U.S. flag

An official website of the United States government

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Institute of Medicine (US) Committee on State Food Labeling; Porter DV, Earl RO, editors. Food Labeling: Toward National Uniformity. Washington (DC): National Academies Press (US); 1992.

Cover of Food Labeling

Food Labeling: Toward National Uniformity.

Show details

ICase Study: Requirements for Labeling Bottled Water

In the course of its deliberations, the Committee identified a number of instances in which individual States acted in the absence of a Federal regulation. One of the best examples of this practice concerns bottled water. The Committee prepared the following case study of bottled water to examine the type of problems that some States believe to have existed and to explore their regulatory response.

In the past several years, the regulation of bottled water by the Food and Drug Administration (FDA) has been seen by some as an unfortunate paradigm of the effects of deregulation.1 For its part, FDA has candidly stated that it does not believe that bottled water poses a risk to the public health and so has not devoted substantial resources to its regulation. That decision apparently has not been shared by 23 States, which have adopted statutes or regulations governing the quality or labeling of bottled water. The recent worldwide recall of Perrier mineral water following benzene contamination was seen by some as both proof of MAN ineffective regulation (the problem was originally discovered in North Carolina) and the industry's inability to police itself. Hearings conducted by the House Committee on Energy and Commerce's Subcommittee on Oversight and Investigations criticized FDA for its regulation of bottled water (U.S. Congress, 1990; U.S. Congress, 1991), following a U.S. General Accounting Office (GAO) report (GAO, 1991) that took the same position.

The regulation of bottled water requires the consideration of two provisions of the Federal Food, Drug, and Cosmetic Act of 1938, as amended (FDCA), that Congress has directed (through the Nutrition Labeling and Education Act of 1990; NLEA) be studied before preemption decisions can be made. The requirements were that a food not subject to a standard of identity bear a common or usual name [FDCA Section 403(i)(1)], and the requirement that a food meet any applicable standard of quality or disclose on its label that it does not [FDCA Section 403(h)(1)].

In examining FDA's regulation of bottled water, several ambiguities arose. State standards of identity are preempted if they are not identical to corresponding Federal standards. Similarly, State common or usual name regulations are preempted if it is determined that FDA is adequately implementing the common or usual name requirement of FDCA. But does NLEA permit a State to adopt a standard of identity for a product for which no Federal standard exists? The superficial answer would appear to be yes, since NLEA and the regulations proposed to date do not specifically prohibit a State from doing so. Yet the preemption provision applicable to the common or usual name requirement prohibits ''any requirement for the labeling of a food of the type required'' by FDCA Section 403(i)(1). It is unclear, however, whether (1) FDA must have established a specific common or usual name for bottled water, or (2) the general provisions of FDCA Section 403(i)(1) and its implementing regulations are sufficient for preemption of any such State requirements. If it is found that FDA is adequately implementing the common or usual name requirement of the statute under either circumstance, and thus preempting this area of State requirements, the question still remains as to whether a State can name a food by issuing a standard of identity.

Another issue surrounds the peculiarities of the bottled water quality standard. Bottled water is the only food for which FDA has adopted a standard of quality in the absence of a standard of identity. Quality standards usually do not deal with issues of food safety, and State regulation of food safety is not preempted by NLEA. Therefore, ambiguity also surrounds the question of whether States can regulate the safety of bottled water in a regulation that is called a quality standard, or whether they can regulate bottled water by calling the regulation one of food safety when in fact it covers the same ground as the FDA quality standard.

Although these problems ultimately must be faced and resolved by FDA, their resolution is viewed as beyond the scope of this case study. As discussed in Chapter 4, the Committee has decided to leave to FDA the decision of whether a State law or regulation is a standard of identity or a common or usual name regulation, and determination of the consequences that flow from that decision.

