Regulation through Revelation: The Origin, Politics, and Impacts of the Toxics Release Inventory Program

by James T. Hamilton

Overview

The U.S. Environmental Protection Agency's Toxics Release Inventory (TRI) Program required facilities that handle threshold amounts of specific chemicals to report yearly their releases and transfers of these toxic substances. The TRI data have become the yardstick by which regulators, investors, environmental organizations, and local community groups measure company environmental performance. This book tells the story of the TRI from its origin and implementation to its revision and retrenchment. Its mix of case study and quantitative analysis shows how the TRI operates and how the information provided affects decisions in both the public and private sectors.

Product Details

ISBN-13:	9780521855303
Publisher:	Cambridge University Press
Publication date:	08/29/2005
Edition description:	New Edition
Pages:	358
Product dimensions:	5.98(w) x 9.02(h) x 0.94(d)

About the Author

James T. Hamilton is Charles S. Syndor Professor of Public Policy at Duke University, where he has taught since 1991, as well as a professor of economics and political science there. Professor Hamilton has written or coauthored six books, including All the News That's Fit to Sell: How the Market Transforms Information into News (2004). For his accomplishments in teaching and research he has received awards such as the David N. Kershaw Award from the Association for Public Policy Analysis and Management (2001), the Kennedy School of Government's Goldsmith Book Prize from the Shorenstein Center (1999), and Trinity College's (Duke) Distinguished Teaching Award (1993). Professor Hamilton's scholarly publications reflect his interests in the economics of regulation, public choice/political economy, environmental policy, and the media.

Read an Excerpt

Cambridge University Press
0521855306 - Regulation through Revelation - The Origin, Politics, and Impacts of the Toxics Release Inventory Program - by James T. Hamilton
Excerpt

Introduction

Searching for the real-life impacts of an environmental regulation can be hard. Industrial plants don't have signs on them that say, "Warning: (Costly) Environmental Regulations at Work." Neighborhood yards don't have placards that say, "Another (Statistical) Case of Cancer Avoided Here." Yet if a regulation improves environmental protection, plant decisions and neighborhood environments change. The overall impact of environmental policies consists of the aggregation of effects of particular regulations in particular places.

Consider the effects of one regulation - the U.S. Environmental Protection Agency's Toxics Release Inventory (TRI) program - in one place - Baton Rouge, Louisiana. Under the Emergency Planning and Community Right-to-Know Act of 1986, plants (primarily in manufacturing) with ten or more employees that produce or use quantities of toxic chemicals above a threshold must file reports with the EPA. For each of the nearly 650 chemicals covered in the TRI, a facility annually submits a report if it meets a chemical's reporting threshold. The TRI form tracks a plant's releases and transfers of a chemical broken down by where the toxic ends up: air, land, underground injection, surface water, public sewage, or off-site transfer (primarily to storage or disposal facilities). The form also contains information on facility management of production-related wastes and tracks pollution prevention activities at the site. Since the first release of TRI data in 1989 covering 1987 toxics data, the TRI has become a prime measure of a plant's or company's environmental performance. In the 1987 figures, Louisiana ranked third in the nation (behind California and Texas) in total TRI releases and transfers. The state's large number of chemical plants helped land six Louisiana parishes on the list of top 25 counties in the nation with the largest sum of TRI releases and transfers, including the parish of East Baton Rouge.¹ If the operation of the TRI influences the releases of toxics, Baton Rouge may be a likely place to look for impacts.

