This paper is extracted from the report “Managing The Effects of Nanotechnology” and has two purposes. The first is to describe the menu of possibilities for government action to deal with the adverse effects of nanotechnology. If there are important alternatives that are not described here it is because of inadvertence or ignorance.
The second purpose is to provide evidence relevant for determining what needs to be done to manage nanotechnology. When I began this work, my initial assumption was that there was no need for new statutory authority. As I learned more about the unique aspects of nanotechnology and thought more about the weaknesses of existing statutes, I was increasingly led to the conclusion that a new law is needed. This paper, however, is not an advocacy piece for a new law. It would have been written quite differently if that were its purpose. Rather, it is a policy analysis, intended to give the reader the information relevant for thinking about a course of action.
In short, the paper is intended to inform, not persuade.
Existing Regulatory Authorities
In the U.S. political system, it has never been easy to pass new laws regulating commercial products. In the current political climate, it is close to impossible. Thus, those who see a need to address the health and environmental effects of nanotechnology have focused their attention on existing regulatory authorities.
These authorities are examined below. For each of the major laws, I look at the extent to which that law can be applied to nanotechnology, and at the strengths and weaknesses of the regulatory regime if it were applied.
It is important to keep in mind that adequate legal authority is not the only requirement for a successful regulatory program. Adequate resources of personnel and money and the will to use the resources and authority also are necessary. Many of the programs discussed below do not have the resources necessary to fulfill their legal obligations. I will note some particularly egregious examples in the discussion of individual programs.
Toxic Substances Control Act (TSCA)
President Nixon’s Council on Environmental Quality originally proposed TSCA as a way of dealing with numerous fears—founded and unfounded—related to toxic chemicals.
Enacted in 1976, its primary goals were to prevent new chemicals from being marketed without adequate safeguards, and to fill gaps in the regulation of existing chemicals. The law was passed with lukewarm support by a Democratic Congress.
TSCA’s coverage is quite broad, with the result that it is considered the primary vehicle for regulating nanotechnology. A broad coalition of environmental groups, including the Natural Resources Defense Council and Greenpeace, has called for regulating nanotechnology under the provisions of TSCA.
TSCA’s coverage is broad in at least two respects. First, unlike most of the environmental statutes, it is not limited to covering one part of the environment (air,water, etc.). Thus, its regulatory reach encompasses all potential adverse effects—not just those occurring in some particular medium. This is appropriate for nanotechnology materials and products, the effects of which are not likely to be limited to a particular part of the environment.
Second, the law is directed at “chemical substances and mixtures” broadly defined as “any organic or inorganic substance of a particular molecular identity”.
Types of products specifically covered by other statutes—such as pesticides, drugs, cosmetics, and medical devices—are excluded from TSCA’s coverage. The environmental coalition cited above asserts that “all engineered nanomaterials are ‘new chemical substances’ under TSCA” and should be regulated as such.
Even if nanotechnology materials do not meet the TSCA criterion for being new chemicals, there is another provision of the act that could bring them under its regulatory umbrella. This is the provision for “significant new use” rules (SNURs). To simplify somewhat, the EPA Administrator can declare that an existing chemical be regulated as if it were a new chemical if it is put to uses that might change its effects, for example, by increasing human exposure.The new use provisions would not be a feasible method of regulating nanotechnology if each particular nanomaterial had to be subject to a SNUR because that approach would require an unrealistically large amount of time and resources. However, TSCA provides that any action that can be taken with respect to individual chemicals also can be taken with respect to “categories” of chemicals. Categories are defined to include a group of chemicals that are similar in “physical properties”, a definition that would seem adequate to cover nanotechnology materials because nanotechnology materials share similar physical properties, such as size.
Perhaps the most challenging problem raised by trying to include nanotechnology materials under TSCA is the importance of structure. Even assuming that nanotechnology materials are within TSCA’s jurisdiction at the stage when a chemical is shaped into a basic nanotechnology form (e.g., nanodot, nanotube), the exposure and toxicity from the ultimate product cannot be predicted at that point. Thus, efforts to deal with the adverse effects of nanotechnology must focus on the stage of the ultimate product. However, it may be unreasonable and impractical to expect the manufacturers of basic nanotechnology forms to keep track of all possible uses. Furthermore, by the next stage in the process when the nanotechnology material is combined into larger structures and/or other materials, it may have lost the molecular identity that defines what TSCA covers.
Labeling on the basic nanotechnology form that requires reporting of uses - as well as tracking of the nanotechnology supply chain—may provide a partial solution to this difficulty. But this is a clear example of the problems that arise when trying to apply existing legal categories to nanotechnology products.
