Skip to content

Shouldn’t We Weight?

Accounting for equity in social cost calculations has huge consequences. 

I have been really uncool for most of my life and continue to be proud of that fact. All the “cool” kids on twitter these days are railing about how carbon pricing has failed (which it hasn’t), and how the social cost of carbon is useless (it isn’t). Well, I don’t care. Mullets were briefly considered to be cool again, and I continue to maintain that they were a #haircutfail. The social costs of carbon and methane will continue to play an important part in helping us figure out which policies pass a benefit cost test and which ones don’t. This has to be an integral part of our road to net zero, no matter whether it makes your mullet curl or not.

The literature on how to calculate the social cost of carbon is slowly evolving in some non-mainstream field journals in economics, but much more rapidly in the general science literature. This is good, as there is real climate science involved in the calculation of these numbers and economists left alone will gladly violate one or multiple laws of thermodynamics without knowing they did. 

While the Social Cost of Carbon has been getting all the attention, its little sibling – the Social Cost of Methane (SCM) – has been one of the big “unknowns” in this literature. EPA had cobbled together a number, which stood up to scrutiny, but a recent publication by a group of Berkeley folks in and around the lab led by David Anthoff in the journal of journals – Nature – just engaged the Warp drive. 

Here’s a quick refresher. Methane is a potent greenhouse gas that comes from both ends of a cow, leaky natural gas pipelines, landfills and wastewater treatment, to name but a few anthropogenic sources. It is invisible and a much more potent greenhouse gas than its famous cousin- CO2. Its lifespan, once released, is also much shorter. For those of you that run, in terms of performance as a greenhouse gas, this is similar to comparing Usain Bolt to Eliud Kipchoge.

The above mentioned paper makes some significant headway in the modelling of the physical impacts of methane on the climate system, which to this guy is important but about as exciting as watching grass grow, because I lack the training and appreciation for it. Specifically, they update the model to account for the recent 25% upward revision of the radiative forcing of methane. They also do a bunch of really cool new modelling stuff around hindcasting (checking how well your model predicts the past) and built in a Bayesian framework – nerd alert!

But what got my interest is what the authors do with respect to Equity Weighting (EW DAVID!!!!).  What is Equity Weighting you ask? Imagine a world with rich people and poor people. Newsflash, you live in this world and the inequality is getting worse by the day! The thought experiment is one where a given consumption loss (from climate change for example) causes a bigger loss in well-being to a poor person than a rich person. Equity weighting recognizes this and assigns a higher weight to climate damages occurring to low income folks (actually, low income regions like sub-Saharan Africa). 

Is this crazy Berkeley kumbaya stuff? No! This is consistent with standard economic theory. Also, is this some absurd tool, no one uses? Nein! The German government actually employs an equity weighted social cost of carbon. 

How do you do this? It’s actually non trivial and really cool. In a setting with uncertainty (such as settings where the future is involved), you can play with a parameter called the “normative parameter of inequality aversion”, which sort of determines how big the relative losses in welfare are to individuals with different levels of wealth from a reduction in consumption. Or in plain language, it determines how much more screwed you are in dollar terms from a consumption loss if you’re poor versus if you’re rich. So what the model does is that it calculates the marginal damages from one additional ton of methane in each region and then assigns more weight on the damages in low income regions when we add them all up – for a variety of values of  this “normative parameter of inequality aversion”. Once you do this, you can calculate an “equity weighted SCM”. 

There is one more super nerdy choice the modeller needs to make, which relates to normalization (or in econ terms, picking the numeraire good). If you pick consumption in high-income areas as the numeraire good, you get a higher $ number for the SCM. So, when we express the SC-CH4 as the welfare equivalent loss of consumption in the US we get a higher number than if we express it as the welfare equivalent consumption loss in Sub-Saharan Africa. So if you read the paper and you see “US numbers” and “Sub Saharan Africa numbers”, David Anthoff reminded me on twitter that these US numbers are NOT “damages that occur in the US” but rather “world damages expressed as the welfare equivalent consumption loss in the US”.

So what happens if you do this. It turns out a lot! Accounting for these societal concerns about equity results in SCM numbers that differ by more than an order of magnitude (fancy speak for ten-fold) between low- and high-income regions! Why is this important? If we take a single number for the entire world, we would compare the cost of emissions to the same number – no matter whether you are in the rich US of A or the not so rich countries of Sub Saharan Africa. Equity weighting creates heterogeneity here, which results in higher-cost opportunities for methane reduction being the right thing to do for a rich country, because the dollars spent on it don’t hurt as much.

So why should we care about this? My mullet wearing friends may like this approach as equity weighting generates higher numbers for the SCM. But we should aspire to be more correct, rather than simply seeking a higher number. This paper certainly does that. It is interdisciplinary science at its very best. For the super nerds here, check out the supplemental information… It’s amazing work. 
But to me, the most convincing argument is simply an ethical one, where an equivalent loss in consumption to a poor person is more damaging to a poor person than a rich person. Benefit cost analyses should incorporate this IMHO. If one had higher resolution models, calculating a equity weighted global SCM would also create a relative benefit for the poorer areas of the US that are also likely to see the largest climate impacts. Before I wrote this blog post, I thought about consulting a philosopher about this, but then recalled that David Anthoff is one by training (he turned to economics later in life!).

Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.

Suggested citation: Auffhammer, Maximilian. “Shouldn’t We Weight?” Energy Institute Blog, UC Berkeley, May 17, 2021,

Maximilian Auffhammer View All

Maximilian Auffhammer is the George Pardee Professor of International Sustainable Development at the University of California Berkeley. His fields of expertise are environmental and energy economics, with a specific focus on the impacts and regulation of climate change and air pollution.

