Project Management II Research Paper

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 615

Toxic Chemical Governance Failure in the United States: Key Lessons and Paths Forward

ARIANA M. CHIAPELLA, ZBIGNIEW J. GRABOWSKI, MARY ANN ROZANCE, ASHLIE D. DENTON, MANAR A. ALATTAR, AND ELISE F. GRANEK

Over 40 years of regulations in the United States have failed to protect human and environmental health. We contend that these failures result from the flawed governance over the continued production, use, and disposal of toxic chemicals. To address this failure, we need to identify the broader social, political, and technological processes producing, knowing, and regulating toxic chemicals, collectively referred to as toxic chemical governance. To do so, we create a conceptual framework covering five key domains of governance: knowledge production, policy design, monitoring and enforcement, evaluation, and adjudication. Within each domain, social actors of varying power negotiate what constitutes acceptable risk, creating longer-term path dependencies in how they are addressed (or not). Using existing literature and five case studies, we discuss four paths for improving governance: evolving paradigms of harm, addressing bias in the knowledge base, making governance more equitable, and overcoming path dependency.

Keywords: toxic chemicals, governance, innovation, path dependency, coproduction

Despite decades of legislation (Wagner 2007) and  public-interest litigation (Lind 2015), contemporary society remains saturated with environmental pollution risks of its own production (Beck 2008). Globally, these failures of environmental protection cause millions of pre- mature deaths per annum and cost society billions of dollars in economic damages (Landrigan et  al. 2017). These per- vasive and increasing environmental threats often remain unknown until publicized by private citizens, scientists, non- governmental organizations, or the media. Warnings issued with increasing frequency by the scientific community (e.g., Ripple et al. 2017) are all too often met with policy gridlock and a lack of substantive government action. Despite the existence of cleaner technologies whose economic and social benefits exceed transition costs, environmental pollution has become the leading global cause of preventable death (Landrigan et al. 2017).

Significant gains have been made in reducing global primary emissions of some highly regulated chemicals (SC 2017), although debate continues on the overall impacts of shifting global geographies of the production of toxic risks (Rasli et  al. 2018). The United States serves as an excellent case study on the multifaceted nature of governing toxic chemical risks. It has lagged behind the European Union

in adopting the precautionary principle, especially with regards to importing consumer and industrial products (Becker 2010), and in dealing with emerging contaminants (Bao et  al. 2015). And, as elsewhere, partial solutions have led to unintended outcomes, such as increases in ozone exposure concomitant with declines in particulate pollu- tion due to the widespread adoption of catalytic converter technology (HEI 2019).

As an interdisciplinary group of scholars crossing the domains of environmental science, public administration, and political science, we attribute failures within the United States to a more general failure of environmental governance (Mol 2016). A focus on governance highlights how differ- ent domains of social decision-making define and manage risks and responsibilities associated with the production and distribution of toxic substances. Governance also centers on the long-running concerns of professional versus layperson knowledge (Brown 1992, Brulle and Pellow 2006), contesta- tions over “facts” in the “post-truth” era, and the degree to which administrative power can shift regimes of environ- mental governance established through years of legislation (Revesz 2019).

Existing work documents how failures of environmental governance results in environmental injustice through the

BioScience 69: 615–630. © The Author(s) 2019. Published by Oxford University Press on behalf of the American Institute of Biological Sciences. All rights reserved. For Permissions, please e-mail: [email protected]. doi:10.1093/biosci/biz065 Advance Access publication 26 June 2019

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

616 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

inequitable distribution of exposure to toxic chemicals based on racial and socioeconomic identities (Landrigan et  al. 2017). This unevenness has resulted from racist and oppor- tunistic practices of uneven permitting and enforcement (Morello-Frotsch and Shenassa 2006) and contributes to the framing of governance failures as someone else’s problem (Pastor and Morello-Frotsch 2018). Simultaneously, toxic chemical risks are ubiquitous and systemic in nature, affect- ing humans across the globe regardless of their socioeco- nomic class (Schwarzenbach et  al. 2010). Existing support for high environmental quality across the political spectrum (Feinberg and Willer 2013), combined with rising rates of developmental and chronic diseases (Landrigan et al. 2017), indicates that there is an urgent need to frame both risks and policy proposals in a way that mobilizes those of diverse political orientations.

The current political climate in the United States indicates significant resentment against the political establishment, typified by a resurgent anti-administrative state agenda remi- niscent of the 1980s (Hejny 2018) with significant negative consequences for public and environmental health (Cutler and Dominici 2018). On the upside, the current adminis- trative swing has exposed the long-standing pattern of elite interests disproportionately writing, lobbying, and adjudicat- ing environmental laws in their narrowly defined self-interest and has increased mobilization of nongovernmental organi- zations, community based organizations, and science-based advocacy organizations (Mol 2016). This political landscape highlights a need for scientists to engage directly with increased public scrutiny (Latour 2004) by calling for demo- cratic governance to employ best available knowledge for protecting the quality of our environment and public health.

With this goal in mind, we provide a conceptual overview of the governance structure of toxic risk management in the United States. We use our conceptual framework to analyze several high-profile case studies and discuss a proposed set of principles, ongoing initiatives, and challenges of improv- ing toxic chemical governance in the United States.

Toxic substances policy in the United States Literature in the United States has documented numerous instances of failure across diverse classes of pollutants, natural systems, and regulatory contexts (e.g., Davies and Mazurek 1998, Paavola 2006, Fletcher 2009). Common causes of attributed failure include a failure to regulate classes of toxic chemicals (Mesnage et  al. 2015), standards inadequate to achieve protection (Vogel and Roberts 2011, Boone et  al. 2014), and nonenforcement of existing regulations (Farber 1999). We define failure as unacceptable levels of human and environmental exposure to toxic chemicals during their pro- duction, use, transport, and fate in the environment.

Although existing regulations and policies written by legislatures and enacted by executive and administrative branches of government (e.g., federal, state, tribal, and local agencies) ostensibly act in the public interest, other social actors actively shape their design and language (e.g., lobbying

from industry and citizen groups; Davies and Mazurek 1998, Cash et  al. 2006) to constrain their effectiveness. In addi- tion, manufacturers, installers, and users of potentially toxic substances routinely evade effective regulation through legal and illegal means (Lynch and Stretesky 2014). In response to these recognized drivers of failure, remediation efforts gen- erally prioritize stricter regulation based on the perceived risks of the substance in question. Tactics include limiting harmful exposure and environmental releases via command and control regulation, mitigating ongoing exposures with funds generated from regulation, and, in some cases, pro- viding incentives for eliminating sources of risk by shifting to alternative technologies (Wilson and Schwarzman 2009). However, the technological capabilities of manufacturing often evolve faster than their regulatory apparatuses, and industries themselves have built up a technological, intellec- tual, and regulatory ecosystem that has effectively excluded many greener technologies (Woodhouse 2006).

In the United States, the current policy framework around toxic substances remains highly fragmented among jurisdic- tions of federal agencies such as the Environmental Protection Agency (EPA), the US Department of Agriculture, the Food and Drug Administration (FDA), and the Department of Health and Human Services. Some states have additional regulations, such as California’s proposition 65, requiring the state to publish and annually update a list of known chemi- cal carcinogens or reproductive toxicants (Nelson 2013). Mirroring jurisdictional fragmentation resulting from sec- tor-specific regulations, variation exists for different media (e.g., soil, air, water; Caliman and Gavrilescu 2009, Rudel and Perovich 2009).