Currently, TSCA’s new chemical notification requirements exempt several categories of chemicals. The law authorizes the EPA Administrator to make exemptions, but what is exempted is defined by rules promulgated by the Administrator. The exemption most relevant to nanotechnology exempts chemicals produced in volumes of 10,000 kilograms or less per year (Code of Federal Regulations, Title 40, Ch.1, part 723.5). This is equivalent to anything less than about 11 tons a year, and would exclude almost all nanotechnology products. It clearly does not make sense if TSCA is to be applied to nanotechnology. EPA would need to amend the low-volume exemption to exclude nanotechnology materials, but crafting the exclusion would be challenging.
TSCA is a law with dramatic strengths and weaknesses. Its strengths are the flexible broadness of its coverage and the wide range of measures it allows to be taken to deal with chemical risks. The measures include almost any conceivable requirement the EPA Administrator would want to impose.
TSCA’s strengths are, not coincidentally, balanced by hedges and obstacles that make it difficult for EPA to take action. The entire law is premised on the balancing of risks and benefits. Such balancing invites controversy and litigation. The act stacks the deck against EPA in litigation in at least three important ways.
First, the technical standard of judicial review in the act is: “supported by substantial evidence in the rulemaking record”. This standard is very difficult to meet, and it contrasts with the much easier “arbitrary and capricious” standard applied to the Clean Air Act, Clean Water Act and most other environmental statutes. The result is that it is very difficult for EPA to defend rules promulgated under TSCA.
A second weakness is that TSCA implicitly assumes that no knowledge about a chemical means that there is no risk. The most relevant section is the epitome of a “Catch 22.” It states that if EPA does not have enough information “to permit a reasoned evaluation of the health and environmental effects of a chemical,” it can delay or prohibit its manu facture only if it can show that the chemical “may present an unreasonable risk” - which is precisely the thing that it cannot show.
There is another criterion that in theory can be used for EPA action. This is that the chemical will be produced in “substantial quantities” and that there will be significant environmental or human exposure. In practice, this criterion only rarely can be used, because most new chemicals initially are produced in small volumes, and because the likelihood of significant exposure is difficult to establish. The problem is even greater for nanotechnology materials because quantity or volume may not be a relevant indicator of potential risk.
A third problem is that the act requires EPA to meet a variety of requirements before it can regulate a chemical. The difficulty of these requirements was illustrated dramatically by the Corrosion Proof Fittings case that struck down EPA’s proposed regulation of asbestos, in part because of inadequate analysis - even though EPA had spent 10 years doing the analysis to support the regulation.
For example, because TSCA requires that a proposed regulation be the “least burdensome” regulation, the court criticized EPA for not analyzing the costs and benefits of all other possible ways of regulating asbestos.
These and other shortcomings contained in the act are sufficient to make TSCA a weak regulatory instrument. Furthermore, as with most government regulatory programs, there simply are not enough people in the toxic substances office to perform the tasks required by the law.
Occupational Safety and Health Act (OSHAct)
The OSHAct was passed in 1970 and has been amended frequently since. The act combined the occupational safety programs housed in the Department of Labor with the occupational health functions exercised by the Department of Health, Education and Welfare (now Health and Human Services).
The Department of Labor was given responsibility for administering the OSHAct. The OSHAct’s basic mechanism is for the Occupational Safety and Health Administration (OSHA) in the Department of Labor to set standards and to enforce the standards through inspections and penalties for noncompliance. An occupational safety and health standard is defined in the act as “a standard which requires conditions, or the adoption or use of one or more practices, means, methods, operations or processes, reasonably necessary or appropriate to provide safe or healthful employment and places of employment.” This language is certainly broad enough to cover nanotechnology.
The difficulties with using the OSHAct to deal with nanotechnology are the same that arise with most of the environmental statutes. Detection of nanotechnology products requires expensive and sophisticated equipment, and it is often unclear which parameters are the relevant ones to measure from the standpoint of toxicity. For practical purposes, whether in the setting of a factory or the ambient environment, detection and control methods (e.g., filters) may not be currently available or may be too expensive or too cumbersome.
One further OSHA weakness should be noted. Like EPA, OSHA traditionally has been starved for resources. In FY 1980 there were 2,950 OSHA employees. Twenty-five years later, with a greatly expanded economy and a larger number of workplaces, there were 2,208 OSHA employees. EPA has been able to perform its basic functions because state agencies do most of the labor-intensive work, such as inspections and enforcement. However, for implementing the OSHAct, only 21 states have OSHA-approved plans that allow them to enforce OSHA standards. Three additional states have approved state plans for protecting state and local government employees only. There are millions of workplaces in the United States, and the chance of being visited by an OSHA inspector is not high.
Food, Drug, and Cosmetic Act (FDCA)
The FDCA is a one-hundred-year-old act originally passed to prevent poisonings from quack patent medicines and to clean up grossly unsanitary conditions in food processing plants. Over the years, it has been frequently amended, its scope expanding to cover an increasing number of areas.