8 thoughts on “Shouldn’t We Weight? Leave a comment

  1. Nice post, and yes, equity matters. But does Germany really use the SCC concept? And then, even equity weighted? Totally new to me but possible. Would be curious to learn more about this.

  2. [] One innovative carbon removal startup is putting ‘oil’ back underground. | Grist Shaun Meehan was burned out. It had been about a year and a half since he left his job at an aerospace company to found a clean energy startup in San Francisco with three of his friends, but the …


  3. Maximilian, I’ve always been one of those uncool kids too. So from one to another: could you provide a link or links to any studies that show carbon pricing can be considered a success? Because from what I’ve seen, the cool kids on Twitter have a point.

    “The social costs of carbon and methane will continue to play an important part in helping us figure out which policies pass a benefit cost test and which ones don’t.”

    Social costs must be compared not in isolation, but summed with the costs of ameliorating them. A question which is likely to be unpopular in California, where renewables are all the rage: are those in low-income communities more impacted by high electricity prices than the effects of climate change? Whether from cross-subsidies, or new-home prices elevated by mandatory solar panels, or nuclear decommissioning costs – adoption of renewables in California has unquestionably raised the price of electricity, with low-income communities bearing most of the burden.

    • Carl
      Let’s start at the beginning in California that led us to this place: nuclear power plants in California have unquestionably raised the price electricity. The capital cost for Diablo Canyon representative of a 900% cost overrun and a 50% additional toll created by the 1995 Settlement Agreement composes almost 50% of the above market costs in PG&E’s portfolio. Although SONGS was quite as egregious, it also laid the groundwork first for the QF Gold Rush in the 1980s and then the AB 1890 in 1996.

      • “Let’s start at the beginning in California that led us to this place…”

        You can start wherever you like Richard, but I refuse to accept ridiculous claims like “nuclear power plants in California have unquestionably raised the price electricity”, that ratepayers are “paying twice for Diablo Canyon”, that there was a “900% cost overrun”, etc., for the simple reason they have no basis in fact. If you want to provide specific links in support I’ll consider them. But again, I need to remind you of a fundamental condition of online debate: it’s a claimant’s job to provide support for his argument. Without it, he has only as much credibility as another dude on the internet.

        Among claims on your website I can immediately refute is the following: “Diablo Canyon was infamous for increasing in cost by more than ten-fold from the initial investment to coming on line…”.

        Richard J. Gilbert, Berkeley’s own Emeritus Professor of Economics, Chair of the Department of Economics from 2002 to 2005, and current Chair of the Berkeley Competition Policy Center, would disagree with you. In “An Economic Evaluation of the Costs and Benefits of Diablo Canyon”, he and co-author Alan J. Cox wrote:

        “Construction of the plant began in 1966 with units 1 and 2 scheduled for completion in 1972 and 1974. The original total cost was anticipated to be about $1 billion (1985 dollars)…the final book cost at completion was $5.67 billion.”

        To justify your ten-fold increase you would have to compare nominal instead of constant costs, a mistake made by freshman economics majors. You weren’t doing that, were you?

        Now – about PG&E’s exaggerated costs for Once-Through-Cooling (OTC) mitigation, which they claimed justified the shutdown of Diablo Canyon:

        • Both PG&E’s Proposal and its Testimony ignored that framework for an OTC mitigation settlement was already negotiated and focused on land conservation and artificial reef.

        • In 2000, the Central Coast Regional Water Quality Board created the framework for an OTC settlement with PG&E. Michael Thomas from Board oversaw the process, and hired Peter Raimundi from UC-Santa Cruz who worked with PG&E consultant John Steinbeck of Tenera Consultants (confirmed by personal interviews with all three).

        • The Regional Water Board – not the State Water Board, as PG&E claimed – decides on OTC compliance. Per Michael Thomas: “Both boards have a role, but the Regional Water is who decides whether to adopt cooling towers.” California has deferred to Raimondi, Thomas and Steinbeck.

        • According to Raimondi’s presentation to the State Water Board, and based on interviews with Steinbeck and Thomas: “An artificial reef of sufficient size and with appropriate design and placement could compensate for the majority of impacts associated with entrainment at DCPP….The estimated cost for the construction of an artificial reef ranged from 15 million to 50 million dollars.”

        • All the parties rejected cooling towers, including the Water Board. Thomas: “I don’t think they are feasible or optimal. There have been multiple studies for towers that aren’t feasible. We hired our own consultants separate from PG&E and they came to same conclusion.”

  4. “just engaged the Warp drive”

    Besides faster-than-light travel, another common feature in Star Trek is time-travel.

    Equity weights were introduced into the climate debate by Azar and Sterner (1996, Ecological Economics) and independently (trust me) by Fankhauser, Tol and Pearce (1997, Environmental and Resource Economics). David told me at the time that equity weight we’re a 70s thing.

    Anthoff learned about equity weights during his master’s, and worked on two extension (Anthoff, Cameron and Tol, 2009, Ecological Economics; Anthoff and Tol, 2010, Journal of Environmental Economics and Management)

    I’m glad that equity weights are now prominently discussed in a widely read journal, but let’s not pretend this is something new.

  5. Very interesting, and an approach that we in the European Commission are considering for the international aspects of our methane abatement work. Specifically, this has important outcomes in agriculture and in state stability and this work should be developed in these contexts.

    Always open to dialogue on these important topics.

    Brendan Devlin
    European Commission

%d bloggers like this: