[Federal Register Volume 70, Number 110 (Thursday, June 9, 2005)]
[Rules and Regulations]
[Pages 33694-33701]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 05-11406]


=======================================================================
-----------------------------------------------------------------------

DEPARTMENT OF HEALTH AND HUMAN SERVICES

Food and Drug Administration

21 CFR Part 165

[Docket No. 2004N-0416]


Beverages: Bottled Water

AGENCY: Food and Drug Administration, HHS.

ACTION: Final rule.

-----------------------------------------------------------------------

SUMMARY: The Food and Drug Administration (FDA) is amending its bottled 
water quality standard regulations by revising the existing allowable 
level for the contaminant arsenic. As a consequence, bottled water 
manufacturers are required to monitor their finished bottled water 
products for arsenic at least once each year under the current good 
manufacturing practice (CGMP) regulations for bottled water. Bottled 
water manufacturers are also required to monitor their source water for 
arsenic as often as necessary, but at least once every year unless they 
meet the criteria for the source water monitoring exemptions under the 
CGMP regulations. This final rule will ensure that the minimum quality 
of bottled water, as affected by arsenic, remains comparable with the 
quality of public drinking water that meets the Environmental 
Protection Agency's (EPA's) standards.

DATES: This rule is effective January 23, 2006. The Director of the 
Office of the Federal Register approves the incorporation by reference 
in accordance with 5 U.S.C. 552(a) and 1 CFR part 51 of certain 
publications in 21 CFR 165.110(b)(4)(iii), as of January 23, 2006.

FOR FURTHER INFORMATION CONTACT: Jennifer A. Burnham, Center for Food 
Safety and Applied Nutrition (HFS-306), Food and Drug Administration, 
5100 Paint Branch Pkwy., College Park, MD 20740, 301-436-2030.

SUPPLEMENTARY INFORMATION:

I. Background

    In the Federal Register of January 22, 2001 (66 FR 6976), EPA 
published a final rule issuing a National Primary Drinking Water 
Regulation (NPDWR) containing a maximum contaminant level (MCL) of 0.01 
milligram per liter (mg[sol]L) or 10 parts per billion (ppb) and a 
Maximum Contaminant Level Goal (MCLG) of zero for arsenic to address 
potential public heath effects from the presence of arsenic in drinking 
water. This rulemaking finalized a proposed rule that EPA published in 
the Federal Register of June 22, 2000 (65 FR 38888). EPA's effective 
date of March 23, 2001, for this rule was temporarily delayed for 60 
days to a new effective date of May 22, 2001, in accordance with the 
memorandum of January 20, 2001, from the Assistant to the President and 
Chief of Staff, entitled ``Regulatory Review Plan'' (66 FR 7702, 
January 24, 2001). On May 22, 2001, EPA announced that it would further 
delay the effective date for the rule until February 22, 2002, to allow 
time to complete a reassessment of the information on which the revised 
arsenic standard is based. On February 22, 2002, the arsenic MCL of 
0.01 mg[sol]L in public drinking water rule became effective, and water 
systems must comply with the new standard for arsenic in public 
drinking water by January 23, 2006. On March 25, 2003 (68 FR 14501 at 
14503), EPA revised the rule text in its January 2001 final rule that 
established the 10 ppb arsenic drinking water standard to express the 
standard as 0.010 mg[sol]L, in order to clarify the implementation of 
the original rule. EPA made this change in response to a concern raised 
by a number of States and other stakeholders that State laws adopting 
the Federal arsenic standard as 0.01 mg[sol]L might allow rounding of 
monitoring results above 0.01 mg[sol]L so that the effective standard 
(in consideration of rounding of results) would be 0.014 mg[sol]L (or 
14 ppb), not 0.010 mg[sol]L (10 ppb).
    Under section 410(b)(1) of the Federal Food, Drug, and Cosmetic Act 
(the act) (21 U.S.C. 349(b)(1)), FDA is required to issue a standard of 
quality regulation for a contaminant in bottled water not later than 
180 days before the effective date of an NPDWR issued by EPA for a 
contaminant under section 1412 of the Safe Drinking Water Act (SDWA) 
(42 U.S.C. 300g-1), or make a finding that such a regulation is not 
necessary to protect the public health because the contaminant is 
contained in water in public water systems but not in water used for 
bottled water. The effective date for any such standard of quality 
regulation is to be the same as the effective date of the NPDWR. In 
addition, section 410(b)(2) of the act provides that a quality standard 
regulation issued by FDA shall include monitoring requirements that the 
agency determines to be appropriate for bottled water. Further, section 
410(b)(3) of the act requires a quality standard for a contaminant in 
bottled water to be no less stringent than EPA's MCL and no less 
protective of the public health than EPA's treatment technique 
requirements for the same contaminant.
    In accordance with section 410 of the act, FDA published in the 
Federal

[[Page 33695]]

Register of December 2, 2004 (69 FR 70082), a proposal to adopt EPA's 
MCL for arsenic as an allowable level in the quality standard for 
bottled water. In the 2004 proposal, FDA tentatively concluded that the 
MCL that EPA had established based on available toxicological 
information for arsenic in public drinking water was adequate for the 
protection of public health. As a consequence, bottled water 
manufacturers would be required to monitor their finished bottled water 
products for arsenic at least once each year under the CGMP regulations 
for bottled water. Bottled water manufacturers would also be required 
to monitor their source water for arsenic as often as necessary, but at 
least once every year unless they meet the criteria for the source 
water monitoring exemptions under the CGMP regulations. Interested 
persons were given until January 31, 2005, to submit comments.