In addition to poor policy design and fragmentation, the current policy framework leaves many chemicals un- or under-regulated. The primary federal toxic chemical regula- tion, the Toxic Substance Control Act (TSCA; implemented in 1976), has grandfathered in nearly 62,000 previously unregulated chemicals without evaluation of risk (Vogel and Roberts 2001), a number not including the manufactur- ing by-products of those chemicals or their environmental derivatives. A hard-fought 2016 amendment to TSCA estab- lished a schedule for evaluating the estimated 85,000 existing chemicals in the marketplace, shifted toxicological analyses toward a risk-based framework, limited the ability of com- panies to claim commercial confidentiality, and has elimi- nated the consideration of cost in risk assessment (Frank R. Lautenberg Chemical Safety Act 2016). With a risk-based framework, the burden of proof for evaluating potential harms to humans and the environment is placed on the regulatory agency, who will only regulate a chemical if it is shown to pose a risk to human and environmental health in a highly specified exposure pathway. TSCA remains in litigation over fundamental procedural issues, including the process of prioritizing different substances for evalua- tion, definitions of unreasonable risk, and whether the EPA should consider the feasibility of replacement substances in prioritization (Bergeson and Graham 2017, CW 2019).

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 617

Despite the 2016 TSCA requirements for EPA to evaluate all new chemicals before market release, EPA remains under- funded and understaffed for timely evaluation. Exacerbating the situation, TSCA does not require companies to provide toxicological data, and the annual evaluations of 20 high- risk and 20 low-risk chemicals cannot keep pace with new chemical production (Botos et al. 2018).

In addition to TSCA, substances that pose threats to human and environmental health are regulated by a num- ber of other regulatory instruments including the Clean Air Act, the Clean Water Act, the Safe Drinking Water Act, the Resource Conservation and Recovery Act, and various work- place regulations under the Occupational Safety and Health Administration—all of which rely on similar processes of analyzing risk to determine the extent to which they should be regulated (Steward 1995). In addition, chemicals intended for human consumption as food stuffs, pharmaceuticals, tobacco products or derivatives, and personal care products undergo their own regulatory procedures through the Food and Drug Administration. Further fragmenting the regula- tory environment, the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), regulates the sale of agricultural chemicals not registered with the EPA. Although this process requires stringent manufacturer testing, labeling, and peri- odic recertification (every 15 years) of pesticides, designa- tion of critical economic importance can outweigh human and environmental risks, especially if those substances have become widespread, or engineered into crop production systems. Perhaps because of this it is rare for a pesticide to be denied re-registration unless there is overwhelming evidence of human and ecological harm. Both FIFRA and TSCA suf- fer in effectiveness because of their definitions of risk, and the ease of industry influence on their decision-making processes.

In contrast, the European Union’s Registration, Evaluation, Authorization and Restriction of Chemicals (REACH) legis- lation, uses precautionary principles, putting the burden of proof of safety on the industries that produce them (Silbergeld et  al. 2015). Under the precautionary principle, new substances and their derivatives are assumed to pose risk until proven otherwise, and the responsibility for prov- ing the absence of risk is placed on both the producer and the regulator. REACH also requires producers to provide the European Chemicals Agency with toxicological information, and uses a spectrum of safety standards matched with appro- priate use restrictions and mandatory labeling.

All of the above policies, struggle with inadequate resources for chemical assessment and are vulnerable to lob- bying and uneven adjudication. A primary challenge with REACH, for example, is that EU member states are respon- sible for implementing the chemical evaluation process, leading to inconsistencies in implementation. Despite the improvements made to TSCA, it still falls short of REACH’s founding precautionary principles. This comparison indi- cates the importance of strong guiding principles in effective toxic chemicals governance. Although the incorporation of

precautionary principles will be an important step toward better regulation of toxic chemicals in the United States, it alone is inadequate. Effective protection from toxic chemi- cal risks will require changes in governance practices and an evolution of the technologies and practices that generate risk to align with public and environmental health goals.

A framework for understanding toxic chemical governance Governance includes the social practice of designating rules, standards, and norms according to which actors and institu- tions negotiate and make decisions (Rogers and Hall 2003), including what knowledge is considered valid and useful (Wynne 2003). A focus on governance identifies how the pres- ent conditions of our society, environment, and technological infrastructures are interdependent with the forms of expertise and political authority deemed necessary to manage harms to humans and the environment (Scott 1998, Jasanoff 2004, Latour 2004). In particular, governance highlights the process of classifying potentially hazardous substances as risks, the consideration of different forms of knowledge or expertise in that process, and the path dependency—or inertia—result- ing from prior decisions. We describe these concepts below and use them later to evaluate five high profile cases of toxics regulation success or failure in the United States.

The construction of risk around toxic chemicals can be defined as a social process of emphasizing some dangers over others (Douglas and Wildavsky 1983). Formal risk analysis involves calculating the probability of a speci- fied level of chemical exposure multiplied by the probable consequences of that exposure (Bocking 2004). However, in practice, such an analysis relies on a set of assump- tions about social behavior alongside physiological and toxicological data and often disregards risks experienced by affected communities (Bocking 2004, Beck 2008). In this sense, standard risk analysis treats risks to public health and the environment as end-of-pipe problems and unplanned releases as public relations problems. Such thinking ignores that the generation of risk results from choices about how chemicals can be produced. These choices produce systems that have global consequences, normalize the production of toxic byproducts, and have significant sunken costs in facili- ties and the development of economic sectors dependent on those kinds of inputs (Beck 2008).

Although different social actors perceive and calculate risks in different ways, risk management is generally seen as an activity worthy of professional expertise. Expertise in this sense refers to the social practices of designating individuals, institutions, technologies, and methods as sources of author- itative knowledge (Scott 1998, Wynne 2003, Bocking 2004, Jasanoff 2004). Expertise often has disciplinary bound- aries, which prevent synthesis across and within disci- plines (Cartwright 1999). As social actors vie for legitimacy within networked political, financial, environmental, social, and technical systems (Grabowski et  al. 2017) institutions take on more stable forms, routing social decision-making

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

618 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

processes into established mechanisms and fora exhibiting different forms of path dependency.

Path dependency refers to the way in which future pos- sibilities are seen as constrained by present conditions and largely results from decisions about financial, institutional, intellectual, and bureaucratic investments in social ways of doing, infrastructures, and technology (Jasanoff 2004, Woodhouse 2006, Beck 2008). Path dependencies may lead to the generation of systemic bias in what type of knowledge is produced and considered relevant, which is often con- tested by popular movements (Hess 2015). Disrupting path dependency generally requires major events, a form of punc- tuated equilibrium (Pierson 2000). Systemic path dependen- cies result when agents within institutions prevent change despite widely recognized problems (Sydow et al. 2009). For example, toxic chemical risks have often been framed as by- products or externalities or as attributes of chemicals to be managed when, in fact, they are embedded within “normal” operations (Perrow 1984, Beck 2008).