Five types of nanotechnology applications are within the purview of the FDCA: drugs, medical devices, biologics, cosmetics and food. (Medical devices are, roughly, any mechanical thing used for treatment or diagnosis of disease.)
Drugs, Biologics and Medical Devices
Drugs, biologics and medical devices are regulated quite differently from cosmetics. Drugs, biologics and medical devices must receive approval from the Food and Drug Administration (FDA) before they can be sold. The approval process is time-consuming and rigorous. The burden of proof is on the manufacturer to show that the product is safe.
The process has been criticized by some as too time-consuming, and by others as not sufficiently rigorous. Recently, there have been instances of political interference in FDA decisions, notably the decision to delay approval of the morning-after birth control pill. However, overall, the process for approving drugs, biologics and medical devices works reasonably well.
Cosmetics are quite a different story. Although the FDCA has a lot of language devoted to cosmetics, it is not much of an exaggeration to say that cosmetics in the United States are essentially unregulated. The FDCA prohibits the marketing of “adulterated or misbranded” cosmetics in interstate commerce. “Adulterated” is roughly defined as injurious to health.
“Misbranded” means that the label is false or misleading or does not contain required information. However, manufacturers of cosmetics are not required to register with FDA, are not required to file data on product ingredients, and are not required to report cosmetic-related injuries to FDA. If by some chance FDA discovers a cosmetic that is adulterated or misbranded, it has no authority to recall the product or take action against the manufacturer. All FDA can do is ask the Justice Department to bring suit to have the product removed from the market.
FDA regulation of food focuses on packaging and food additives. As with FDA’s regulation of drugs, there have been some intense food-related controversies (e.g., saccharin).
FDA’s determination not to regulate genetically modified food as a food additive has been harshly criticized and the agency has been faulted for its inadequate monitoring of pesticide residues on food. However, FDA’s legal authority under FDCA is adequate, and there is not a gross disparity between resources available and the regulatory tasks to be performed.
In light of these facts, it appears that nanotechnology based drugs, biologics and medical devices - and probably nanotechnology-based food additives and packaging - are best regulated under the FDCA authorities. On the other hand, although it would be neater legally and bureaucratically to regulate nanotechnology cosmetics under FDCA, the public would be better protected by regulating cosmetics under some alternative regime.
There are many more environmental laws than most people realize—at least several hundred, even using a narrow definition of “environmental.” The vast majority deal with very specific, narrow subjects, or they are minor amendments to existing laws. The three major environmental statutes are the Clean Air Act (CAA), the Clean Water Act (CWA) and the Resource Conservation and Recovery Act (RCRA).
The basic mechanism of CAA and CWA is to set standards and to enforce them through permits issued to pollution sources. The CAA standards are mostly ambient and emission standards, and the CWA standards are mostly technology-based. But RCRA is different. Instead of regulating pollution sources, it sets technology standards for disposal sites and establishes a “cradle-to-grave” reporting system for hazardous wastes.
The preceding one-paragraph description of three very long (CAA is 448 pages) and complex laws is obviously a gross oversimplification.
However, it is hard to imagine how these laws could be used to manage the adverse effects of nanotechnology products. As was noted above in the section on the OSHAct, it is difficult to detect nanomaterials except with sophisticated laboratory equipment. If these materials cannot be detected, the provisions of the environmental laws are inoperable. Furthermore, even if these materials can be detected, the only practical control in many situations would be to impose a complete ban on their release into the environment. Such a ban probably could be better handled through a product-oriented law like TSCA. The technology based parts of the environmental statutes would be inoperable because there is no agreed-upon best available technology (BAT) for the removal of nanotechnology particles from air, water or waste streams.
The processes used to manufacture nanotechnology materials present a different situation. A recent study examined the potential risks from manufacturing five types of nanomaterials and compared the risks with those from a half dozen nonnano processes ranging from petroleum refining to wine making. The risks from the two types of processes (nano and non-nano) were found to be in the same general range – a finding that is not surprising given that the manufacture of nanotechnology materials uses inputs and processes similar to those in other industries.
The study excluded risks attributable to the nanomaterials themselves. While existing environmental laws presumably can handle ordinary emissions from plants manufacturing nanotechnology materials, it is not clear whether those laws can deal with emissions of nanotechnology materials.
It may be necessary to invoke the environmental laws if disposal or leakage from manufacturing facilities is a significant source of nanotechnology materials in the environment. From what we know now, use and disposal of nanotechnology products is probably a much greater source of environmental exposure.
This brings us back to TSCA-like statutes. TSCA section 6 authorizes the EPA Administrator to prohibit or otherwise regulate “any manner or method of disposal” of a hazardous substance or mixture “by its manufacturer or processor or by any other person who uses, or disposes of, it for commercial purposes.” But this is probably only a theoretical point. In fact, because of other provisions in TSCA, it is unlikely that the provision would be invoked or, if it were, that it would be upheld in the courts.