II. Comment on the Proposed Rule

    FDA received four letters, each containing one or more comments, in 
response to the December 2, 2004, proposal. The comments were received 
from two trade associations and two consumers. Two letters generally 
support the proposal with one containing comments suggesting 
modifications to various provisions of the Analysis of Economic Impacts 
section. The agency's responses to these suggestions are addressed in 
that section. Two letters raised issues that are outside the scope of 
this rulemaking (the appropriate agency to regulate bottled water and 
EPA's requirements for testing frequency) and therefore are not 
addressed here.

III. Conclusion

    The agency is adopting the allowable level for arsenic in the 
quality standard for bottled water as proposed (69 FR 70082). 
Therefore, FDA is establishing in Sec.  165.110(b)(4)(iii)(A) (21 CFR 
165.110(b)(4)(iii)(A)), which includes allowable levels for inorganic 
substances, an allowable level for arsenic at 0.010 mg[sol]L and 
removing the existing entry for arsenic in Sec.  165.110(b)(4)(i)(A).
    With respect to analytical methods for the determination of 
chemical contaminants, FDA is making the following changes in Sec.  
165.110(b)(4)(iii). In the new Sec.  165.110(b)(4)(iii)(E)(14), FDA is 
incorporating by reference EPA approved analytical methods (66 FR 6975 
at 6988) for determining compliance with the quality standard for 
arsenic in bottled water. These methods are contained in the manual 
entitled ``Methods for the Determinations of Metals in Environmental 
Samples-Supplement 1,'' EPA/600/R-94/111, May 1994, which is 
incorporated by reference in accordance with 5 U.S.C. 552(a) and 1 CFR 
part 51. The source for this manual containing the two methods is the 
National Technical Information Service (NTIS), PB95-125472, U.S. 
Department of Commerce, 5825 Port Royal Rd., Springfield, VA 22161. FDA 
believes that these methods are sufficient to use for determining the 
level of arsenic in bottled water.
    Therefore, upon the effective date of this rule, January 23, 2006, 
any bottled water that contains arsenic at a level that exceeds the 
applicable allowable level will be deemed misbranded under section 
403(h)(1) of the act (21 U.S.C. 343(h)(1)) unless it bears a statement 
of substandard quality as provided by Sec.  165.110(c)(3).

IV. Environmental Impact

    The agency has previously considered the environmental effects of 
this rule as announced in the proposed rule (69 FR 70082, December 2, 
2004). No new information or comments have been received that would 
affect the agency's previous determination that there is no significant 
impact on the human environment and that an environmental impact 
statement is not required.

V. Analysis of Economic Impacts

A. Final Regulatory Impact Analysis

    FDA has examined the economic implications of this final rule as 
required by Executive Order 12866. Executive Order 12866 directs 
agencies to assess all costs and benefits of available regulatory 
alternatives and, when regulation is necessary, to select regulatory 
approaches that maximize net benefits (including potential economic, 
environmental, public health, public safety, and other advantages; 
distributive impacts; and equity). Executive Order 12866 classifies a 
rule as significant if it meets any one of a number of specified 
conditions, including: Having an annual effect on the economy of $100 
million, adversely affecting a sector of the economy in a material way, 
adversely affecting competition, or adversely affecting jobs. A 
regulation is also considered a significant regulatory action if it 
raises novel legal or policy issues. The Office of Management and 
Budget (OMB) has determined that this final rule is a significant 
regulatory action as defined by Executive Order 12866.
    This final regulatory impact analysis revises the analysis set 
forth in the proposed rule (69 FR 70082) in response to comments 
received. Except as we indicate below, the analysis in this final rule 
is the same as the analysis of the proposed rule.
1. Need for Regulation
    We did not receive any comments on the discussion of the need for 
regulation in the analysis of the proposed rule. To briefly summarize 
the discussion in the analysis of the proposed rule, under section 410 
of the act, when the Environmental Protection Agency (EPA) issues a 
regulation establishing an MCL for a particular contaminant in drinking 
water, FDA is required to issue a standard of quality regulation for 
that contaminant in bottled water or make a finding that such a 
regulation is unnecessary to protect the public health. FDA's quality 
standard must also include appropriate monitoring requirements. If FDA 
does not issue a quality standard for arsenic in bottled water by 180 
days before the effective date of EPA's regulations or make a finding 
that such a regulation is not necessary to protect the public health, 
then EPA's regulation becomes applicable to bottled water as well as 
drinking water.
2. Regulatory Options
    We considered five regulatory options in the analysis of the 
proposed rule:
    Option One--Re-establish a quality standard for arsenic in bottled 
water that maintains the current allowable level of 0.05 mg[sol]L.
    Option Two--Take no action. Under this option, EPA's regulation on 
arsenic in drinking water would become applicable to bottled water.
    Option Three--Establish a quality standard for arsenic in bottled 
water that adopts EPA's MCL for arsenic in drinking water of 0.010 
mg[sol]L. Under this option, bottled water producers would be subject 
to CGMP monitoring requirements in 21 CFR 129.35 and 129.80.
    Option Four--Establish a quality standard for arsenic in bottled 
water that sets the allowable level of arsenic at 0.02 mg[sol]L.
    Option Five--Establish a quality standard for arsenic in bottled 
water that sets the allowable level of arsenic at 0.005 mg[sol]L.
    One comment stated that bottled water should be regulated by EPA, 
not FDA. This comment maintained that economies of scale suggest that 
EPA should oversee bottled water as well as tap water and that it is 
wasteful for us to spend public money to change bottled water 
regulations in a way that mirrors EPA's regulations for tap water.

