Back in February, a climate advocacy organization was pressuring the International Energy Agency (IEA) to include recommendations in its policy analyses for how to transition away from fossil fuels. While it is totally expected that such groups will engage in advocacy aligned with their cause, what they may not realize is that those actions can inadvertently hurt their own causes by undermining the credibility of impartial scorekeepers.

The IEA is a multinational body that performs energy-related analysis for the Organization for Economic Cooperation and Development (OECD). OECD member nations are generally rich and synonymous with terms like “the West” or “global North.” The IEA originated during the 1970s oil crises with the objective of furnishing policymakers with the information needed to make effective energy policy decisions, based on empirical data. Since then, the IEA has fulfilled an important role by providing quality data analysis. Highlights include reports like one focused on critical minerals and data on big and small energy industry developments alike.

But the IEA has also delved into topics that are more controversial, namely laying out what the world “must” do to achieve global climate targets. The quality of the IEA’s analyses on these topics is strong, but they can wrinkle feathers because they start to veer into recommendations that rely more on analyst preferences rather than staying within an objectively nonpartisan, unbiased framework. The unfortunate effect of this is that it can reduce the IEA’s credibility with precisely the skeptical audiences that they are more capable of reaching than issue advocacy groups are.

The U.S. version of the IEA is the Energy Information Administration (EIA), a strictly nonpartisan organization whose sole job is to provide data analysis in as credible a fashion as possible. The EIA has its share of successes and failures, but every analysis the EIA puts out is highly respected because it is simply a straight data presentation with limited commentary. To the extent that policy developments are discussed in these analyses, they are only mentioned to explain how the policy increases or decreases effects observed in data, without judgment as to what is good or bad. As a result, the EIA has maintained strong credibility with both the right and the left.

As a counterpoint, the Environmental Protection Agency (EPA) is managed by political appointees, and even though it issues data analysis with every major rule the agency proposes, these are almost guaranteed to be panned by the opposing political party every time. When the EIA put out an analysis on the Clean Power Plan, it was taken seriously on both sides of the aisle, but the EPA’s analysis was praised or critiqued along party lines.

Ultimately, institutional credibility matters. Once an organization starts to lose credibility, it is very hard to win it back. If the IEA followed the requests of issue advocates, the effect would not be for the IEA to convince skeptic audiences that issue advocacy groups were right, it would just be to make skeptics no longer trust the IEA. If policymakers retrench into only believing “their” side of analysis, and no unbiased central scorekeeper exists, the quality of debates will erode, and policy effects will suffer. A real-world example of this is seen in the widespread mistrust of the Centers for Disease Control and Prevention recommendations related to COVID-19, which reflected mistrust of politicians’ posturing on an issue that should be apolitical.

Ultimately, it’s important for policymakers not to be surrounded by yes-men. Part of that means accepting that sometimes credible institutions are going to say things we disagree with. Keeping institutions like the IEA credible means acknowledging that they are not an extension of efforts to expand our personal worldviews. In a word, it is humility that is missing from modern policy debates, and we would be better served by more of it instead of less.

Every Friday we take a complicated energy policy idea and bring it to the 101 level.