For the 2001 reporting year, the 22 plants in East Baton Rouge that submitted TRI forms generated 9.8 million pounds of environmental releases and the 16 facilities in West Baton Rouge reported more than 900,000 pounds.² The Baton Rouge TRI forms chronicle the release of a range of chemicals, including hydrochloric acid, styrene, propylene, and nickel compounds. The facilities reporting are as diverse as the substances they emit. Edo Specialty Plastics (15,500 pounds of on-site releases; plastic pipe industry) is a yellow warehouse-style building just down the road from the Country Club of Louisiana, a development of half-million-dollar homes, and the estate of televangelist Jimmy Swaggart.³ Westside Galvanizing Services (2,002 pounds of on-site releases, 946,653 pounds of total waste managed; metal coating and allied services industry) is a cluster of buildings at the end of a private drive. Heat and smoke are visible in the plant's large manufacturing shed, and noise marks the production process. Nearby neighbors include the West Baton Rouge Correctional Facility, a burned-out sugarcane field, and an ExxonMobil gas plant. Intercontinental Terminals (4,600 pounds of on-site releases, 334,200 pounds of waste managed; chemicals and allied products industry) is right next to the levee. The mix of huge stacks and pipes frames a view of the state capitol dome visible in the distance. Capitol Steel Inc (nickel compounds released; fabricated structural metal industry) is a cramped site of welding, noise, and truck traffic. The plant is on a busy street populated by deteriorating houses, small businesses with bars on the windows, and shops advertising participation in the WIC (Women, Infants, and Children) and Food Stamps poverty programs.

Although the TRI form requires a plant to list on-site and off-site releases of toxics and to provide the name and number of a public contact, nothing requires facilities to respond to questions about what goes on inside the plants. Of the Baton Rouge TRI plants I reached, just five public or technical contacts responded to questions, and only on the condition that I not reveal their names and employers.⁴ A clear sample bias emerged, given that the facilities with the largest TRI figures refused to respond to questions. At the five plants where contacts were willing to talk about the TRI, most had one person responsible for filling out the forms. The time required ranged from a facility were a person spent one day to compile a single TRI form to a plant with ten forms filed, which took approximately 80 hours to complete. One facility sent data to an environmental consultant in another state to complete the report. Complaints about the TRI from those filling out the forms included the thresholds that triggered reporting and the incomplete guidance provided on lead reporting. When asked about the efficacy of the reporting requirements, they said the data had initially shocked managers when it first came out, that it allowed managers to follow what was going on at plants, and that Louisiana's Department of Environmental Quality did focus on lowering the states's TRI figures. As relatively small reporters, however, they said they received little public contact about their emissions. As one filer put it, "There are many bigger fish to fry in Louisiana."

At least three groups have used the TRI data extensively in Louisiana - environmental activists, regulators, and academics. Wilma Subra, a chemist living in New Iberia, Louisiana, won a MacArthur Fellowship for her work in helping communities deal with the impact of hazardous chemicals. Describing her work with TRI data, she notes:

I have used TRI in my work with citizens all over the world. I use the data to track trends in the releases and transfers. I also use the data to compare accidental releases and upset conditions. Where toxic monitoring of ambient air is performed, I use TRI to be sure the correct chemicals are being monitored. I teach the community members the importance of the TRI and how to use it. The EPA has performed mobile monitoring in fenceline communities. I use TRI to identify which facilities are releasing the chemicals detected...The data has been one of the most important sources of information that has caused the facilities to change their behavior. After all it is self reported and the facilities must stand behind what the data represents.⁵

The state Department of Environmental Quality (DEQ) uses the data to monitor facility performance. Commenting on the TRI's impact, Delecia Lafrance of the DEQ observes:

As indicated in the decreasing trends over time since the inception of the TRI program, facilities have significantly reduced releases and become more aware of their annual release totals. By disclosing TRI information to the public, industries are more accountable to citizens and the environment. The public accessibility of TRI data has encouraged facilities to look at process alternatives, chemical substitutions, and alternative technology that generate less waste and/or decrease releases.⁶

Paul Templet, a professor at Louisiana State University, uses TRI data to write about environmental policy. Analyzing the operation of the TRI, he concludes:

I believe it has changed the local plants behavior both directly and indirectly. Directly by exposing the levels of releases causing plants to reduce releases. Indirectly, because people like me use it to point out how my state compares to others and publish the results. I was also Secretary of the Louisiana Dept. of Env. Quality and used the data to compile lists of the largest dischargers and to call for reductions. I also used it to condition the facility's tax subsidies.⁷

