When Analysts Fail to Show Their Work, the Fossil Fuel Industry Benefits
By Dr. Lindsey Gulden
A recent carbon capture and storage analysis has me reflecting on the importance of analysts and researchers embracing transparency to avoid misleading decision-makers and the public. I’m a data scientist working to accelerate the energy transition and increase corporate accountability, and transparency is fundamental to both efforts.
Across technical fields like economics, public health, and climate science, it’s all too easy – and becoming all too common – to mislead journalists and decision-makers alike, to say nothing of the general public. Sometimes the misleading is accidental; sometimes it’s not. As someone who’s spent a career developing and evaluating technical models of all sorts – climate models, economic models, crop-production models, and commodity-pricing models – I am concerned that a mundane but essential component of trustworthy, high-quality, model-based analysis is often lacking: transparency.
To mislead with a complex model, an expert must simply fail to fully disclose the model settings and data that they fed the model. Experts need not voice a lie; they only need to avoid making their work freely and fully reproducible by someone else.
The May 2024 Rhodium Group report on the future carbon capture and storage (CCS) for U.S. heavy industry epitomizes this problem. Given carbon capture’s poor track record and the increasing skepticism regarding its value beyond increasing oil production via enhanced oil recovery (EOR), I was surprised to see Rhodium forecast substantial adoption of CCS by 2040. I dug into the details to learn more. What I found was effectively a black-box model used to justify conclusions that are both wildly optimistic and particularly misleading with regard to carbon capture’s role in climate-change mitigation.
To assess the quality of Rhodium’s conclusions, we mirrored and then extended Rhodium’s analysis using a cash-flow model consistent with their limited descriptions (work detailed here). We found that, when a full breadth of reasonable future scenarios is considered, CCS is only consistently profitable when it is coupled with natural gas processing facilities and used for EOR, a technique that squeezes more oil from existing reservoirs.
When CCS-EOR is coupled with natural gas processing facilities, ancient, naturallypreviously stored CO2 is piped to the surface with other gasses. Some of this geologicancient CO2 is vented directly to the atmosphere; the rest is reinjected back underground into aging oil reservoirs to flush out more oil. Not least because each ton of injected CO2 increases petroleum extraction by about three barrels of oil, CCS-EOR for any industry–but especially natural gas processing–is net carbon positive. The overall process emits more CO2 into the atmosphere than is injected underground. fThe result: the only CCS implementation that makes business sense is the version of CCS that most damages the environment.
Despite the fact that only CCS with geologic storage can ever be net carbon negative, Rhodium fails to specify whether their forecasted substantial adoption of CCS across heavy industry is CCS with geologic storage or CCS-EOR. Rhodium also fails to fully specify the model, data, and assumptions used to arrive at its conclusions. We found that Rhodium’s scenarios were peppered with unrealistic or unspecified assumptions. For example: all cases, including the one with ostensibly ‘low’ oil prices, assumed oil prices that were consistently much higher than the historical average price. Use of anomalously high oil prices substantially increases the rosy appearance of a rosy easy profits after subsidiesorfuture for CCS-EOR. As with other technical analysis, these conclusions are too easily misunderstood, co-opted and misused, often by industries – like the fossil fuel industry – to mislead decision-makers and the public.
In fact, Rhodium made very similar an wildly optimistic projections about the role of CCS in 15 years ago, wildly optimistic, that were widely cited at the time to promote CCS. These projections proved entirely wrong. You would think they might ask what the got wrong then, before doing it again?
Lack of transparency is becoming an increasingly common problem across studies cited in public policy debates, and it is often used by industries and special interests to cloak their objectives in a veil of technical legitimacy. Drilled Media reported on the role of respected consultancies in helping to legitimize otherwise suspect conclusions regarding the fossil fuel industry’s role in the energy transition. Drilled Media highlighted the multinational consulting group ICF and its lack of transparency in studies produced for petroleum-industry trade groups such as the National Ocean Industries Association and the American Gas Association.
Given the financial incentives of petroleum-industry trade groups, one might reasonably conclude that the lack of full transparency in these studies is purposeful deception. Without independent parties being able to fully inspect and stress-test the models, data, and assumptions used to generate the reports, oil and gas interests cite the reports’ questionable conclusions with impunity. All too often, neither journalists nor decision-makers are able to understand the studies, let alone refute their methods and conclusions.
Demanding transparency is especially important in matters of public policy and the public interest. Full disclosure of reports’ data and methods should be a minimum bar that all research and analysis must clear to be deemed credible and trustworthy. Those entrusted to protect the public interest must demand transparency from the authors of complex technical analyses such that others are able to vet the work. Citing reports whose methods and data are cloaked (intentionally or not) in obscurity is neither good technical practice nor in the public interest. If a report’s authors have not made clear and publicly available all code of their model, all input data, all model settings, and all of their assumptions, we can’t fully trust their conclusions. The work may be sound, but it merits a closer look before we write the headline.
Lindsey Gulden is an independent data scientist and a member of the Science Roundtable on Carbon Capture. She worked for ExxonMobil for a decade and is now working to advance climate action.