Drawing on these concepts of risk, expertise, and path dependency, we present a conceptual framework of the cur- rent toxic chemical governance system in the United States. We break down the overall governance system into five interdependent domains in which toxic chemical regulations are interpreted, implemented, and evaluated: knowledge production, policy design, monitoring and enforcement, evaluation, and adjudication (figure 1). Each domain oper- ates simultaneously in time and space, although problems can flow from one to another (e.g., failures of enforcement often result in adjudication).

Knowledge production. Although different forms of expertise and knowledge are embedded in all domains, the knowledge base refers to the overall organization of information pertain- ing to toxic chemicals. This includes “facts,” information, and the accepted methods for producing them, which invokes the ways institutions, values, norms, and discourses within a social system decide what type of knowledge is legitimate or useful (Jasanoff 2004, Stehr 2015). Many stakeholders are involved in toxic chemical knowledge creation, including affected communities, the scientific community, the media, and industry representatives. Each stakeholder group con- structs their knowledge differently, leading to different claims about toxic chemicals. These varied claims and perspectives on what constitutes legitimate knowledge, and how it is and should be produced, lead stakeholders to identify and catego- rize threats to health and the environment in radically differ- ent and often incompatible ways (Wynne 2016).

Policy design. Policy design includes processes for describ- ing present conditions, framing goals, creating incentives or regulations to achieve those goals, and assigning rights and responsibilities to different social actors within an over- arching policy architecture. This domain heavily influences monitoring, enforcement, and evaluation activities, and sets the stage for adjudication. It is here that the interests and

values negotiated within the knowledge base become codi- fied into legislation via regulations, incentives, and budget allocations. Top-down policy architecture is more easily implemented but less flexible for local stakeholders, whereas bottom-up approaches are adaptive and flexible, but can be difficult to create given disagreement among stakeholders or lead to unequal environmental regulations across the coun- try (Bocking 2004). Canonical descriptions of the policy process divide participants into decision-makers, generally referring to elected officials, and stakeholders, includ- ing affected communities, industries, and special interest groups. It has been observed that local affected communities engaged in the policy process often demand a precautionary approach to protect their local human and environmental safety (e.g., Bullard and Johnson 2009), whereas industry interests push for limiting regulation and including policy language that allows them to continue current business operations (e.g., Boone et al. 2014).

Monitoring and enforcement. Monitoring and enforcement refers to the mechanisms of observing regulated activities and the ability to coerce compliance with standards and operating procedures as written. Enforcement can take place via three primary approaches. First, the formal regulatory arena: local, state, and federal executive and regulatory agen- cies issue fines for limit exceedances and issue release per- mits, among other codified approaches to compliance. This requires sufficient resources for detecting and correcting vio- lations. Second, self-regulated monitoring and enforcement: In the absence of close regulatory oversight, private-contract auditing agencies oversee industry groups to ensure compli- ance, often via certification programs. Third, complaints by affected communities who identify misconduct, draw media attention, and place political pressure on industry groups to comply with regulations. This often happens when there are limitations in agency resources. Proper enforcement requires adequate policy design, including initial political will, coherence in writing legislation, and consistent, long- term political and financial support for monitoring and enforcement efforts (Wagner 2007).

Programmatic and policy evaluation. Programmatic and policy evaluation refers to evaluation of policies designed to man- age chemical exposure, production, and transportation, and the creation of alternative technologies and practices. This domain is tightly linked to policy design, whereby program and policy evaluations should inform future policy designs. Major actors involved in this domain include federal, state, and local regulatory government agencies, affected com- munities, industries, the scientific community (including nongovernmental agencies and nonprofits), and the media. Judicial agencies can also perform policy evaluations in response to publicity of toxic chemical risk to human or environmental health, and initiatives within industry to change their practices or use of certain chemicals may also be involved. This is the domain in which stakeholder claims

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 619

about the impacts of policies are evaluated, then either used to evolve policy or disregarded.

Adjudication. Adjudication is the legal process by which disputes are settled, policies are interpreted (e.g., claims of harm and liability), and enforcement activities are contested (e.g., ongoing TSCA litigation pertaining to procedural rules for chemical risk evaluation). A key part of adjudica- tion pertains to the formal determination of compliance, liability, harm, and responsibility to parties involved in litigation.

More broadly, adjudication is the process by which knowledge claims, policy efficiency, and the distribution of benefits or burdens of a particular substance are vetted by the judicial branch of government. Affected or scientific communities, nongovernmental interest groups, govern- ment agencies, and industries often initiate adjudication (Hoffman 1999) as a means of changing activities within the other governance domains. These changes may include policy design modifications, enforcement of compliance, increased monitoring and enforcement, and promoting or contesting evaluation. Common forms of adjudication include petitions to state and federal agencies, settlements with regulatory and private sector entities, or lawsuits.

Although the courts can settle issues of human and environ- mental failures, this approach is, by definition, reactionary and can only interpret legislation to nullify or clarify obliga- tions, or set appropriate enforcement actions, through slow, costly, and often adversarial means (Silbergeld et  al. 2015). Adjudication can preemptively affect policy as legislatures shy away from creating unenforceable policies.

Application of governance system conceptual framework We expand our conceptual framework of five governance domains (i.e., knowledge production, policy design, moni- toring and enforcement, evaluation, and adjudication) to determine how patterns of flawed governance lead to unsafe exposure of select chemicals. We create a qualitative evaluative framework for governance issues related to risk definition, knowledge production, and path dependency across the five domains (figure 2). Based on aspects of the governance literature discussed above, we consider whether risk, expertise, and path dependency are succeed- ing, failing, or partially succeeding, corresponding with a numerical ranking (see figure 3). Risk is qualitatively evaluated as succeeding if there is plurality and consensus of how the risk is framed and it is considered failing if risk

Figure 1. Conceptual framework of US toxics governance domains, and the actors who contribute to each domain. At the base of each box, a summary of common ways these domains contribute toward governance failures are listed. Affected communities bear the costs and risks of contaminants, or benefit from current practices. The scientific community can include academic, government, and industry scientists, who identify and define risk, but also develop the technologies that create risks. Industry refers to private companies creating and owning the technologies producing or using chemicals. News or media outlets procure and disseminate information to the public about risks.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

620 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

Figure 2. Evaluative framework for examining governance issues of risk, expertise, and path dependency in each governance domain. Green highlighting indicates success (agreement), grey indicates partial success (partial agreement), and red indicates failure (disagreement).

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 621

Key findings Most cases ranked between failing (score = 1) and partially succeeding (score = 2) out of a total possible score of success (score = 3; figure 2). Overall, we found that although most cases had robust risk knowledge that included diverse per- spectives for framing risk, ongoing issues with path depen- dency, and in some cases policy regression, were common throughout the cases evaluated.

Lead in school drinking water (score: 1.4/3). Schools serv- ing children across socioeconomic strata nationwide have unsafe lead levels in drinking water fountains (Wines et  al. 2016), because the Safe Drinking Water Act regulates water lead levels at water treatment facilities, but not at the tap. Updated “lead-free” plumbing rules maintain allowable lead content, and legacy plumbing and water infrastructure management can cause significant lead leaching. Well- publicized cases include schools in Washington, DC in 2000 and 2004, Seattle, WA in 2004, Flint, MI in 2014, Newark, NJ, New York, NY and Portland, OR in 2015 and 2016. The persistence of this issue is caused by failures within the enforcement, monitoring and evaluation, and adjudication domains, along with path dependences within all domains (figure 4a).