Writers for the Baton Rouge daily newspaper, The Advocate, also use the TRI in environmental coverage. A 2003 story in The Advocate triggered by the release of the 2001 TRI data quoted plant officials talking about their efforts to reduce releases. Overall, Louisiana TRI releases dropped from 154.5 million pounds in 2000 to 123 million pounds in 2001, prompting Mary Lee Orr of the Louisiana Environmental Action Network to observe, "I think that governments and industry are not only realizing that lost product [e.g., chemicals lost as pollution] is bad for business, it's bad for the community and bad for workers."⁸ Mark Schleifstein, a staff writer for The Times-Picayune in New Orleans, frequently uses the TRI data in industry stories. Asked to comment on the impact of the data, he replied:

A back of envelope estimate is that the first years of the list caused about $5 billion in investment in chemical plants in Louisiana for "debottlenecking" and other process changes aimed at reducing emissions. A company owning several fertilizer plants spent $70 million reconstructing its waste piles to reduce its numbers. Shell Oil developed a recycling method for hydrochloric acid to drop its injection well numbers. At many chemical plants we visited for the 1989 series, executives told us that the key critics of their numbers were their boardrooms. Typical response when numbers came out, said several execs, was "Let me get this straight. You're allowing x million pounds of product go up in the air? water? underground?" Louisiana Department of Environmental Quality has had varied response to the data, depending on which governor has been in office. All have been aimed at publicizing the numbers and pointing out where they've fallen. During a four-year stretch under Gov. Buddy Roemer (88-92), the agency was more aggressive and used TRI as a tool to jawbone companies into more rapid emissions reductions. Louisiana Environmental Action Network, umbrella group representing more than 100 enviro orgs across the state, has used TRI data from time to time to target specific industries, types of waste disposal, etc. Ditto Greenpeace.⁹

When asked how well the TRI has worked, Schleifstein replied:

Based on the experience in Louisiana, it's been extremely successful. Emissions have dropped by 82 percent since 1987, although they've been pretty much level since 1995. Pretty good for a program that simply requires the publication of numbers.

Focusing on a single area, Baton Rouge, shows how the TRI affects people. For those filling out the forms at reporting facilities, the TRI's requirements can mean a day or week of their lives each year. For a small set of environmental activists or regulators, the TRI can help them achieve the goals of their careers - pollution prevention and reduction. If the scrutiny generated by the TRI results in fewer emissions of air carcinogens or lower releases of toxics into waterways, the TRI can (for a set of people who may never realize it) improve human health. This approach to analyzing the TRI misses a key set of questions, however - where did the TRI come from? How has it operated and changed over time?

This book answers those questions by telling the story of the TRI as a regulatory program. The creation of an information provision program may appear an unlikely event, given that the policy imposes costs concentrated on industries likely to oppose the measure and creates benefits diffused among a public unlikely to lobby actively for its creation. The chemical plant accident in Bhopal, India, in December 1984, however, generated public scrutiny of toxics use and production and emboldened a set of political entrepreneurs in the House and Senate to insert a reporting provision in a broader hazardous waste bill. The Emergency Planning and Community Right-to-Know Act of 1986 (EPCRA) established the initial set of chemicals, industries, and facilities that would have to provide the EPA with yearly reports on their toxic releases and transfers. The legislation also marked the first time that Congress required an agency to create a publicly accessible electronic database, which later become known as the Toxics Release Inventory. The first TRI figures, capturing toxics information from 1987, were released in 1989 with great fanfare. More than 19,000 facilities had submitted more than 74,000 reporting forms, which detailed 22.5 billion pounds of releases and transfers for the 320-plus chemicals then listed on the TRI.¹⁰ The ensuing attention led companies to pledge reductions, environmentalists to use the data to press for change, and regulators to point to the TRI as evidence for the power of information. During the 1990s, the TRI expanded to include more chemicals and industries. By reporting year 2001, close to 25,000 facilities were submitting TRI forms, which by now contained data on releases and on the management of production-related waste. For the set of chemicals and industries continuously tracked from 1988 to 2001, total releases and transfers had dropped 54.5%.

