Environmental Data on the Internet: A Wired Public Setting Environmental Policy

Peter L. Gray

The author is a partner in the Washington, D.C., office of McKenna & Cuneo, where he specializes in environmental regulation and litigation. He may be contacted at peter_gray@mckennacuneo.com.

[30 ELR 10122]

During its first 15 years of existence, the U.S. Environmental Protection Agency (EPA) sought to improve environmental quality through conventional "end-of-pipe" command and control regulation. In the late 1980s, EPA shifted to a new paradigm in environmental protection: to enhance environmental protection by encouraging "voluntary" pollution prevention. EPA believes that if companies are given the appropriate incentive they will identify the best control for the least cost. Emissions trading under the Clean Air Act (CAA)¹ is an example of an appropriate incentive.

The focus of this Dialogue is an inappropriate incentive: EPA's attempt to create public pressure on industry to reduce pollution through "right-to-know." Give the public information on the nature and amount of chemicals that industry releases into the environment, the theory goes, and the public will motivate industry to reduce pollution in a way that EPA cannot—by protesting, by boycotting and, most importantly, by shining the media's spotlight on the company. President Clinton articulated this "public pressure paradigm" on October 30, 1999, during his weekly radio address: "By requiring industries to tell communities how much they pollute the air and water, we empower citizens to fight back, and create a powerful incentive for industry to pollute less."²

To facilitate such public pressure, EPA has made the public's right-to-know one of the Agency's top priorities. EPA has sought to expand the data it collects from regulated entities and to make the collected data more readily accessible via the Internet. One such example is EPA's effort to expand the chemical release data collected under § 313 of the Emergency Planning and Community Right-To-Know Act (EPCRA).³ To enhance public access to this and other types of environmental data, EPA has built a centralized, environmental data warehouse, known as "Envirofacts," which is publicly accessible through the Internet.⁴ To oversee the maintenance of these massive, publicly available environmental databases, the Agency established the Office of Information.

At the heart of EPA's public pressure paradigm is a highly suspect notion: that every time industry voluntarily reduces chemical emissions in response to public pressure, there is an incremental gain in environmental quality. That is not necessarily true. To measure the effect of environmental policy, one must first establish a goal, then perform a cost-benefit analysis on various strategies for achieving that goal, and select the most cost-effective strategy. It is EPA's job to make that evaluation, not the public's. Environmental policy driven by right-to-know is an abdication of EPA's responsibility. It is dangerous, and should be more closely examined by its proponents.

The following discussion examines this dangerous trend by focusing on shortcomings of two environmental databases established by EPA—the Toxic Release Inventory, established pursuant to § 313 of EPCRA and the Risk Management Plan repository, established pursuant to § 112(r) of the CAA.⁵

The Toxic Release Inventory

Under § 313 of EPCRA, facilities in certain Standard Industrial Classification (SIC) codes that handle "toxic chemicals" in excess of specified thresholds must estimate the amount of such chemicals released into the environment during a calendar year and submit such estimates to EPA and state authorities.⁶ EPA, in turn, compiles the data in a publicly accessible computerized database known as the Toxic Release Inventory (TRI).

According to EPA, § 313 has been one of the Agency's most effective pollution prevention programs. EPA attributes this success to the public availability of the § 313 TRI database. Although § 313 imposes no discharge limits or other regulatory controls on toxic chemicals, the requirement to publicly disclose the quantity of toxic chemicals being released into the environment has led companies to voluntarily reduce discharges.

It is no mystery why § 313 has led to reduced emissions—subject companies hope to avoid unwanted media scrutiny. Consider a company in the following predicament: a company estimates the total quantity of a toxic chemical [30 ELR 10123] released into the environment from its facility during a calendar year, and submits that information to EPA. The Agency enters such data into the TRI database. A public interest group, such as the Environmental Defense Fund, downloads the TRI database, identifies the facilities releasing the largest amount of toxic chemicals, and issues a press release naming America's Top 10 Polluters, one of which happens to be the company's manufacturing site. To avoid the potential adverse consequences of being listed (depressed stock price, public boycotts of its product, etc.), the company decides to reduce its emissions of chemicals subject to § 313.⁷

The question is whether society is better off as a consequence of these reduced emissions. Before rushing to judgment, consider several of the shortcomings of the TRI database. First, § 313 offers no basis for the public to distinguish between toxic chemical releases that pose a risk and those that do not. Some listed "toxic chemicals" (such as mercury, cyanide, and strychnine) are in fact toxic, while others (such as phosphoric acid) are not. Without knowledge of the relative toxicity of listed chemicals, data on the quantity released is rather meaningless. Nevertheless, EPA continues to publish on an annual basis a report containing scary top 10 lists (Most Polluted States, Top 10 Polluting Companies, etc.) based on the quantity of listed toxic chemicals being released.

