outcomes (Chang et al., 2017). Similarly, if decision makers and countries are measured on CO2 emissions, they will be inclined to reduce them.

The second psychological principle that this proposal exploits is that short-term reward and punishment systems are generally more effective than long-term systems. In Nudge, Thaler and Sunstein recount a story of a newly hired assistant professor (“David”) who was yet to officially receive his PhD. David was faced with the tedious task of fulfilling the formal requirements of his PhD, including writing the doctoral thesis, instead of working on more exciting and newer projects associated with his new position. Despite plenty of incentive to complete his requirements (e.g. being treated as an instructor instead of assistant professor until the task is done and being able to put away less money towards retirement), he could not muster the motivation to finish.

Thaler then intervened by making a deal with David. David would write a series of $100 checks to Thaler, payable on the first day of each of the next few months. Thaler would cash the check if David did not complete a new chapter of his thesis during that month. Moreover, Thaler would use the money to throw a party to which David would not be invited. By shifting to this system of shorter-term penalties, David’s behavior immediately changed and he completed his PhD requirements on time. Similarly, by more frequently reporting CO2 emissions data (and thereby more frequently applying reputational pressure for effective decisions), decision makers should feel more short-term motivation to reduce emissions and thus should be more likely to act.

Clearly, publishing regular emissions data by will not, by itself, be enough to reduce emissions. However it promises to help focus attention on the central problem of emissions reductions and the effectiveness of specific policies that might otherwise be ideologically divisive, unpopular due to cost or counter-intuitive. Focusing attention on emissions in this way will be necessary to navigate many of the battles to come as we attempt to stave off the worst effects of global climate change.

Consider for example the recent IPCC report (2018) which argues that there is already too much CO2 in the atmosphere, so that even aggressively reducing emissions over the next few years will likely not be enough to limit warming and ward off the worst effects of climate change. According to the report it is crucial that, in addition to reducing emissions, governments invest heavily in “carbon neutral” and “carbon negative” technologies that involve capturing, storing, and recycling carbon (Carbon 180, 2018). Though these technologies are surprisingly advanced, costs have dropped significantly in recent years (Service, 2018), and industrial actors are poised to scale-up (Temple, 2018), no political leaders have opted to vocally take on a leadership role in this area. This is true even in a country like France, where the public is generally supportive of measures meant to combat climate change and its president has tried to position himself as global leader on the issue.

The lack of leadership may in part be explained by the fact that the costs of investing in such technologies would be visible and cognitively salient to voters, but the benefits (i.e. the likely impact in reducing emissions) would seem far more abstract. By putting into place regular reporting of emissions data, we can nudge our leaders into making better bets, like bets on carbon capture or other promising approaches. If we nudge enough important decisions in the right direction, these can add up to make a real difference in the fight to prevent the worst effects of climate change.



Ariely, D. (2010). You are what you measure. Harvard Business Review.

Barron, A.R., Fawcett, A.A., Hafstead, M.A.C., McFarland, J.R., Morris, A.C. (2018). “Policy Insights form the EMF 32 Study on U.S. Carbon Tax Scenarios.” Climate Change Economics 9(1):1840003.

Caron, J, J Cole, R Goettle IV, C Onda, J McFarland and J Woollacott (2018). Distributional implications of a national CO 2 tax in the U.S. across income classes and regions: A multimodel overview. Climate Change Economics, 9(1), 1840004.

Chang, W., Atanasov, P., Patil, S., Mellers, B., Tetlock, P. (2017). Accountability and adaptive performance under uncertainty: A long term view. Judgment and Decision Making, 12(6), 610-626.

Carbon 180 (2018). Cheat Sheet: What you need to know from the National Academy’s report on Carbon Removal. Medium.

LCI (2018).

Shanjun Li & Christopher Timmins & Roger H. von Haefen (2009). “How Do Gasoline Prices Affect Fleet Fuel Economy?,” American Economic Journal: Economic Policy, American Economic Association, vol. 1(2), pages 113-37, August.

Service, R. (2018). Costs plunge for capturing carbon dioxide from the air. Science. doi:10.1126/science.aau4107

Temple, J. (2018). The carbon capture era may finally be starting. MIT Technology review.

Thaler, R. & Sunstein, S. (2009). Nudge. Penguin Books. New York, NY


  • comment-avatar
    Radu Umbres 16 November 2018 (11:21)

    Making emissions data more relevant
    Thanks Brent for this great post and a really valuable idea.
    The only question is if simply CO2 levels by themselves are salient enough. Perhaps a more powerful tool is to track alongside these fairly abstract numbers, which make much sense for climatologists but perhaps less so for laymen, some other data. I am not an expert, but perhaps figures on the rise of planetary ocean levels, the likelihood of severe weather phenomena, the relative risk of species extinction etc could drive home the point better, especially for the wider public.