The analysis in the book roughly follows a regulatory time line: votes to set up the program, early implementation of information provision, revisions to the program, assessments of the impact of the TRI, and broader lessons to be learned about regulatory implementation from the operation of the TRI. I use a wide range of methods in the book, including a statistical analysis of roll-call votes, a stock market event study, a detection-controlled estimation model of regulatory noncompliance, and qualitative analyses of the legislative, administrative, and judicial actions that defined the TRI. A common theme unites these analyses - the impact of information provision about pollution quantities depends strongly on the distribution of other types of information. The initial legislative votes to set up the TRI were influenced by the lack of information voters have about technical amendment votes in Congress. Firm compliance with the TRI in the early years of the program turned on another type of information, the knowledge that the program even existed and imposed reporting requirements. Company incentives to reduce pollutants depended on the degree that environmental information circulated to investors, who in turn bought/sold company shares depending on the release of TRI information. Attempts to reduce the impact of the program were unsuccessful when Republicans led a high-profile fight in Congress to reform risk regulation. But later attempts to scale back the program's reach were successful because they involved less-visible regulatory actions. Finally, concerns about the use of information by terrorists caused the government to reduce the availability of chemical release information. Overall, the evolution of the TRI shows how the presence or absence of information aside from the TRI figures affected the impact of the TRI data.

Chapter 1 shows that information provision as a policy tool generated opposition from its inception. The amendment adding toxics reporting to the Superfund reauthorization bill passed in the House in December 1985 by a vote of 212 to 211. The greater the amount of toxic releases and transfers in the congressional district that would be tracked in TRI, the more likely the legislator was to oppose the provision of information about pollution. The eventual text of EPCRA, which President Reagan signed in October 1986, specifically defined who would have to report and also outlined administrative procedures that would allow interest groups to petition the EPA to add or delete chemicals from reporting requirements. Chapter 2 describes the rulemaking that led to crafting of the TRI reporting form, which was published in February 1988. Firms had until July 1, 1988, to submit their reports on toxics releases and transfers in 1987. The EPA quickly fashioned a reporting program and collection mechanism, so that on June 19, 1989, the EPA was able to release the national TRI figures. The analysis in Chapter 2 shows that the information was news to at least two sets of people, investors and reporters. The release of the TRI data generated drops in stock market value for some of the reporting firms, though this was less likely for companies already known to have significant pollution problems. Reporters were also less likely to write about a firm's TRI releases if the company was in a pollution-intensive industry or had a track record of involvement at Superfund sites.

Chapter 3 chronicles the early years (1987-1992) of the TRI's operation. Two complaints about the initial TRI were that nearly a third of those who should have filed may not have submitted forms and that the scrutiny generated by the program focused on pounds of pollutants rather than risks. In Chapter 3, I analyze TRI nonreporting in Minnesota, which in 1991 launched an extensive program to find facilities that should have submitted 1990 TRI forms. The results show that a plant's failure to file a TRI form relates more to ignorance about reporting requirements than to evasion of the regulation. Though Minnesota added a substantial number of facilities to the TRI through its enforcement program, these facilities turned out to be relatively small polluters, so overall TRI totals for the state did not greatly increase. To explore the impact of risk on firms' toxic emissions, I examine reported changes at the plant level of the release of carcinogens into the air between 1988 and 1991. Controlling for the quantity of air toxics released in 1988, I find that plants whose emissions generated higher numbers of expected cancer cases did reduce their emissions more between 1988 and 1991. The nature of the community bearing the pollution risk also affected firm decisions. The higher the voter turnout in the area, a proxy for residents' likelihood of collective action, the greater the reductions in a plant's release of air carcinogens.¹¹