Applying Existing Authorities
Several kinds of incremental steps could be taken to manage nanotechnology within the existing legal framework. This section explores three approaches: coordinating existing laws; amending existing laws to deal with nanotechnology; and strengthening existing laws. These are in no way mutually exclusive.
A coordinated and comprehensive program for dealing with the effects of nanotechnology that is based on existing laws would: clarify which laws would govern in particular circumstances; specify how different laws would relate to one other if more than one applied; and ensure that there were no gaps or significant duplications in coverage. Regulation of biotechnology provides relevant experience for establishing such a program.
On June 26, 1986, the President’s Office of Science and Technology Policy (OSTP) issued a “Coordinated Framework for Regulation of Biotechnology”. The framework delineates the policies and responsibilities of six different agencies with respect to regulating biotechnology research and products.
It allocates jurisdiction and responsibility among the agencies both for approving commercial biotechnology products and for conducting biotechnology research. Wherever possible, it gives responsibility to a single agency for reviewing product safety. Where this is not possible, it designates a lead agency, and mandates consolidated or coordinated reviews.
As part of the same effort, the White House established an OSTP-chaired Biotechnology Science Coordinating Committee consisting of representatives from each of the agencies. Among other things, the committee helped get some agreement on consistent definitions of the genetically engineered organisms subject to review - a key element in any coordinated approach. The committee also established the principle that agencies should use scientific reviews of comparable rigor.
The biotech Coordinated Framework has been in place for 20 years, and, as noted by a 2004 comprehensive review, “The adequacy of the Coordinated Framework has been a matter of disagreement from the beginning.” Interestingly, the review report notes that, “While the policy remains that genetically engineered products should receive the same regulatory treatment as similar, conventionally produced products, in practice agencies have developed a hybrid system that effectively treats biotechnology products differently. In part, this evolution has resulted from the difficulty of fitting biotechnology products into pre-existing legal categories…and in part due to the perceived public interest in affording GE [genetically engineered] products greater scrutiny”. The same evolution could take place with nanotechnology products.
In theory, a framework for nanotechnology could be established similar to the biotechnology framework. However, nanotechnology probably covers an even broader array of types of products than does biotech. Also, it is too early to know the types of products that will incorporate nanotechnology, so an agreement might be more difficult. A greater obstacle is the difference—with respect to physical and biological behaviour – between nanotechnology products and larger-scale products of the same molecular composition. EPA cannot regulate carbon nanotubes as if they were just small amounts of carbon compounds, because the nanotubes behave quite differently in the environment, and probably in the human body as well. The same is true of other types of nanotechnology products. This means that an effective regulatory mechanism must involve internal changes within each of the agency programs and not just allocate responsibilities among agencies.
The current federal coordinating mechanism for nanotechnology is the Nanoscale Science Engineering and Technology (NSET) subcommittee of the National Science and Technology Council Committee on Technology. NSET has established a Nanotechnology Environmental and Health Implications Working Group. The subcommittee and the working group have focused on research and development. NSET could evolve to deal with regulatory and policy issues as well as research, but it does not have the membership or authority to do so now.
Amending Existing Laws
Some of the aforementioned problems – and questions about whether particular laws (e.g., TSCA) cover nanotechnology - could be addressed by amending individual statutes to make clear that they do cover nanotechnology. While this would avoid a lot of litigation, winning the approval of numerous congressional committees and the executive branch for such legislation would be a formidable challenge.
There also would be the major substantive problem of deciding how to define what is covered. Is it possible to define nanotechnology just by the size of the material? What if the nanotechnology material is combined with a non-nanotechnology material? If one manufacturer makes carbon nanotubes and another manufacturer makes a textile that incorporates the tubes, do you regulate both? If the nanotubes are used in a medical device, what role would FDA play? What happens in the future when nanotechnology is combined with genetic engineering? While it might be possible to arrive at reasonable answers to these kinds of questions, it would require a lot of time, effort and foresight.
Strengthening Existing Laws
Even assuming that existing laws could be amended to clarify and make explicit their coverage of nanotechnology - and that the patchwork of existing laws could be stitched together in a coordinated framework that would perform better than it has for biotech - one still would be left with the weaknesses contained in these laws.
In the current political climate, it is inconceivable that these weaknesses could be remedied. It would be easier, politically and substantively, to draft and enact a new law focused on nanotechnology. I turn to an outline of such a law in the next section.
Even if there is agreement that there should be a new law, it will take months—and probably years—before a new law is enacted. In the interim, regulators would have to use existing laws in the best possible way. Thus, how the current laws can be applied to nanotechnology, and what resources will be required to effectively apply them, are inescapable and important issues.
This article has been broken into parts due to its original length. The parts can be found as follows: