Among EA-minded people interested in preventing climate change, it seems Clean Air Task Force (CATF) is seen very favorably. Why? The “Climate Change Cause Area Report” by Founders Pledge (PDF) gives an overview.
CATF’s work is introduced as follows:
“It was founded in 1996 with the aim of enacting federal policy reducing the air pollution caused by American coal-fired power plants. This campaign has been highly successful and has been a contributing factor to the retirement of a large portion of the US coal fleet.” (p. 5)
On p. 88, you will read:
“Do they have a a good track record? CATF have conceived of and led several successful advocacy campaigns in the US, which have had very large public health and environmental benefits. According to our rough model, through their past work, they have averted a tonne of CO 2 e for around $1.
Is their future work cost- - effective? Going forward, CATF plans to continue its work on power plant regulation and to advocate for policy support for innovative but neglected low carbon technologies.
Given their track record and the nature of their future projects, we think it is likely that a donation to CATF would avert a tonne of CO 2 e for $0.10-$1.”
On p. 91:
“CATF was founded in 1996 to advocate for regulation of the damaging air pollution produced by the US coal fleet, initially focusing on sulphur dioxide (SO 2 ) and nitrogen oxide (NO x ). They later advocated for controls on mercury emissions. The theory of change was that the cost of emission controls for conventional pollutants and mercury would result in the retirement or curtailment of coal plant operation resulting in reductions in CO 2 (and other) emissions. CATF conceived of the campaign goal, designed the strategy, and led the campaign, in turn drawing in philanthropic support and recruiting other environmental NGOs to the campaign.”
Where do the numbers come from? The spreadsheet states “subjective input” in several cells. The “Climate Change Cause Area Report” by Founders Pledge (p. 129--) states that “CATF is typical of research and policy advocacy organisations in that it has worked on heterogeneous projects. This makes it difficult to evaluate all of CATF’s past work, as this would require us to assess their counterfactual impact in a range of different contexts in which numerous actors are pushing for the same outcome.” The report then asks e.g. how much CATF “brought the relevant regulation forward”, and the answers seem to rely strongly on assessment by CATF. Nonetheless, it makes assessments like “Our very rough realistic estimate is therefore that CATF brought the relevant regulation forward by 12 months. The 90% confidence interval around this estimate is 6 months to 2 years.” On p. 91 you can read: “Through each of these mechanisms, CATF increased the probability that regulation was introduced earlier in time. Our highly uncertain realistic estimate is that through their work, CATF brought regulation on US coal plants forward by 18 months, with a lower bound of 9 months and a higher bound of 4 years. CATF believe this to be a major underestimate, and have told us that they could have brought the relevant regulation forward by ten years.”
While of course it’s fine to give subjective estimates, they should be taken with a grain of salt. It seems the comparison is much more reliant on such subjectivity than when you evaluate charities with concrete, repeatedly applied health interventions.
What, if anything, could be biased?
Additional to the (probably unavoidable) reliance on self-information, the following paragraph made me wonder:
“CATF have told us that at the time the campaign was conceived, major environmental organisations were opposed to reopening the question of plant emissions after the Clean Act Amendments of 1990, as they feared the possibility that legislative debate would unravel other parts of the Act. 216 This is based on conversations at the time with the American Lung Association, Environmental Defense Fund, and the Natural Resources Defense Council.”
How can we know whether such fears were justified ex ante? How do we guard against survivorship or hindsight bias?
Among EA-minded people interested in preventing climate change, it seems Clean Air Task Force (CATF) is seen very favorably. Why? The “Climate Change Cause Area Report” by Founders Pledge (PDF) gives an overview.
CATF’s work is introduced as follows:
“It was founded in 1996 with the aim of enacting federal policy reducing the air pollution caused by American coal-fired power plants. This campaign has been highly successful and has been a contributing factor to the retirement of a large portion of the US coal fleet.” (p. 5)
On p. 88, you will read:
“Do they have a a good track record? CATF have conceived of and led several successful advocacy campaigns in the US, which have had very large public health and environmental benefits. According to our rough model, through their past work, they have averted a tonne of CO 2 e for around $1.
Is their future work cost- - effective? Going forward, CATF plans to continue its work on power plant regulation and to advocate for policy support for innovative but neglected low carbon technologies.
Given their track record and the nature of their future projects, we think it is likely that a donation to CATF would avert a tonne of CO 2 e for $0.10-$1.”
On p. 91:
“CATF was founded in 1996 to advocate for regulation of the damaging air pollution produced by the US coal fleet, initially focusing on sulphur dioxide (SO 2 ) and nitrogen oxide (NO x ). They later advocated for controls on mercury emissions. The theory of change was that the cost of emission controls for conventional pollutants and mercury would result in the retirement or curtailment of coal plant operation resulting in reductions in CO 2 (and other) emissions. CATF conceived of the campaign goal, designed the strategy, and led the campaign, in turn drawing in philanthropic support and recruiting other environmental NGOs to the campaign.”
How does the evaluation work? A spreadsheet with an evaluation shows benefits of the policy impact.
Where do the numbers come from? The spreadsheet states “subjective input” in several cells. The “Climate Change Cause Area Report” by Founders Pledge (p. 129--) states that “CATF is typical of research and policy advocacy organisations in that it has worked on heterogeneous projects. This makes it difficult to evaluate all of CATF’s past work, as this would require us to assess their counterfactual impact in a range of different contexts in which numerous actors are pushing for the same outcome.” The report then asks e.g. how much CATF “brought the relevant regulation forward”, and the answers seem to rely strongly on assessment by CATF. Nonetheless, it makes assessments like “Our very rough realistic estimate is therefore that CATF brought the relevant regulation forward by 12 months. The 90% confidence interval around this estimate is 6 months to 2 years.” On p. 91 you can read: “Through each of these mechanisms, CATF increased the probability that regulation was introduced earlier in time. Our highly uncertain realistic estimate is that through their work, CATF brought regulation on US coal plants forward by 18 months, with a lower bound of 9 months and a higher bound of 4 years. CATF believe this to be a major underestimate, and have told us that they could have brought the relevant regulation forward by ten years.”
While of course it’s fine to give subjective estimates, they should be taken with a grain of salt. It seems the comparison is much more reliant on such subjectivity than when you evaluate charities with concrete, repeatedly applied health interventions.
What, if anything, could be biased?
Additional to the (probably unavoidable) reliance on self-information, the following paragraph made me wonder:
“CATF have told us that at the time the campaign was conceived, major environmental organisations were opposed to reopening the question of plant emissions after the Clean Act Amendments of 1990, as they feared the possibility that legislative debate would unravel other parts of the Act. 216 This is based on conversations at the time with the American Lung Association, Environmental Defense Fund, and the Natural Resources Defense Council.”
How can we know whether such fears were justified ex ante? How do we guard against survivorship or hindsight bias?