Discussing the operation of "the TRI program" is a misnomer, because the scope and implementation of the TRI have changed frequently over time. In Chapter 4, I show how the reporting requirements and their impacts have shifted in reaction to EPA's formal rulemaking process, congressional activity, presidential politics, court decisions, and informal agency activity. Interest groups were able to use the administrative procedures spelled out in EPCRA to petition the EPA to add and delete chemicals from the TRI list. The course of the TRI changed even more with changes in policymakers. Under President Clinton, the EPA added 286 chemicals to the reporting list and required seven more industries to file reports. When Republicans gained control of the House and Senate in 1995, they tried (unsuccessfully) to reduce the scope of TRI reporting through regulatory reform proposals that never made it into law. Under President George W. Bush, the EPA shifted attention to ways to reduce the reporting burdens associated with the TRI. Throughout the program, the EPA also used informal actions, such as the creation of the 33/50 voluntary pollution program or creation of software that combined TRI data with risk measures, to focus attention on particular industry segments or facilities.

A common set of factors may influence the operation of regulatory programs: public opinion, media coverage, enforcement policy, and office politics. Chapter 5 explores these factors in the evolution of the TRI. The high level of public interest in toxics that helped generate EPCRA dissipated over time, even as the regulatory program it created lived on. Media coverage in the popular press focused on political battles, such as the 1995 attempts to pass regulatory reform legislation or the 1996 election year focus on the TRI in speeches by President Clinton. The trade press continued to cover the TRI when it fell out of public view, in part because rulemakings continued to generate disputes in the regulatory community. With the election of George W. Bush and Republican control of Congress, the direction of TRI policy changed. Budgets shrank, inspections dropped, and the focus of agency policy became how to reduce the reporting burden. Actions by the Office of Management and Budget (OMB) particularly directed the EPA's attention to lessening the costs to industry of reporting. The budget for the TRI program dropped 39% from FY 1997 to FY 2004. The agency was able to maintain operations in part through changes in technology, such as the development of the TRI-ME reporting software and development of the Internet-accessible Central Data Exchange for firms to submit their reports. These actions lowered costs to filers and reduced the transaction costs of data processing for the agency.

The academic research summarized in Chapter 6 demonstrates that the TRI changed decisions and debates. Quantitative studies show that the toxics information brought news to investors, reporters, and residents. Case studies reveal how the TRI alerted some managers to environmental releases they did not previously know about and led many to undertake actions to reduce their TRI figures. Environmentalists used the data to develop national reports, pressure firms for reductions, and press legislators for changes in environmental policy. Regulators used the data to target particular industries, firms, and plants for scrutiny and enforcement actions. The impact of the TRI on decisions made at a plant appeared to vary with the nature of the surrounding community, the characteristics of the parent company, the environmental laws in the state, and the enforcement of traditional command and control regulations. The existence of the TRI, coupled with advances in geographic information systems (GIS) technology, helped fuel environmental equity debates about the distribution of pollution exposure across demographic groups. The operation of the TRI also demonstrated the feasibility of using information provision as an environmental policy tool and encouraged the development in other countries of Pollutant Release and Transfer Registers (PRTRs).

Though the TRI requires plants to report their releases and transfers of toxic chemicals, much of what goes on inside these facilities remains unknown. What are the costs of producing the TRI data? What resources go into the changes in production process or pollution control aimed at reducing TRI figures? How accurate are firm estimates of releases? What fraction of reported reductions in TRI emissions are due to information provision versus other factors, such as variations in production or the enforcement of command and control regulations? Answering these questions would better establish the costs of the program and the chain of causation between information provision and altered decisionmaking. Though current research on the TRI shows the benefits of new information generated by the reporting requirements, more information and analysis are needed in the future to understand the net benefits of the program.