FDA Adoption of Standard of Quality and Good Manufacturing Practice Regulations

FDCA Section 401 (a) [21 USC §341(a)] authorizes FDA to establish for any food definitions and standards of identity, reasonable standards of quality, and reasonable standards of fill of container. A food that is subject to a quality standard is misbranded if it does not conform to that standard or does not declare on its label that it does not conform [FDCA Section 403(h)(1)]. Until the passage of NLEA, the adoption or amendment of quality standards for all products required the use of formal rulemaking under FDCA Section 701(e).2

Exercising its authority under FDCA Section 401, FDA established quality standards for bottled water in 1973. In proposing the standard, the agency explained that "[b]ottled water is increasingly being used as a source of drinking water.... The consumer expects bottled water to meet the minimum criteria established for public drinking water supplies" (FDA, 1973a; p. 1019).

The quality standards were based on the 1962 Public Health Service standards for public drinking water supplies. FDA noted then that the recently created Environmental Protection Agency (EPA) had assumed the responsibility for establishing drinking water standards and that FDA intended to revise the bottled water standards to keep them compatible with EPA drinking water standards.

FDA received 33 comments on the proposed standards. These comments raised most of the issues about bottled water that continue to be of concern today. One comment suggested that FDA adopt more stringent standards than those adopted by EPA because many consumers assume that bottled water is of a higher quality than tap water. FDA replied:

Although some consumers may assume, and some promotion of bottled water may encourage the assumption that bottled water is of a higher quality then tap water, them is no Federal requirement to this effect. The quality of tap water and bottled water can vary widely due to the source itself ... as well as to treatments these waters may receive during processing. Because of these source and treatment variables them is no basis for assuming that bottled water is of a higher quality than municipal tap water (FDA, 1973b; p. 32558).

Several comments also suggested that bottled mineral water be subject to the quality standard. FDA concluded that mineral water was "inherently different' from bottled water and that a separate quality standard would be developed. Optimistically, the agency stated that the lack of a quality standard for mineral water would be a "temporary situation" (FDA, 1973b; pp. 32558–32559).

Several State health agencies objected to the proposal because it did not provide for the safety of bottled water. FDA replied that "if bottled water contains any substance at a level injurious to health, it will be deemed to be adulterated and appropriate regulatory action will be taken, whether or not it meets the standard of quality" (FDA, 1973b; p. 32559).

As noted earlier, FDA departed from its usual procedure when it first proposed a standard of quality for bottled water without at the same time proposing a standard of identity. The agency's initial Federal Register notice did not explain the rationale for this decision, but several commenters, urged FDA to regulate the use of such terms as "spring," "well," and "distilled water.'' FDA declined and stated:

[T]here is no need for a requirement that the source or treatment of the water be declared on the label of bottled water. Bottled drinking water can be produced from various sources of water, and various types of treatment of the water can be used in manufacturing bottled water of an acceptable quality. If the manufacturer decides to provide information in the labeling or in advertising relating to bottled water, stating or implying that it is the product of a specific source of water or that the water has been treated in a specific manner, such information must be truthful, factual, and not misleading in any respect. Sec. 403(a) of the act provides that a food shall be deemed to be misbranded if its labeling is false or misleading in any particular. The Commissioner concludes that this statutory authority is sufficient to provide for regulatory action in instances where false or misleading statements concerning the source or treatment of bottled water am made and that specific statements to this effect in the standard are unnecessary (FDA, 1973b; p. 32561).

In addition to citing the general prohibition against false and misleading label claims in FDCA Section 403(a), FDA also could have relied on FDC Section 403(i), which deems a food misbranded unless it bears its common or usual name. If the source of a particular bottled water was a municipal water supply but the water was labeled as "spring water," it would not only be labeled in a false or misleading manner but would also be misbranded for failing to bear its common or usual name. Although objections to the final rule were filed, FDA concluded that they did not justify changing the regulation or conducting a hearing, and the quality standard became effective on June 19, 1975 (FDA, 1975).

In 1974, passage of the Safe Drinking Water Act codified the division of labor for regulating water between FDA and EPA. In addition to directing EPA to promulgate national primary drinking water standards, the Safe Drinking Water Act also added Section 410 to FDCA. Section 410 directs FDA to consult with EPA whenever the latter issues interim or revised national primary drinking water standards and, within 180 days of EPA's promulgation, either amend the bottled water standard or explain its rationale for not doing so in the Federal Register. From 1975 to 1979, FDA met or came close to meeting the 180-day time limit. Since then, however, FDA has not even come close to meeting the statutory deadline for acting on EPA actions.