Heavy metals in light industry (score: 1.8/3). Bullseye Glass in Portland, Oregon creates art and architectural glass prod- ucts. Because of regulatory exemptions for small-scale industry, they lacked scrubbers and were releasing heavy metals, including known carcinogens such as cadmium and arsenic (Donovan et al. 2016). Portland residents filed eight complaints across multiple decades to the state Department of Environmental Quality, but no action was taken until the media was notified that the US Forest Service found levels of

Figure 3. Each indicator was evaluated by level of failure and success (y-axis) and criteria based on attributes of each indicator (x-axis).

is understood only from one perspective or is highly con- tested. Expertise is qualitatively evaluated as succeeding if there are multiple forms of knowledge and participants in addressing the risk, including those most affected by the decision; it is considered failing when only one form of expertise or knowledge is used and the perspectives of affected parties are disregarded. Path dependency is considered succeeding if the system is evolving to address emerging risks and challenges and is considered failing if it is regressing or failing to evolve despite an acknowledged need to do so.

We then apply this evaluative framework (figure 2) to five high-profile case studies. We chose a set of toxic substances case studies based on their representativeness within infrastructure systems (lead and SOx), manu- facturing (heavy metals in light industry), agriculture (glyphosate), and consumer products (bisphenol-a, BPA). For each case, we assembled literature reviewing the evi- dence base for each case, collected popular media accounts describing the policy responses, and examined relevant legislation and enabling policies of their regulation (sup- plement 1). Each coauthor described the conditions of a case by each domain, and then used the subjective scoring system to rank the robustness of risk framing, representa- tiveness of expertise, and the degree of path dependency per the evaluative framework in figure 3. For each case, we averaged the group’s scoring for risk, expertise, and path dependency within each governance domain to compare the perceived level of success or failure between cases, and then discussed the group’s findings to achieve consensus on a final ranking. Applying this evaluative framework draws out where failures in toxics governance are rooted in each case and where there are similarities and differences across cases.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

622 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

Figure 4. Application of indicators to five case studies across multiple chemical classes, including (a) lead in school drinking water, (b) heavy metal emissions from light industry, (c) sulfur and nitric oxide emissions, (d) bisphenol-a, and (e) glyphosate use in agriculture. For references, see supplement 1.

A

B

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 623

C

D

Figure 4. Continued.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

624 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

cadmium almost 50 times above Oregon’s benchmark of 0.6 nanograms per cubic meter during a moss air monitoring project (Donavan et al. 2016). A class action lawsuit against Bullseye Glass was subsequently filed, and a cease and desist order was issued for any uncontrolled furnaces. Regulatory gaps are now addressed by the Cleaner Air Oregon initiative, but still exist at the federal level. Based on our evaluative framework, we determined that risk evaluation for this case was moderately successful, and that the release of unsafe levels of heavy metals can mainly be attributed to failures of expertise and path dependency (figure 4b).

Sulfur and nitric oxides (score: 2/3). Sulfur oxides (largely SO2) result from burning sulfur or sulfur-containing materials, mostly coal, but present in all fossil fuels. Nitric oxides (NOx), which form during hydrocarbon combustion under an excess of oxygen, are both harmful to human health (affecting respiratory, cardiovascular, and neurological sys- tems) and the built and natural environment as the leading causes of acid rain and deposition (Likens 1974, McCubbin and Delucchi 1999). Risk of unsafe exposure still exists because of industry influence on policy design and monitor- ing and enforcement and because of failures of expertise and

path dependency within the policy evaluation and adjudica- tion domains (figure 4c).

Bisphenol-a in consumer products (score: 1.3/3). This endocrine disruptor (compound that interferes with proper hormone signaling) is present in many plastics to mitigate brittleness. BPA became notorious when scientists identified that it can leach into food and drinks and onto skin, and consumption or absorption of this compound, especially early in development, may increase cancer risk because of its endocrine-disrupting properties (e.g., Seachrist et al. 2016). Despite these findings, the use of BPA is still allowed in most products, although market pressure has resulted in its phaseout, and the FDA has removed it from the list of allowable additives in baby and children’s food and drink products. Although some risk has been mitigated thanks to moderately successful knowl- edge production and adjudication, failures are still pervasive around proper monitoring and enforcement because of privi- leging of expertise and systemic path dependency, and the use of replacement chemicals with uncertain toxicity (figure 4d).

Glyphosate in agriculture (score: 1.4/3). Glyphosate is the active ingredient in Roundup, one of the most commonly

E

Figure 4. Continued.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 625

applied pesticides in the United States, where laws require reasonable certainty of no harm as a prerequisite for pes- ticide certification. Industries typically determine such risk by assessing health effects at increasingly higher doses (dose–response); however, scientists have found that low, environmentally relevant concentrations of glyphosate can mimic and interfere with hormone signaling (endocrine disruption) and may also be associated with non-Hodgkin’s lymphoma (Mesnage et  al. 2015). Despite this growing body of scientific literature on the risks of glyphosate, the threshold for maximum glyphosate residues on food and animal feed—known as the tolerance level—continues to increase, and glyphosate was recertified for use in 2015 (Benbrook 2016). The persisting risk of glyphosate expo- sure can be attributed to pervasive failures across all gover- nance domains, particularly with respect to policy design, monitoring and enforcement, and evaluation. Industry influence over the governance processes has continued to affect ongoing adjudication processes; although harmed individuals have achieved some postharm compensation litigation is ongoing (figure 4e).

Paths forward Our analysis of a diverse set of failures to protect public and ecological health from toxic risks indicates a strong need to improve the overall governance of toxic chemical production and use throughout the United States. Four major patterns emerge from our analysis of governance failures. First, gov- ernance allowing the production and release of toxic chemi- cals with inadequate assurance of safety leads to inevitable harm to human and environmental systems. Second, certain forms of knowledge, particularly those that favor industry over public and environmental health, are privileged when assessing the extent and risk of this harm. Third, knowledge inequality is exacerbated by unequal formal mechanisms for resolving disputes over the assessment, mitigation, and redressing of harms. And lastly, path dependency of techno- logical, administrative, and knowledge-producing systems makes effective change difficult and perpetuates harm, despite regulatory action.

For each of these interrelated issues, we identify paths forward based on a reinterpretation of the purpose of toxic chemicals governance, provide examples of developing real world initiatives addressing them, and discuss chal- lenges to their continued development and success.

Issue 1: Incomplete paradigms of mitigation and risk management— the inevitability of harm from toxic chemical production. It is clear from our case studies of lead, glyphosate, SOx, heavy metals, and BPA that many toxic chemical risks are persistent both in their sources and their biological consequences. These risks often only become known after enough harm has accrued to communities to elicit a social response (Mesnage et al. 2015, Silbergeld et  al. 2015). Patterns of enforcement and adjudi- cation indicate that present regulatory processes generally only mitigate or act retroactively, not preventatively. At the

current rate of evaluation under TSCA, new chemicals are being manufactured faster than existing chemicals are being evaluated, especially those produced outside of the United States (Bernhard et  al. 2017). Even for regulated chemicals that have reporting requirements, existing data sets fail to communicate the frequency of chemical exposures or releases that are occurring, leading to enforcement failures as evidenced by the pervasive presence of toxic chemicals in global ecosystems and human populations (Schwarzenbach et al. 2010, Bernhardt et al. 2017). Good governance should therefore incorporate a paradigm shift around toxic chemi- cals management from one of mitigating risk to one of elimi- nating risk and supporting clean production to improve the long-recognized need for coordinated global and regional governance (Vogel 1997).