Chapter 7 concludes with lessons from the regulatory implementation of the TRI and lessons for regulatory implementation in other environmental programs. From many perspectives, the operation of the TRI is a successful story of policy innovation. The EPA was able to set up within a short period of time the first publicly accessible electronic database mandated by Congress. The TRI did provide individuals in the private and public sectors with new information about toxics. People did learn, their actions did change, and debates were altered by the existence of the TRI. The development of the Internet and diffusion of computers have both lowered the costs of reporting in the program and raised its benefits by raising accessibility.

The operation of a single regulatory program cannot define the overall success or failure of particular approaches to environmental regulation. Yet the story of the TRI does offer a series of lessons that may serve as hypotheses to be tested with other environmental policies. These lessons include the following. Perceived flaws in regulation emerge more from politics than from lack of foresight. The impact of regulations on the ground varies with changes in who occupies the White House and Congress. Administrative procedures, and judicial review, do allow interest groups another shot at influencing the course of regulatory policy. Ideas spread. Information provision can work. Intermediaries lower the costs to the public of public information. The impact of information is not uniform. Regulators learn over time. The following chapters illustrate these lessons through the story of the TRI.

ONE
Legislating an Incomplete Contract

Environmental regulations follow a long paper trail. They start life as a proposed bill in Congress, travel through committee hearings, generate floor debates, and emerge as a final bill passed by the House and Senate and signed by the President. The exact wording of the regulations comes later in a public rulemaking process, though the actual meaning and impact of the rules may depend on future court cases, agency guidance documents, and enforcement decisions by regulators. For the Toxics Release Inventory program, the path from law to regulation runs through an unlikely spot - the Union Carbide plant in Bhopal, India. On the night of December 2-3, 1984, a leak at the chemical plant generated clouds of methyl isocyanate that killed more than 2,000 people. Worldwide media attention generated debates over how to prevent such accidents and exposures. By the end of December 1984, four subcommittees in the U.S. House of Representatives were investigating Bhopal. In 1985, multiple bills were introduced to provide residents with information about what chemicals were used by plants in their neighborhoods.¹ Speakers frequently referred to Bhopal during debate, generating 82 references to the incident in the Congressional Record in 1985-1986. In 1986, Congress finally passed the Emergency Planning and Community Right-to-Know Act (EPCRA), whose implementation gave rise to the regulatory program known as the Toxics Release Inventory program. In this chapter, I examine how theories of regulation help explain the politics behind the passage of EPCRA, the design of its legislative language, and the rationales for mandating information provision about pollution.

The political economy of EPCRA demonstrates the impact of information on delegated decisionmaking. Environmental regulations, like other forms of government rules, depend on a series of delegated decisions. Voters delegate to Congress members the power to write laws that control pollution, though the likely ignorance of voters about the activities of their representatives may influence the degree to which Congress members respond to their constituents. Congress members delegate decisions over the exact wording of environmental regulations to agencies, in which hidden actions by regulators and hidden information may allow bureaucrats to pursue their own agendas. Information provision programs such as the TRI delegate the generation and reporting of information to firms, though the latitude involved in estimating pollution figures leaves companies discretion in developing their toxic emission reports. The origin, implementation, and overall impact of the TRI ultimately depend on how information is distributed among voters, policymakers, firms, and interest groups. This chapter analyzes how voter information affected the passage of EPCRA, how concerns about delegated decisionmaking were reflected in the text of the legislation, and theories of how and why the TRI might work (though these theories were not a significant part of the debate over the creation of the program).

---

© Cambridge University Press

Table of Contents

1. Legislating an incomplete contract; 2. Defining terms: rulemaking and the initial TRI data release; 3. Spreading the word in the public and private sectors; 4. Politics of expansion and contraction; 5. Lifecycles in the regulatory environment; 6. The impact(s) of the TRI; 7. Lessons from and for regulatory implementation.

Editorial Reviews

“…everyone should applaud Hamilton's superb scholarship…Highly recommended.” –CHOICE, R.E. O'Connor, National Science Foundation

From the Publisher