In 1975, FDA adopted Good Manufacturing Practice (GMP) regulations for bottled water, including bottled mineral water (21 CFR §129.1 et seq.). Among other things, these regulations specify the kinds of facilities that must be used and the process controls that are required to ensure a safe product. The regulation also requires that "product water" (i.e., the water that is to be bottled) come from an approved source. "Approved source" is defined as a source of water and the water itself "that has been inspected and the water sampled, analyzed, and found to be of a safe and sanitary quality according to applicable laws and regulations of State and local government agencies having jurisdiction" [21 CFR §129.3(a)]. The regulation also requires that bottled water manufacturers test source water (from other than a public water system) and bottled water as often as necessary but including at least weekly for microbiological contamination and at least yearly for chemical contamination. Testing of source water ''shall be consistent with the minimum requirements set forth in [the bottled water quality standard]" [21 CFR §129.35(a)(3)(ii)].

On January 29, 1988, the International Bottled Water Association (IBWA) filed a citizen petition with FDA seeking the amendment of the bottled water quality standards to define the various kinds of bottled water, including mineral water. IBWA also sought amendment of the bottled water GMPs to require source testing and mandatory annual testing by all producers.

Congressional Scrutiny

The Subcommittee on Oversight and Investigations of the House Committee on Energy and Commerce held a hearing on FDA oversight of bottled water on April 10, 1991. In a memorandum dated April 9 to members of the Subcommittee on Oversight and Investigations outlining the expected testimony, the chairman, John D. Dingell (D-Mich.), said that testimony would show that

FDA has, in three major ways, abdicated its responsibility to regulate the product: by tailing to promptly adopt quality standards, as required by statute; by failing to define bottled water products, despite an 18-year old pledge to do so; and by failing to consider the bottled water industry's proposal to adopt a regulatory model code to assure consumer confidence (Dingell, 1991).

The fundamental difference between the way FDA has traditionally viewed bottled water and the way the subcommittee and perhaps Consumers now view the product was evidenced by the following statement by Chairman Dingell:

Finally, this hearing will address whether consumers can be confident that what they are purchasing is pristine, pure water, possessing unique characteristics that make it inherently superior to tap water by virtue of its source and its physical properties (Dingell, 1991).

From FDA's point of view, however, it had never purported to regulate bottled water as a product with "unique characteristics that make it inherently superior to tap water." To the contrary, FDA had rejected such an implication in 1973 when it adopted the quality standard for bottled water. It stated then that there is no Federal requirement that bottled water be of higher quality than tap water and "there is no basis for assuming that bottled water is of a higher quality than municipal tap water" (FDA, 1973b). Thus, the subcommittee's criticism seemed largely based on FDA's failure to regulate bottled water in the way the subcommittee now believes it should be regulated, rather than in the way FDA had consistently maintained it would be regulated.

The subcommittee's criticism of FDA also included what appeared to be an unfair assertion that there were 22 recalls of bottled water "for reasons ranging from the detection of mold to contamination by kerosene" (Dingell, 1991). An examination of these recalls shows that the actual situation was not nearly so dire. Five of the 22 were of one product—the benzene-contaminated Perrier mineral water, which was available in unflavored form and in four flavors. These cases were Class II recalls. Similarly, six Class III recalls (the lowest level of regulatory action) concerned mold and yeast contamination of six flavors of the same bottled water product. These products were sweetened, thus making them flavored soda waters rather than bottled waters. Although they were arguably misbranded for failing to state their common or usual name, they were not subject to the bottled water quality standard and should not have been counted as such. Two other manufacturers recalled five products. The subcommittee's list also included a recall of club soda even though club soda is not traditionally considered a bottled water nor is it subject to the bottled water quality standard. A Class I recall of isopropyl alcohol was listed as involving bottled water because it was erroneously labeled as distilled water. This case was a drug recall, not a recall of bottled water (Dingell, 1991). When all of these factors are taken into account, it appears that bottled water recalls were instituted by only seven manufacturers in 1990, an unexceptional record for an industry with more than 450 producers. For the purpose of this study, only six recalls involved actual or potential misbranding charges.