Principle 1: The right to be free from toxic chemical risks. Enshrining the right to be free from harm from toxic chemicals in policy will provide a clear articulation of our overall goals as a soci- ety with regards to what rights are sacrosanct and which can be negotiated (Hayward 2002). Ambitious policy goals of eliminating the production of toxic chemicals and support- ing the right of humans and ecosystems to be free from harm caused by toxic chemicals will enable transformation of the complex systems producing toxic chemical risks (Jasanoff 2004, Woodhouse 2006). Given the economic benefits that industrialized countries have already realized by engaging in compliance-based environmental regulation (Wallace 1995), further benefits could be realized by addressing the interdependent threats of anthropogenic climate change and global pollution, all while revitalizing US manufacturing and providing millions of jobs in the process (Bain et al. 2016).

To overcome the significant political, economic, and tech- nological inertia of addressing these interdependent threats, we can look to the precedent of using purity as a rhetorical tool for political mobilization and for overcoming industry special interests (Barkan 1985). Existing research recog- nizes a high degree of support for protecting environmental purity and human health across the political spectrum, despite ideological differences over the role of government in regulating businesses, requiring sustained public mobili- zation to enact significant legislative reform (Feinberg and Willer 2013). Although such mobilization can set a legisla- tive agenda for technological and economic evolution, a need remains for generating knowledge to enable systemic transformation (McCormick and Kautto 2013).

Issue 2: Biased and incomplete knowledge. Across our cases, we observed a consistent privileging of certain forms of knowledge in defining and managing risks, which generally favors biophysical laboratory science over field observa- tion, including epidemiological, anthropological, and social science accounts of experienced risk and harm. Even after harm becomes known, industry and responsible parties will consistently challenge accounts of harm while hiding behind the same scientific uncertainty that would cast doubt

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

626 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

on their initial risk assessments. This tactic is present at the forefront of litigation over glyphosate and lead in school drinking water. Our case studies mirror larger systemic problems in risk assessment, including affiliation bias in the risk assessment arena (Slovic 2016), targeted attacks on independent researchers (Reeves 2015), and the large vol- ume of industry-sponsored toxicological risk assessments (Hartung 2009).

Knowledge production around toxic substances in the United States remains fragmented by the physiochemical and toxicological properties of regulated materials. As chem- ical classes affect different exposure pathways, placement of chemicals within the overall economic system (e.g., during their production, release in the environment, or use in con- sumer products) is an important consideration in proper regulation. However, there is little systematic coordination in the production of knowledge of contaminant classes based on their chemical structure and mode of action, or how these classes are used and released. Current toxic chemical governance uses a narrow approach to knowledge production instead of, for example, evaluating substances on the basis of classes with shared chemical structure, such as organochlorines or brominated flame retardants, or even on the basis of shared mode of actions, such as level or type of carcinogenicity. A class-based approach may lead to more effective and efficient regulation and protection from chemi- cal risks (e.g., Sanderson et al. 2004), and has been partially adopted by the current TSCA.

The failure to include broad expertise in the gover- nance process has cascading effects: Policy design does not adequately prevent failures, and often does not provide architecture for effective monitoring and enforcement. As path dependency is rigid, effective policy evaluation is often nearly impossible. This means that adjudication is necessary to attempt to address grievances, whereas effective change is made difficult by poor policy design, lack of monitoring and enforcement, and the institutional challenges to quality policy evaluation, including major limitations on building a knowledge base for alternative chemical production.

Principle 2: Support diverse knowledge systems. Evolving the knowledge base entails supporting the generation and syn- thesis of diverse forms of knowledge for a more robust understanding of the complex nature of toxic chemical risks and the resulting sociotechnical transformations needed. At present, advances have been made in funding independent evaluations of chemical toxicity and in improving both labo- ratory and field-based methods for assessing toxic chemical risks. However, some promising technologies, such as the use of cell cultures and metabolic micro arrays instead of animal testing, could dramatically cut the costs of risk assessment but require sustained investment in order to penetrate a field dominated by animal testing (Hartung 2009). These advances in laboratory science should also be interdepen- dent with field-based, public health, and experiential knowl- edge of toxicity. Increases in knowledge generation and

synthesis about the impacts of toxic chemicals also need to inform and integrate research on alternative modes of clean production for substances of similar function, all which could be funded by implementing fees on the production of certain chemical classes (Thornton 2000). Building such a diverse knowledge system is not without its challenges, many of which can be overcome by providing an inclusive, representative, outcome focused, and independently evalu- ated research process for different classes of toxic chemicals (Reed et  al. 2014). However, as our case studies indicate, integrating diverse knowledge requires substantive changes throughout the rest of the governance system.

Issue 3: Uneven and unequal governance. In cases in which laboratory science presents significant evidence of risk of widely used chemicals, such as BPA, glyphosate, and lead in plumbing, unequal policy and enforcement mechanisms privilege the material interests of powerful actors over the health and well-being of communities and ecosystems. A lack of resources for adequate regulatory enforcement and policy and program implementation is symptomatic of the skewed priorities of the existing governance system. Some of our case studies exhibited partial success in one or more domains. For example, adjudication in glyphosate under the logic of compensation, allows for continued opera- tions, serving as a bandage to mitigate core weaknesses in policy design and enforcement. This model of governance disproportionately affects vulnerable populations, including children, elderly, low-income individuals, and future genera- tions in favor of industry (Elliott et al. 2004, Landrigan et al. 2017).

Part of the reason for this uneven and unequal governance is the influence industry has on shaping the present policy sphere. These types of failures result from targeting public opinion (Robbins 2007) and from lobbying and influencing legislators (Fredriksson et al. 2003, Hall and Deardorff 2006) to the point that legislation drafted by industry associations can become law (Potter 2011). This legislative capture is often reinforced by regulatory capture, occurring when an executive agency meant to protect the public interest instead protects the industry it regulates (Shapiro 2012). Arguments for this close relationship between regulators and industries hinge on the idea that the two entities are supposed to col- laborate to provide economic growth while protecting public values and interests (Lind 2015). By extension, the relatively limited influence on the policy process exerted by envi- ronmental and public health interest lobbyists, has caused them to invest in legal expertise, has resulting in a system of “regulation by litigation” (EPA 2017) by the “public interest law complex” (Lind 2015). Overall, these tensions highlight that although some adjudication can lead to substantive enforcement actions, without significant policy change and associated governance evolution, seeking financial redress from toxic industries may perversely promote increased or dirtier production as companies must finance compensatory penalties from their operating budgets.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 627

Principle 3: Inclusive, transparent, and accountable institu- tions. Overall, effective governance comes from increasing the representativeness and transparency of democratic pro- cesses, and allowing for the direct involvement of affected communities in policy design and implementation. Such a principle supports two primary initiatives: building a collaborative governance body and identifying cross-scale institutional links needed to address the complexity of con- temporary global industrialization.