  • comment-avatar
    Brent Strickland 16 November 2018 (15:53)

    Question about relevance
    Hi Radu. Thanks for your comment. My bet (though this is perhaps worth following up on experimentally) is that publishing CO2 levels by themselves is the best bet. There are a couple of reasons for thinking this. First, they would likely be immediately comprehensible in some over-informed countries, or in less informed populations, they could become comprehensible. We could make the same “criticism” about unemployment data: too abstract, too statistical, etc… But the western world seems to have had very little trouble dealing with those, even if some people may have found them confusing when they first started being released. Secondly, and related to the last point, I think there is a need to more sharply focus public attention on C02 specifically. I find in France and in the US that among people who recognize that there is a general problem related to climate change and who agree that the problem is caused by humans, they are often deeply confused about what’s causing it. People will say all sorts of odd things are responsible for climate change: insecticides, plastic in the oceans, gmo’s, holes in the o-zone etc… This lack of focus on the appropriate underlying mechanism often leads them to support inefficient solutions to the problem, such increasing organic food output. My worry in publishing rises in sea level, drops in biodiversity, etc… is that this may actually worsen the causal focus problem, and not guide people towards making smart decisions politically or personally.

  • comment-avatar
    Hugo Mercier 20 November 2018 (06:27)

    Ways this could backfire?
    Thanks Brent! Very interesting idea, which I could imagine working. However, one could also imagine (at least) two ways in which it could backfire. The first, which you indirectly mention, is an issue of time frame. People pay more to fill up their car’s tank now, but the effects on CO2 emissions would be (1) delayed by years (2) likely impossible to discern among other factors.

    The second is that this might encourage the bons élèves to relax. For instance, thanks to nuclear power, France as a whole emits comparatively little CO2. Mightn’t looking at these numbers, and at those of other countries, make French people want to make less effort, thinking they’ve already done their part? (Obviously, this might be compensated by the opposite effect among the mauvais élèves). (Didn’t something like this happen with an intervention aimed at reducing electricity consumption? I only have a faint memory of this…)

  • comment-avatar
    Brent Strickland 20 November 2018 (17:40)

    Hi Hugo,

    Thanks for the comments, both of which are very reasonable. For the first point about timing, there is certainly a question about what’s optimal here. First I would point out that many measures could be taken which would have an immediate impact, for example, putting into place carbon capture and recycling at cement factories (who by the way are responsible for something like 5% of CO2 emissions worldwide if memory serves). If you do that tomorrow, you’ll see a result in the very short term. Secondly, it’s worth keeping in mind that frequent reporting can encourage medium term and even longer term decisions. The reason is that frequent reporting of X makes the variable collectively more salient, and therefore things that are good for X may count for more, even if the benefit is medium or long term. For example, despite regular reporting for unemployment data, Macron’s government put into place economic policies that they only expected to start showing true returns in 2-3 years. They’re essentially making the bet that the economic results will be visible by the time the next electoral cycle comes around. I don’t see any reason why people wouldn’t approach regularly reported CO2 stats in a similar way. For the causal remark I think my approach, if anything, would help isolate causally responsible actions. For example, imagine that France were to put into place a standard requiring 20% of all concrete used in buildings to be reinforced with captured carbon (companies like Solidia and Carbon Cure are already doing this). The policy may go into effect today, but we would only see a drop in emissions once LafargeHolcim and other cement producers increase their capture and recycling capacities. At that fairly precise moment in time we would know almost exactly how much of a drop to expect, and would likely be able to see the impact on CO2 emissions (which may be much harder if you report only once a year).

    For the second issue of motivation, yes they talked about something similar to this in the nudge book. Basically if you just showed who the “bon eleves” and “mauvais eleves” were, the bon eleves weren’t motivated to drop energy consumption (though this could be outweighed by the drop in energy consumption in the mauvais eleves, as you suggest). However the nudgers were able to figure out ways to keep the positive effect of showing people they were under the benchmark while eliminating the laziness effect for the overachievers. They did this by showing frowny faces for the underachievers and smily faces to overachievers (in addition to their quantitative position relative to a benchmark). Obviously emoticons for an entire country, emoticons would not work and even be a little silly, but my point here is that the way you frame the information could very likely be tweaked to work around this. For example, perhaps each country should only look at it’s past history and just focus on bettering itself, or perhaps things could be framed as a type of “championship” where countries strive to be the winner (e.g. in per capita yearly reductions).