FDA's failure to adopt the industry's requested regulatory scheme does not necessarily mean that the agency was in error. Industry rarely seeks Federal regulation unless it perceives some economic benefit in doing so. In a letter to an association member, the IBWA president stated that "once we have succeeded in establishing federal regulation of our industry through the FDA, we will be in a position to make the strongest public statement possible that we are the most regulated beverage sold in the United States" (Dingell, 1991, attachment G). While being able to make that statement might be helpful for marketing purposes, Americans consume far more colas and soda waters annually, and these beverages are subject to neither the bottled water quality standard nor the bottled water GMPs. They are consumed without health hazard and not regulated at the level of oversight that the bottled water industry is seeking, apparently for marketing purposes.

A GAO report on FDA's regulation of bottled water, prepared at the request of the subcommittee, took FDA to task for failing to observe the time limits set by Congress for adopting EPRs primary drinking water standards (GAO, 1991). GAO also criticized FDA for inspecting bottled water plants only on the average of every 5-3/4 years, failing to require bottlers to use certified laboratories or to report results to FDA (authority that the agency claims it does not have), and failing to define terms frequently found on bottled water labels, such as "spring," "pure," and other quality or source claims. GAO recommended that FDA seek legislation authorizing it to use certified laboratories to test water and to report those results to the agency. It also suggested that Congress might wish to amend FDCA Section 410 to automatically adopt EPA's standards, unless FDA acts.

The GAO report also suggested that the bottled water industry is causing considerable potential health hazards to the public. However, FDA believes that the evidence supports its view that, by and large, the bottled water supply in this country is safe and adequately labeled and simply does not deserve a higher level of regulatory scrutiny.

Fred Shank, director of FDA's Center for Food Safety and Applied Nutrition, gave candid testimony to the subcommittee about the way FDA uses its limited resources to police the U.S. food supply. Dr. Shank (1991) stated unequivocally that

... we have no reason to question the safety of bottled water. Based on our experience, FDA considers bottled water to have a low potential for contamination or for causing sickness ... (p. 1).

Bottled water establishments are included under the general food safety program. In this program bottled water plants generally are assigned low priority for inspection. FDA bases its priorities on factors such as the potential for public health problems and the violation rate of the industry. When compared to products such as low acid canned foods and products where Listeria or Salmonella have a significant potential to develop, bottled water products must take a back seat. Our experience over the years ... has not shown that there is a significant problem with bottled water products... (p. 12).

[W]e believe bottled water is a safe product. As an industry, bottled water has fewer compliance problems than most other food industries. That is why FDA classifies bottled water as a low risk product. Even the compliance problems the bottled water industry has with FDA have not been of a type that can be classified as a hazard (p. 20).

Notwithstanding this vigorous defense of FDA's actions, it is not surprising that the congressional attention has caused FDA to rethink some of its positions. The subcommittee was told that the agency was reconsidering the coverage of the bottled water standards to include mineral water and the bottled water component in flavored beverage products fabricated from bottled water ingredients. It is also considering a new quality standard requiring the water component of seltzer, tonic water, colas, and similar products to meet quality standards based on EPA's primary drinking water standards. The agency has not spoken to the need for standards of identity or common or usual names.

State Actions

A total of 23 States have expressed their dissatisfaction with FDA's regulation of bottled water by adopting laws or regulations to provide additional controls. Although these State laws vary, in general, they address two basic issues: the nomenclature of various types of bottled waters and the purity of the products. (These laws and regulations are summarized in Appendix H.) Many States have adopted the Association of Food and Drug Officials' (AFDO) model bottled water regulation, or some variation of it (AFDO, 1984). The AFDO model regulation contains definitions for the following waters: artesian well, bottled, demineralized, drinking, light mineral, mineral, mineralized, natural, purified, spring, and well. The model regulation requires that all bottled waters bear one of the defined names and that artificially carbonated waters disclose the addition of carbonation. It also requires compliance with the FDA standard of quality, the bottled water GMPs, and source water and finished product sampling (21 CFR Part 129).