Second, the data in the TRI only presents a picture of releases from those industries that are actually subject to § 313. In this regard, the statute only requires facilities in SIC codes 20-39 (the manufacturing sector) to submit toxic chemical release reports (and only to the extent such covered facilities exceed reporting thresholds). EPA has, through rulemaking, expanded the types of facilities subject to § 313 reporting to include some additional SIC codes.⁸ Nevertheless, large segments of industry are not subject to § 313 reporting. This limitation looms large when attempting to hypothesize how reducing chemical releases might lead to improved environmental quality. At best, the TRI database presents an incomplete picture of the sources of a pollutant and thus reduction in releases from TRI sources may not be indicative of improved environmental quality.

The significance of these limitations of the TRI database came to a head in the recently concluded dispute over the listing of phosphoric acid as a § 313 toxic chemical. In 1990, The Fertilizer Institute (TFI) petitioned EPA to remove phosphoric acid from the § 313 list. In its petition, TFI provided extensive evidence that phosphoric acid—a fertilizer ingredient, that also is used in soft drinks as a preservative—does not meet any of the three criteria in § 313 for listing a substance as a toxic chemical (specifically, acute human health effects, chronic human health effects, or significant adverse environmental effects).⁹

Eight years later, EPA denied TFI's petition, stating that phosphoric acid meets the environmental effects criterion, i.e., that "the chemicalis known to cause or can reasonably be anticipated to cause, because of (i) its toxicity, (ii) its toxicity and persistence in the environment, or (iii) its toxicity and tendency to bioaccumulate in the environment, a significant adverse effect on the environment . . . ."¹⁰ As support for its ruling, EPA stated that phosphoric acid releases are linked to "eutrophication," a condition where a water body has received excess nutrients causing algae to bloom. As the algae population increases, the dissolved oxygen level in the water body decreases, which can lead to increased mortality of resident fish.

TFI filed a legal challenge to EPA's denial of its delisting petition in the U.S. District Court for the District of Columbia, arguing that EPA's interpretation of the environmental effects criterion effectively reads the toxicity requirement out of EPCRA § 313(d)(2)(C), i.e., the requirement that the chemical "is known to cause, because of its toxicity, . . . a significant adverse effect on the environment." On April 15, 1999, Judge Gladys Kessler granted TFI's motion for summary judgment and reversed EPA's denial of the petition.¹¹

Although EPA appealed the decision to the D.C. Circuit, on October 28, 1999, EPA withdrew the appeal, and thereafter proposed to delist phosphoric acid.¹²

Although TFI's position has been vindicated, the public pressure paradigm has done its damage. During the 14 years that phosphoric acid has been listed as a toxic chemical under EPCRA § 313, EPA identified phosphoric acid as one of the top 10 toxic chemicals in terms of quantities released into the environment. As a result of media and public interest group scrutiny, phosphoric acid producers spent millions to reduce discharges of phosphoric acid. Was this money well spent from an environmental quality perspective? Who can tell, since EPA did not have to perform a cost-benefit evaluation. But consider the following: on the cost side are the massive capital outlays by the fertilizer industry seeking to reduce phosphoric acid releases in response to public pressure. (Also noteworthy—though perhaps not directly relevant to cost-benefit analysis—is the impact of these costs on the fertilizer industry, which has gone through a tremendous consolidation, partly because high environmental costs have eroded already razor thin profit margins. Certainly no one would suggest that listing phosphoric acid as a § 313 chemical has led to this situation, but it contributed.)

On the benefit side of the cost-benefit evaluation, it would be difficult to quantify any environmental benefits that have resulted from reducing discharges of phosphoric acid from manufacturing facilities. As stated in TFI's 1990 delisting petition, releases of phosphoric acid from manufacturing facilities account for less than 1 percent of the phosphates in U.S. waters. The remaining 99 percent comes from sources that are not subject to § 313 reporting. Thus, to the extent society wished to address eutrophication, it would make no sense to target the manufacturing sector for discharge reduction. Yet, that is precisely what happened. This is the danger of letting public pressure dictate environmental policy.(NEWLINE)

[30 ELR 10124]

Risk Management Plan

Another environmental database that EPA is hoping to use to advance its right-to-know driven pollution prevention paradigm are the risk management plans (RMPs) that facilities must submit to EPA, pursuant to § 112(r) of the CAA, if they handle more than a specified quantity of certain regulated substances.¹³

One of the most hotly debated aspects of the RMP is the requirement that covered facilities prepare an analysis of the consequences of a "worst-case release" of the regulated substance (the so-called off-site consequences analysis or OCA). In preparing the OCA, a reporting facility must identify the population that might be injured by a catastrophic failure of the facility's largest on-site vessel containing a regulated substance. To identify the population at risk, the facility calculates the maximum horizontal distance that an airborne release would travel at a concentration sufficient to cause irreversible injury (the so-called toxic endpoint distance). The facility then draws an imaginary circle around its facility using the toxic endpoint distance as the radius, and people living within the circumscribed area comprise the population at risk.