The creation of a collaborative governance body may help alleviate some of the patterns we have highlighted. Successfully building a collaborative governance body involves bolstering participatory science approaches (e.g., citizen science programs) to narrow the science–policy gap. Specifically, a collaborative governance body would (1) consider and evaluate traditional ecological knowl- edge, scientific knowledge, and the experiential knowl- edge of affected communities (e.g., Bäckstrand 2003); (2) include diverse stakeholders in knowledge exchange (Reed et  al. 2014); and (3) engage procedural elements, such as independent moderation, to ensure a balance of power within the group (Purdy 2012). Collaborative governance is also mutualistic with collaborative knowledge produc- tion, and it decreases monitoring costs and increases industry accountability while empowering communities (Johnson et al. 2014).

Examples of such bodies presently exist, although not without their own challenges. As with toxic chemicals, our oceans are governed by a diversity of laws, regulations, and agencies. To address this fractured governance, the National Ocean Policy Act (NOPA) was passed in 2010, establish- ing the National Ocean Council, a collaborative body that includes representatives of the federal departments and agencies with major jurisdiction over the oceans to share resources and collaborate to implement policy. The NOPA marks the first national effort to implement a holistic, multiagency approach to managing our coasts and oceans, although contrasting priorities between executive adminis- trations have limited its effectiveness (Malakoff 2018). The implementation and challenges of the NOPA indicate that governance of complex social, environmental, and techno- logical systems requires operating horizontally across sectors (industries, media, academic scientists) and vertically across levels (communities, agencies, legislatures, national govern- ments; Cash et al. 2006).

Issue 4: Path dependency and inertia. The path dependency exhibited in each case results from the costs sunken into cer- tain means of production (i.e., technologies producing toxic risks), the persistence of many toxic materials, privileging of knowledge (Brown 1992), and the general absence of self- corrective behavior by industries, barring significant social influence. Current market logics enabled by state regulation have proven inadequate for internalizing the costs of produc- tion and have violated the economic principles of functioning markets (Haldane et  al. 2017). More troublingly, industry

priorities have continued to shape research and development toward minimizing costs and maximizing profits as opposed to alternative means of production (Woodhouse 2006).

In addition, many emergent risks are systemic, in that they emerge from complex interactions between society, the environment and technologies, such as SOx and NOx result- ing from automotive pollution. Although small techno- logical fixes such as improved catalytic converter technology and conversion to electric vehicles are possible, the aggregate influence of car-dependent suburban development has out- stripped gains from cleaner combustion technology. At the same time, innovation in some sectors has been shown to reduce risks from long-entrenched interests, evidenced by the grid purchasing power parity of wind farms over coal, facilitated by direct investments in research and develop- ment, and significant policy support for fledgling industries (Jenkins et al. 2010). A proper innovation-oriented approach can facilitate long-term system evolution, as opposed to the reactionary method of bureaucratizing risk that have led to systemic path dependencies that undermine sufficient toxic chemical governance.

Principle 4: Invest in innovation and real-world deployment. Policies need to identify and incentivize ways of producing sub- stances that meet the goals and needs of contemporary society without exposing people to toxic chemicals. Existing command and control, and other end-of-pipe regulations, although insufficient to protect human and environmental health, have stimulated extensive innovations in industry, creating jobs while improving human and environmental health (Wallace 1995, Pearce and Stillwell 2008). These models can be significantly improved using initiatives such as cradle-to-cradle manufacturing (Braungart et  al. 2007), the bioeconomy (McCormick and Kautto 2013), and the increasingly loud call for a Green New Deal (Jones 2009). Embracing such transitions will support our rights to a pure and high-quality environment. Economically, it will reduce and eventually eliminate compliance costs, increase labor productivity, provide greater long-run certainty over operational costs, reduce the economic burden of healthcare costs on society, and increase the economic advantage of US industries (Braungart et al. 2007, Jones 2009). Although some polluting industries may oppose such initiatives, the above arguments invalidate their rhetorical claims about the need to reduce regulations to protect jobs and economic advantage. In the face of such path dependency, it has become incumbent on the scientific community to evolve industries to eliminate harms from toxic chemicals, espe- cially given their role in accelerating the existential threat of rapid anthropogenic climate change.

Conclusions After over 40 years of modern environmental regula- tions, toxic chemical risks remain pervasive and largely unacknowledged in the United States, despite their sig- nificant negative impacts on public health, the economy,

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

628 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

and life-sustaining ecosystems. Crisis response and risk mitigation have pervaded environmental regulations around toxic chemicals. Effective toxic chemical governance will require sustained effort to produce better knowledge in the service of large-scale industrial and social transformations and the creation of inclusive governance bodies. A transi- tion to a regenerative economy that eliminates the concept of waste and permissible harm is urgently needed. To do so, researchers, industries, communities, policymakers, and the media must continue to craft collaborative visions and produce knowledge that enable public and private invest- ments in clean and ecologically sound technologies and land management practices. Evaluation of existing systems highlights research priorities for those seeking to transform governance to improve human and environmental health. By lifting the veil around the science and technology of producing and managing toxic chemical exposure risks, we can improve democratic governance and insure a healthier, economically robust, and equitable future for all.

Supplemental material Supplemental data are available at BIOSCI online.

Acknowledgments We thank Michael Weisdorf, Melanie Malone, Darrell Brown, Eugene Foster, and Vivek Shandas for their invaluable discus- sion, insight, and critique while developing this article. We additionally thank four anonymous reviewers for insightful feedback. This work was supported by the National Science Foundation Integrative Graduate Education and Research Traineeship (grant no.0966376: “Sustaining ecosystem ser- vices to support rapidly urbanizing regions”). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The authors declare no conflicts of interest.

References cited Bäckstrand K. 2003. Civic science for sustainability: Reframing the role

of experts, policy-makers and citizens in environmental governance. Global Environmental Politics 3: 24–41.

Bain PG et al. 2016. Co-benefits of addressing climate change can motivate action around the world. Nature Climate Change 6: 154.

Bao LJ, Wei YL, Yao Y, Ruan QQ, Zeng EY. 2015. Global trends of research on emerging contaminants in the environment and humans: A lit- erature assimilation. Environmental Science and Pollution Research 22: 1635–1643.

Barkan ID. 1985. Industry invites regulation: The passage of the Pure Food and Drug Act of 1906. American Journal of Public Health 75: 18–26.

Beck U. 2008. World at risk. Polity Press. Becker M, Edwards S, Massey RI. 2010. Toxic chemicals in toys and chil-

dren’s products: Limitations of current responses and recommendations for government and industry. Environmental Science and Technology 44: 7986–7991.

Benbrook CM. 2016. Trends in glyphosate herbicide use in the United States and globally. Environmental Sciences Europe 28: 3.

Bergeson LL, Graham MR. 2017. TSCA framework rules litigation update: Ninth circuit denies EPA’s motion to transfer to fourth circuit; fourth circuit expected to rule on petitioners’ transfer motions imminently.

TSCA Blog. Bergeson & Campbell, P.C. www.tscablog.com/entry/ tsca-framework-rules-litigation-update-ninth-circuit-denies-epas- motion-to.

Bernhardt ES, Rosi EJ, Gessner MO. 2017. Synthetic chemicals as agents of global change. Frontiers in Ecology and the Environment 15: 84–90.

Bocking S. 2004. Nature’s experts: Science, politics, and the environment. Rutgers University Press.

