Fracking, Toxics and Disclosure
The following is a guest blog by EO Technical Committee member, T. Robert Fetter, who conducted a research project as part of the work on responsible shale development supported by Viridian Energy’s Simply Right Natural Gas Fund.
In emerging technologies where stakeholders express concern about potential external damages, but the extent of those damages is uncertain, regulators frequently turn to information disclosure as a tool to encourage pollution-reducing innovation without being overly prescriptive. The low cost and ready availability of technology for collecting, disseminating, and analyzing data makes information-based regulations increasingly attractive and popular. However, it is unclear why and whether companies would change their operational practices in response to information-disclosure requirements. There is limited empirical analysis on the causal relationship between disclosure regulations and changes in companies’ behavior, especially in non-consumer-facing settings.
This study analyzes how disclosure regulations affect firms’ use of toxic additives in hydraulic fracturing. Many state laws now require the public disclosure of the chemicals used, but these state laws were passed at different times. This study uses the differences in regulatory timing as a “natural experiment” and examines whether companies changed fluid formulas earlier in states that passed disclosure regulations first. The results indicate that companies did reduce the toxicity of fracturing fluids earlier in states with earlier regulations, implying that the regulations caused reductions in toxics use, even after controlling for many other factors including the separate effects of time and technological progress.
The chemicals used in hydraulic fracturing operations, some of which are toxic, have been a matter of significant public concern, particularly for wells in close proximity to homes and other sensitive land uses. Regulators and the public have expressed concerns about potential dangers associated with manufacture, transport, use and disposal of the chemicals involved. While the industry generally maintains there is low risk from chemicals used in the process, some operators and vendors have moved toward using less-toxic chemicals as part of “green completion” technology.
Even in the absence of mandatory disclosure regulations, many operators have voluntarily provided information to the public about the chemicals they used. This study uses the voluntary reports to measure the toxic chemicals that companies used prior to mandatory disclosure. While this reflects the best data available, it makes the causal effect of mandatory disclosure difficult to measure. When companies voluntarily choose what fractures they report chemicals for, economic theory suggests they are more likely to report the subset of fractures in which they used lower-toxic formulas. Thus, moving from voluntary to mandatory disclosure might appear to cause an increase in toxics use, and any reduction in toxics use caused by the disclosure law – for instance, due to pressure from external or internal stakeholders – may be outweighed by the apparent increase.
This study teases apart these countervailing effects by analyzing a subset of the data that is limited to wells operated by “frequent voluntary reporter” operators – that is, operators that voluntarily reported chemicals for a large proportion of their wells before disclosure became mandatory. The impact of full reporting, which otherwise might mask the impact of stakeholder pressure, is lower for these operators since they reported chemicals for a larger proportion of their wells before mandatory disclosure. In addition, the study uses data on one state (Pennsylvania) from a period in which companies were required to report chemicals for all wells to the state, but in a form that was relatively inaccessible to public review, at the same time as some companies voluntarily reported data for some wells to a much more accessible and high-profile website. Evidence from these simultaneous reports supports the hypothesis that – on average – when companies can choose whether to report chemicals for individual frac operations, they choose the ones with less toxic chemicals.
The analysis focuses primarily on companies’ use of toxic chemicals that are regulated by the Safe Drinking Water Act or the Clean Water Act, or classified as diesel fuel. The study analyzes the effects of mandatory disclosure on four quantities: the total concentration of toxic chemicals within fracturing fluids, the total quantity of toxic chemicals adjusted for differing levels of toxicity among chemicals, the quantity of chemicals that are frequently mentioned in media reports on hydraulic fracturing risks, and the quantity of chemicals declared “proprietary.” The evidence indicates companies reduced their use of toxic chemicals (the first two measures), but finds no evidence that companies changed their use of high-media-profile chemicals. Taken together, these findings suggest that companies prioritized reductions in toxic chemical use, without regard to whether chemicals were frequently mentioned by the media in conjunction with potential risks of fracturing activities.
All of the state disclosure laws covered by the study allow companies to declare some chemicals as proprietary in nature and therefore avoid releasing information about their chemical identities, although states require differing levels of documentation in order for companies to make this claim. The study finds that companies gradually increased their use of these “proprietary” claims over the same period of time that state disclosure laws came into effect. However, the increased use of these claims rose in both kinds of states – those where disclosure was mandatory, and those where it was still voluntary. In other words, there is no evidence that the disclosure laws caused the increase in “proprietary” claims.
- Operators responded to mandatory disclosure laws by using fracture designs with lower concentrations of toxic chemicals. On average, mandatory disclosure laws resulted in about a 15 percent decrease in the average concentration of toxic chemicals used per fracture, after controlling for a variety of factors. Much larger reductions, on the order of 70 percent or more, occurred about one year after mandatory disclosure laws took effect.
- The companies that respond most to mandatory disclosure may be those that are also most likely to voluntarily report more information. To distinguish the effect of full reporting from the effect of disclosure-related pressure, the study isolated the responses of “frequent voluntary reporters” to the mandatory disclosure regulation. The study was not able to discern a separate effect for operators that are less-frequent voluntary reporters.
- At the same time more states passed disclosure regulations, operators declared a larger share of their fracturing fluids “proprietary”. However, there is no evidence that this behavior was caused by disclosure laws; instead, it appears to be on an independent time trend, perhaps due to increased innovation or greater attention.
The full paper is available at https://sites.duke.edu/trfetter/research/.