Under EPA's RMP regulations, EPA stated that it would make all facility RMP data, including the OCA, available to the public through the Agency's Envirofacts website.¹⁴ But a firestorm of controversy, focusing primarily on the possibility that terrorists might use the OCA data to identify targets for destruction, led Congress to pass legislation prohibiting publication of the OCA data for one year during which EPA is to study the issue further.¹⁵

An equally compelling reason for not posting OCA data on the Internet is that the worst-case release scenario on which the OCA is based is totally unrealistic, leading to gross exaggerations of the population that might be hurt by catastrophic releases.

In this regard, EPA's worst-case release model is premised on the following assumptions from which a facility may not deviate in calculating the toxic endpoint distances. First, a facility must assume that its largest vessel containing the regulated substance ruptures, that 100 percent of the vessel's contents are released in 10 minutes or less, and that the entire quantity of the released material becomes airborne in that same 10 minutes. In reality, even the most volatile chemicals will form pools that can take days to volatilize, as opposed to vaporizing instantaneously. Second, the facility must assume that all active systems designed to mitigate releases (such as flares or containment structures with drain valves) fail. Third, the release occurs during atmospheric conditions that rarely if ever occur simultaneously, i.e., high temperature and steady "F class" wind speed. (Indeed, in many parts of the country, such conditions never occur simultaneously.)

The effect of piling one inaccurate assumption on top of another (and another) is to overstate the toxic endpoint distance by an order of magnitude or more. Had EPA posted such data on its Envirofacts website, there is little doubt that members of the public—prompted to action by the inevitable press reports of "dangers lurking behind chain link fences" and "circle of death" graphics—would have placed enormous pressure on facilities to eliminate on-site storage of regulated substances. This, in turn, may have led facilities to find substitutes for regulated substances used in operations. Would society be better off from an environmental quality perspective? Again, without a cost-benefit analysis, without evaluating the environmental consequences of encouraging an industry sector to shift from Chemical A to Chemical B, we do not have the objective data necessary to answer this question.

Conclusion

Setting environmental policy is highly complex. There are a myriad of variables to consider. The welfare of entire industries often hang in the balance. Consequently, before deciding on a course of action regarding protection of the environment, society expects there to be a full airing and analysis of all the facts by scientific experts. The public certainly has a right to know about the nature and quantity of chemicals handled in their communities. But EPA should not be encouraging the public to set environmental policy through the public pressure paradigm. It is an abdication of authority.

1. 42 U.S.C. §§ 7401-7671q, ELR STAT. CAA §§ 101-618.

2. Radio Address of President William J. Clinton, Oct. 30, 1999 (text available at http://www.pub.whitehouse.gov/uri-res/).

3. 42 U.S.C. § 11023, ELR STAT. EPCRA § 313.

4. See EPA's "Envirofacts" website available at http://www.epa.gov/enviro/index_java.html.

5. 42 U.S.C. § 7412(r), ELR STAT. CAA § 112(r).

6. Id. § 11023(a), ELR STAT. EPCRA § 313(a).

7. If you have any doubts about this scenario, visit the Environmental Defense Fund's "Scorecard" website available at http://www.edf.org/.

8. See, e.g., 62 Fed. Reg. 23892 (May 1, 1997) (codified as amended at 40 C.F.R. § 372.22(b)). Section 313(b)(1)(B) of the Act authorizes EPA to add or delete SIC codes. 42 U.S.C. § 11023(b)(1)(B), ELR STAT. EPCRA § 313(b)(1)(B).

9. See 42 U.S.C. § 11023(d)(2)(A)-(C), ELR STAT. EPCRA § 313(d)(2)(A)-(C).

10. 63 Fed. Reg. 3566 (Jan. 23, 1998).

11. Fertilizer Inst. v. Browner, 163 F.3d 774, 29 ELR 20349 (3d Cir. 1998).

12. 64 Fed. Reg. 68311 (Dec. 3, 1999).

13. 42 U.S.C. § 7412(r), ELR STAT. CAA § 112(r).

14. 61 Fed. Reg. 31668 (June 20, 1996).

15. The Chemical Safety Information, Site Security and Fuels Regulatory Relief Act, Pub. L. No. 106-40, § 3, 113 Stat. 207, 209 (Aug. 5, 1999).