In communications to the Committee, a number of States expressed discontent with FDA's inaction in the alleged face of false and misleading labeling claims for bottled water products. Even a cursory examination of supermarket shelves confirms that there are a number of products that could be the subject of FDA enforcement action, based on their labeling. Although the recent growth of the bottled water industry has made the public and regulators more aware of potentially false and misleading labeling, it can be argued that FDA's prioritization of its enforcement resources is sound. In any event, the Committee has determined that whether FDA actually enforces particular statutory provisions is beyond the scope of its inquiry.

Assessment of Adequacy and Conclusions

Much of the recent controversy, and many of the States' regulatory initiatives, have focused on the safety of bottled water. Although of great importance, this issue falls outside of the scope of NLEA and, therefore, this study. On the other hand, the same States that have expressed concern over Federal quality standards have also believed it necessary to establish nomenclature requirements. Some States and consumer groups believe that the opportunity for public confusion has increased by virtue of the increased number of products in the market and the increasingly aggressive claims made for these products.

In 1973, FDA believed it was evident that bottled water was not any better or purer than tap water. That conclusion may still be factually valid, but the Committee questions whether that view is held by consumers after years of exposure to advertising claims of superiority for bottled and mineral water. As mentioned above, the fact that nearly half the States have established definitions for the different types of bottled and mineral waters on the market is evidence that there is a perception that FDA's efforts here have not been adequate. Although the general misbranding provisions of FDCA could have been used by FDA to prosecute many of the perceived offenders, it is clear that the States believed that the existence of definitions in the form of standards or common or usual names would make their enforcement job easier. Based on the Committee's working principles, it can be concluded that the State laws and regulations that define and/or standardize the names of the various kinds of bottled and mineral waters are appropriately candidates for Federal adoption.


  • AFDO (Association of Food and Drug Officials). 1984. AFDO Model Bottled Water Code, adopted 1984. York, Pa.: AFDO.
  • Dingell, J.D. 1991. Memorandum to members of the Subcommittee on Oversight and Investigations, House Committee on Energy and Commerce, U.S. Congress. April 9.
  • FDA (Food and Drug Administration). 1973. a. Bottled Water, Proposed Quality Standards. Fed. Reg. 38:1019–1020; Jan. 8.
  • FDA (Food and Drug Administration). 1973. b. Bottled Water, Quality Standards; Addition of Flouride and Current Good Manufacturing Practice Regulations; Final Rule. Fed. Reg. 38: 32557–32565; Nov. 26.
  • FDA (Food and Drug Administration). 1975. Quality Standards for Bottled Water, Final Rule. Fed. Reg. 40:21932–21934; May 20.
  • GAO (U.S. General Accounting Office). 1991. Stronger FDA Standards and Oversight Needed for Bottled Water. GAO, Washington, D.C.
  • Shank, F.R. 1991. Testimony of Fred R. Shank, Director, Center for Food Safety and Applied Nutrition, Food and Drug Administration, before the Subcommittee on Oversight and Investigations, House Committee on Energy and Commerce, US. Congress. April 10.
  • U.S. Congress, House. 1990. Proceedings of the Bottled Water Workshop, September 13–14, 1990. A report for the Subcommittee on Oversight and Investigations, Committee on Energy and Commerce. Committee print 101-X Washington, D.C.
  • U.S. Congress, House. 1991. Bottled Water Regulation. Hearing before the Subcommittee on Oversight and Investigations, Committee on Energy and Commerce. Transcript 102–36. Washington, D.C.


1. FDA considers "bottled water" to be "water that is sealed in bottles or other containers and intended for human consumption. Bottled water does not include mineral water or any type of soft drink commonly known as sods water, which is made by adsorbing carbon dioxide in potable water" [21 CFR §103.35(a)]. Most State laws dealing with bottled water, as well as the Association of Food and Drug Officials' model bottled water regulation, include mineral water within their ambit. Unless the context dictates otherwise, this paper includes mineral water within the term "bottled water.''

2. NLEA Section 8 amends FDCA to permit the issuance and amendment of standards of identity, quality, and fill of container for food products other than dairy products and maple syrup by notice-and-comment rulemaking under FDCA Section 701(a).

Copyright © National Academy of Sciences.
Bookshelf ID: NBK234832


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (5.6M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...