Journalists like me often seek out academics for comment and insight on stories related to the energy transition, since these professors have often done in-depth research into various fuel sources and their impacts. The hope is that these sources are relatively unbiased; their loyalty is to the data. But a study published Thursday in Nature Climate Change found that prominent energy policy centres at top-tier universities that are funded by the fossil fuel industry may produce content more favourable to dirty energy than other, similar centres. This is concerning, because it’s not just journalists who seek the council of these academics — it’s policymakers, too.
“Reports by fossil-funded [centres] are more favourable towards natural gas than towards renewable energy, while centres less dependent on fossil fuel industry funding show a pro-renewable energy preference,” Anna Papp, a PhD student in Sustainable Development at Columbia University and one of the authors of the paper, told Earther in an email.
Academic centres focused on energy research have become an increasingly respected and important voice in energy policy conversations, as the U.S. and the world begin grinding the gears on the energy transition. Representatives from places like Columbia’s Centre on Global Energy Policy and MIT’s Energy Initiative have testified in Congress and are often featured on television as experts; some of their reports have even been the subject of their own Congressional hearings. But several of the most prominent academic think tanks working on energy issues also have significant funding from the fossil fuel industry. Columbia’s Centre on Global Energy Policy, for instance, lists its financial partnerships on its website, which include big fossil fuel names like BP, ConocoPhillips, ExxonMobil, Chevron, and Occidental Petroleum. (Full disclosure: While I was employed at a PR firm between 2014 and 2016, Columbia University’s Centre on Global Energy Policy was a client; I worked on some of their press needs and materials.) What’s more, much of the research and whitepapers produced by these centres does not undergo the peer review process that a scientific paper may receive.
“Given longstanding concerns about the objectivity of corporate-funded research — for example, biomedical research — we wanted to better understand industry-funded research in the context of climate change,” Papp said.
Papp and her coauthors set out to see if the reports and materials put out by academic centres who did disclose their fossil fuel funding were different from the centres that had no fossil fuel funding or which did not prominently feature that funding. To more accurately capture attitudes from these energy centres toward certain topics, Papp and her colleagues used a machine learning approach known as “text as data.”
“‘Text as data’ algorithms convert written text to data that can be analysed quantitatively,” Papp said. “A human reader forms an opinion about the sentiment of sentences or paragraphs, e.g., how positive or negative the text is. Of course, labelling sentences manually is extremely time-consuming and subjective. Sentiment analysis tools attempt to replicate this process computationally and quantify emotions contained in text.”
Using a sentiment analysis tool, Papp and her fellow researchers compiled 1,706 research reports, consisting of more than one million sentences, from 26 energy research centres at universities based in the U.S., UK, and Canada published between 2009 and 2020. They chose to focus the analysis on academic centres’ sentiment towards natural gas.
While natural gas has been promoted in the past as a “bridge fuel” between coal and oil and renewables, the massive amount of methane involved in its production has meant that its former climate-friendly branding is coming under serious scrutiny. Despite research showing that the world needs to cease all new fossil fuel exploration immediately in order to keep the world from warming more than 1.5 degrees Celsius, many oil and gas majors continue to promote natural gas as part of a climate solution.
“Natural gas is now the U.S.’s largest energy source for electricity generation,” Papp said. “All major fossil fuel companies produce it. So natural gas is very policy-relevant.”
The researchers found that reports from the three prominently fossil fuel-funded centres in the analysis at Columbia, MIT, and at Stanford’s Precourt Institute for Energy were “more favourable” toward natural gas than renewable energy. The positive sentiment in content from these three centres was “indistinguishable,” the paper found, from content produced by the American Gas Foundation and the American Gas Association, industry groups “whose explicit purpose is to promote the gas industry.” The 23 energy centres that did not prominently advertise fossil fuel funders, meanwhile, produced materials that were favourable to renewable energy and hydropower, and more “neutral” toward natural gas.
“Our scholars carry out their research by following facts and evidence wherever they lead, independently and free from any influence or control by funders or other interest groups,” the Columbia Centre on Global Energy Policy’s website states. “They adhere to strict conflict of interest policies established by Columbia University to protect against real or perceived risks to the integrity of research.” A spokesperson from Columbia’s Centre on Global Energy Policy said that the staff of the Centre had not had time to review the paper but that they had confidence in their scholars and their independent research.
A spokesperson from MIT’s Energy Initiative said they had not seen the paper and had no comment. “Our research reports are the work of MIT faculty, staff and students with no influence–no approval or rejection, no oversight, no opportunity to accept or reject any findings — from any funders, whether MITEI members or not,” they said in a statement.
Stanford’s Precourt Institute for Energy did not get back to us. We’ll update this story if they do.
One of the main challenges of this research is that the three centres identified as the most positive toward natural gas are simply the ones who disclose their prominent relationships with fossil fuel funders. Papp said it was difficult to find information on funding from many centres’ public information sources. It’s not out of the question that fossil fuel companies may be giving money to other energy centres in this study who simply did not disclose the funding publicly. And of the three centres at MIT, Stanford, and Columbia that do disclose larger-scale funding, the research found that less than 25% of the reports produced by these centres between 2009 and 2020 had explicit funding acknowledgements. With the nebulous nature of universities disclosing financial interests and without the guardrails built in by peer-reviewed research, it’s tough to figure out who is paying for the perspectives presented by university energy centres — even as the work they put out becomes more and more important to policymakers.
“The academically-branded but non peer-reviewed research of these energy centres are increasingly used for policy making, but without appropriate disclosures it is difficult for the public and policymakers to determine whether there may be a threat to the impartiality of these products,” Papp said.