Boone MD et al. 2014. Pesticide regulation amid the influence of industry. BioScience 64: 917–922.

Botos A, Graham JD, Illés Z. 2018. Industrial chemical regulation in the European Union and the United States: A comparison of REACH and the amended TSCA. Journal of Risk Research (April 2018). https://doi. org/10.1080/13669877.2018.1454495.

Braungart M, McDonough W, Bollinger A. 2007. Cradle-to-cradle design: Creating healthy emissions: A strategy for eco-effective product and system design. Journal of Cleaner Production 15: 1337–1348.

Brown P. 1992. Popular epidemiology and toxic waste contamination: Lay and professional ways of knowing. Journal of Health and Social Behavior 33: 267–281.

Brulle RJ, Pellow DN. 2006. Environmental justice: Human health and environmental inequalities. Annual Review of Public Health 27: 103–124.

Bullard RD, Johnson GS. 2009. Environmental justice grassroots activism and its impact. Journal of Social Issues 56: 555–578.

Caliman FA, Gavrilescu M. 2009. Pharmaceuticals, personal care products, and endocrine disrupting agents in the environment: A review. Clean Soil, Air, Water 37: 277–303.

Cartwright N. 1999. The Dappled World. Cambridge University Press. Cash D, Adger WN, Berkes F, Garden P, Lebel L, Olsson P, Pritchard L,

Young O. 2006. Scale and cross-scale dynamics: Governance and infor- mation in a multilevel world. Ecology and Society 11: 2045–2065.

[CW] Chemical Watch. 2019. TSCA Highlights (22 April 2019). https:// chemicalwatch.com/tsca-highlights.

Cutler D, Dominici F. 2018. A breath of bad Air: Cost of the Trump envi- ronmental agenda may lead to 80,000 extra deaths per decade. JAMA 319: 2261–2262.

Davies JC, Mazurek J. 1998. Pollution Control in the United States: Evaluating the System. Routledge.

Donovan GH, Jovan SE, Gatziolis D, Burstyn I, Michael YL, Amacher MC, Monleon VJ. 2016. Using an epiphytic moss to identify previously unknown sources of atmospheric cadmium pollution. Science of the Total Environment 559: 84–93.

Douglas M, Wildavsky A. 1983. Risk and Culture: An Essay on the Selection of Technological and Environmental Dangers. University of California Press.

Elliott MR, Wang Y, Lowe RA, Kleindorfer PR. 2004. Environmental jus- tice: Frequency and severity of US chemical industry accidents and the socioeconomic status of surrounding communities. Journal of Epidemiology and Community Health 58: 24–30.

[EPA] Environmental Protection Agency. 2017. Administrator Pruitt Issues Directive to End EPA “Sue & Settle” (31 July 2018). EPA Press Office. www.epa.gov/newsreleases/ administrator-pruitt-issues-directive-end-epa-sue-settle.

Farber DA. 1999. Taking slippage seriously: Noncompliance and creative compliance in environmental law. Harvard Environmental Law Review 23:297.

Feinberg M, Willer R. 2013. The moral roots of environmental attitudes. Psychological Science 24: 56–62.

Fletcher AL. 2009. Clearing the air: The contribution of frame analysis to understanding climate policy in the United States. Environmental Politics 18: 800–816.

Frank R. Lautenberg Chemical Safety for the 21st Century Act. Public Law no. 114–182 (22 June 2016). US Government Publishing Office. www. congress.gov/114/plaws/publ182/PLAW-114publ182.pdf.

Fredriksson PG, List JA, Millimet DL. 2003. Bureaucratic corruption, envi- ronmental policy and inbound US FDI: Theory and evidence. Journal of Public Economics 87: 1407–1430.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

https://academic.oup.com/bioscience August 2019 / Vol. 69 No. 8 • BioScience 629

Grabowski ZJ, Denton A, Rozance MA, Matsler M, Kidd S. 2017. Removing dams, constructing science: Coproduction of undammed riverscapes by politics, finance, environment, society and technology. Water Alternatives 10:769–795.

Haldane JBS, Bowes MD, Palmer KL. 2017. Using economic incentives to regulate toxic substances. Routledge.

Hayward T. 2002. Environmental Rights as Democratic Rights. Pages 237–256 in Minteer BA, Taylor BP, eds. Democracy and the Claims of Nature. Rowman and Littlefield Publishers.

Hall RL, Deardorff AV. 2006. Lobbying as legislative subsidy. American Political Science Review 100: 69–84.

Hanna-Attisha M, LaChance J, Sadler RC, Champney Schnepp A. 2016. Elevated blood lead levels in children associated with the flint drink- ing water crisis: A spatial analysis of risk and public health response. American Journal of Public Health 106: 283–290.

Hartung, T. 2009. Toxicology for the twenty-first century. Nature 460: 208. [HEI] Health Effects Institute. 2019. State of Global Air 2019. Health Effects

Institute. Hejny J. 2018. The Trump administration and environmental policy: Reagan

redux? Journal of Environmental Studies and Sciences 8: 197–211. Hess DJ. 2015. Publics as threats? Integrating science and technology stud-

ies and social movement studies. Science as Culture 24: 69–82. Hoffman AJ. 1999. Institutional evolution and change: Environmentalism and

the US chemical industry. Academy of Management Journal 42: 351–371 Jasanoff S. 2004. States of Knowledge: The Co-production of Science and

the Social Order. Routledge. Jenkins J, Swezey D, Borofsky Y. 2010. Where Good Technologies Come

from: Case Studies in American Innovation. Breakthrough Institute. www.thebreakthrough.org/archive/american_innovation.

Jones V. 2009. The Green Collar Economy: How One Solution Can Fix Our Two Biggest Problems. Harper Collins.

Johnson MF, Hannah C, Acton L, Popovici R, Karanth KK, Weinthal E. 2014. Network environmentalism: Citizen scientists as agents for envi- ronmental advocacy. Global Environmental Change 29: 235–245.

Landrigan PJ et al. 2017. The Lancet Commission on pollution and health. Lancet 391: 462–512.

Latour B. 2004. Why has critique run out of steam? From matters of fact to matters of concern. Critical Inquiry 30: 225–248.

Likens GE, Bormann FH. 1974. Acid rain: A serious regional environmental problem. Science 184: 1176–1179.

Lind M. 2015. Raiding progress. The Breakthrough Institute. https:// t h e bre a kt h ro u g h . or g / i n d e x . p hp / j o u r n a l / p a s t - i s s u e s / i s s u e - 5 / raiding-progress.

Lynch MJ, Stretesky PB. 2014. Exploring Green Criminology: Toward a Green Criminological Revolution. Routledge.

McCormick K, Kautto N. 2013. The bioeconomy in Europe: An overview. Sustainability 5: 2589–2608.

Mesnage R, Defarge N, De Vendomois JS, Seralini GE. 2015. Potential toxic effects of glyphosate and its commercial formulations below regulatory limits. Food and Chemical Toxicology 84: 133–153.

McCubbin DR, Delucchi MA. 1999. The health costs of motor-vehicle- related air pollution. Journal of Transport Economics and Policy 33: 253–286.

Mol AP. 2016. The environmental nation state in decline. Environmental Politics 25: 48–68.

Morello-Frosch R, Shenassa ED. 2006. The environmental “riskscape” and social inequality: Implications for explaining maternal and child health disparities. Environmental health perspectives 114: 1150.

Nelson J. 2013. Proposition 65 in Plain Language. Office of Environmental Health Hazard Assessment. www.oehha.ca.gov/proposition-65/ general-info/proposition-65-plain-language.

Paavola J. 2006. Interstate water pollution problems and elusive federal water pollution policy in the United States 1900–1948. Environment and History 12: 435–465.

Pastor M, Morello-Frosch R. 2018. Gaps matter: Environment, health, and social equity. Generations 42: 28–33.

Pearce A, Stilwell F. 2008. “Green-collar” jobs: Employment impacts of climate change policies. Journal of Australian Political Economy 62: 120–138.

Perrow C. 1984. Normal Accidents: Living with High Risk Technologies. Basic Books.

Pierson P. 2000. Increasing returns, path dependence, and the study of poli- tics. American Political Science Review 94: 251–267.

Potter W. 2011. Green Is the New Red: An Insider’s Account of a Social Movement under Siege. City Lights Publishers.

Purdy JM. 2012. A framework for assessing power in collabora- tive governance processes. Public Administration Review 72: 409–417.

Rasli A, Qureshi MI, Isah-Chikaji A, Zaman K, Ahmad M. 2018. New tox- ics, race to the bottom and revised environmental Kuznets curve: The case of local and global pollutants. Renewable and Sustainable Energy Reviews 81: 3120–3130.

Reed MS, Stringer LC, Fazey I, Evely AC, Kruijsen JHJ. 2014. Five principles for the practice of knowledge exchange in environmental management. Journal of Environmental Management 146: 337–345.

Reeves C. 2015. Of frogs and rhetoric: The Atrazine wars. Technical Communication Quarterly 24: 328–348.

Revesz RL. 2019. Institutional pathologies in the regulatory State: What Scott Pruitt taught us about regulatory policy. Journal of Land Use and Environmental Law 34. Forthcoming.

Ripple WJ, Wolf C, Newsome TM, Galetti M, Alamgir M, Crist E, Mahmoud MI, Laurance WF. 2017. World scientists’ warning to humanity: A sec- ond notice. BioScience 67: 1026–1028.

Robbins P. 2007. Lawn People: How Grasses, Weeds, and Chemicals Make Us Who We Are. Temple University Press.

Rogers P, Hall AW. 2003. Effective Water Governance. Global Water Partnership. TEC Background Papers no. 7.

Rudel RA, Perovich LJ. 2009. Endocrine disrupting chemicals in indoor and outdoor air. Atmospheric Environment 43: 170–181.

Schwarzenbach RP, Egli T, Hofstetter TB, Von Gunten U, Wehrli B. 2010. Global water pollution and human health. Annual Review of Environment and Resources 35:109–136.

Scott JC. 1998. Seeing like a State: How Certain Schemes to Improve the Human Condition Have Failed. Yale University Press.

Seachrist DD, Bonk KW, Ho SM, Prins GS, Soto AM, Keri RA. 2016. A review of the carcinogenic potential of bisphenol A. Reproductive Toxicology 59: 167–182.

Shapiro SA. 2012. The complexity of regulatory capture: Diagnosis, cau- sality, and remediation. Roger Williams University Law Review 102: 221–257.

Silbergeld EK, Mandrioli D, Cranor CF. 2015. Regulating chemicals: Law, science, and the unbearable burdens of regulation. Annual Review of Public Health 36: 175–191.

Slovic P. 2016. The Perception of Risk. Routledge. Sanderson H, Johnson DJ, Reitsma T, Brain RA, Wilson CJ, Solomon KR.

2004. Ranking and prioritizing of environmental risks of pharmaceu- ticals in surface waters. Regulatory Toxicology and Pharmacology 39: 158–183.

Stockholm Convetion. 2017. Global Monitoring Plan for Persistent Organic Pollutants: Second Global Monitoring Report. United Nations Environment Programme. http://chm.pops.int/Implementation/ GlobalMonitoringPlan/MonitoringReports/tabid/525/Default.aspx

Thornton J. 2000. Beyond risk: An ecological paradigm to prevent global chemical pollution. International Journal of Occupational and Environmental Health 6: 318–330

Stehr N. 2015. Knowledge Politics: Governing the Consequences of Science and Technology. Routledge.

Stewart RB. 1995. United States environmental regulation: A failing para- digm. JL and Com 15: 585.

Sydow J, Schreyögg G, Koch J. 2009. Organizational path dependence: Opening the black box. Academy of Management Review 34: 689–709.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Overview Articles

630 BioScience • August 2019 / Vol. 69 No. 8 https://academic.oup.com/bioscience

Vogel D. 1997. Trading up and governing across: Transnational governance and environmental protection. Journal of European Public Policy 4: 556–571.

Vogel SA, Roberts JA. 2011. Why the toxic substances control act needs an overhaul, and how to strengthen oversight of chemicals in the interim. Health Affairs 30: 898–905.

Wagner W. 2007. Commons Ignorance: The Failure of Environmental Law to Provide the Information Needed to Protect Public Health and the Environment. Vandeplas.

Wallace D. 1995. Environmental Policy and Industrial Innovation: Strategies in Europe, the USA, and Japan. Earthscan Publications.

Wilson MP, Schwarzman MR. 2009. Toward a new US chemicals policy: Rebuilding the foundation to advance new science, green chemistry, and environmental health. Environmental health perspectives 117: 1202.

Wines M, McGeehan P, Schwartz J. 2016. Schools nationwide still grapple with lead in water. New York Times (26 March 2016). www.nytimes. com/2016/03/27/us/schools-nationwide-still-grapple-with-lead-in- water.html.

Woodhouse E. 2006. Nanoscience, Green Chemistry, and the Privileged Position of Science. Pages 148–184 in Frickel S, Moore K, eds. The

New Political Sociology of Science: Institutions, Networks, and Power. University of Wisconsin Press.

Wynne B. 2003. Misunderstood Misunderstandings: Social identities and the public uptake of science. Pages 281–304 in Irwin A, Wynne B, eds. Misunderstanding Science: The Public Reconstruction of Science and Technology. Cambridge University Press.

Wynne B. 2016. Misunderstood misunderstanding: Social identities and public uptake of science. Public Understanding of Science 1: 281–304.

Ariana M. Chiapella ([email protected]) and Elise F. Granek are affiliated with the Department of Environmental Science and Management at Portland State University, in Portland, Oregon. Zbigniew J. Grabowski is affiliated with the Cary Institute of Ecosystem Studies, in Millbrook, New York, and with the Urban Systems Lab, at The New School, in New York, New York. Mary Ann Rozance is affiliated with the College of the Environment at the University of Washington, in Seattle. Ashlie D. Denton is affiliated with the Hatfield School of Government, at Portland State University, in Portland, Oregon. Manar A. Alattar is affiliated with the Department of Biology at Portland Community College, in Portland, Oregon.

D ow

nloaded from https://academ

ic.oup.com /bioscience/article-abstract/69/8/615/5523266 by 81695661, O

U P

on 08 A ugust 2019

Copyright © 2019 American Institute of Biological Sciences. Copyright of BioScience is the property of American Institute of Biological Sciences and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.