[[Page 33696]]

 This option is outside the scope of this rulemaking and would not be 
legally feasible at this time. In addition, changing the jurisdiction 
of bottled water from FDA to EPA would generate costs in addition to 
cost savings. We do not have information suggesting that the net 
benefits of this option would be likely to be greater than the net 
benefits of the options that we considered in the proposed rule. 
Therefore, we have not addressed this option in this analysis.
3. General Comments
    (Comment 1) One comment argued that some bottled water 
establishments may need to purify their water using reverse osmosis or 
other methods in order to meet an allowable level of 0.010 mg[sol]L. 
This comment suggested that these establishments would need to change 
the identification of their products from ``spring water'' to 
``purified water.'' The comment noted that this could lead to a loss of 
utility for consumers who prefer spring water if they have limited 
choices for home or office water delivery and can no longer obtain 
spring water from other establishments. Finally, the comment noted that 
this change in how consumers value bottled water could reduce sales for 
the establishments producing that water, although the comment noted 
that that it was unable to estimate this cost.
    (Response) The comment is correct that if bottled water 
establishments need to adopt treatment methods that require them to 
change the identity of their product, (e.g., from ``spring water'' to 
``purified water'') then some consumers might place a lower value on 
that water. If some consumers choose not to continue to consume the 
water after the identity change, then some bottled water establishments 
could face a decline in sales and profits. We would classify any loss 
of profit from shifts in consumer demand as a distributive impact 
rather than a social cost because the sales and profit losses for some 
firms would be offset by countervailing sales and profit increases for 
other firms. The comment did not provide sufficient information to 
estimate the loss in consumer utility or the distributive impact on 
industry. Although the comment only discussed this effect in relation 
to an allowable level of 0.010 mg[sol]L (corresponding to Options 2 and 
3), the same effect might also be relevant to any reduction in the 
current allowable level of 0.05 mg[sol]L, including reductions to 
allowable levels of 0.02 mg[sol]L (Option 4) and 0.005 mg[sol]L (Option 
5). The likelihood of this effect would be greater the lower the 
allowable level. Thus, this effect would be largest under Option 5 and 
smallest under Option 4.
    (Comment 2) One comment suggested that our estimate of the benefits 
of specifying a maximum arsenic level of 0.005 mg[sol]L was too high 
because EPA based their benefit estimate on a flawed interpretation of 
the available data, and we based our benefit estimate on EPA's benefit 
estimate. This comment cited a report issued by the National Academy of 
Sciences (NAS) that the comment claimed concluded that arsenic does not 
cause bladder or lung cancer at levels up to 0.05 mg[sol]L (50 ppb) in 
drinking water (Ref. 1). Although the comment made this point in 
relation to our benefit estimates for Option 5 (allowable level of 
0.005 mg[sol]L), it is also relevant to our benefit estimates for 
Options 2 and 3 (allowable level of 0.01 mg[sol]L) and Option 4 
(allowable level of 0.02 mg[sol]L). The comment also argued that EPA 
based its risk assessment on extrapolating cancer risks from relatively 
high levels of arsenic investigated in some epidemiological studies to 
the relatively low levels that EPA considered when setting an MCL for 
arsenic. According to the comment, the NAS study highlighted the 
uncertainty associated with this extrapolation and also suggested there 
might be a threshold below which arsenic in water would not increase 
the risk of cancer at all. The comment noted that EPA reviewed this 
study prior to issuing a regulation establishing a MCL of 0.01mg[sol]L. 
Finally, the comment cited an article that was published after EPA's 
regulation that reportedly found no association between bladder cancer 
and arsenic in drinking water at a level of 0.10 mg[sol]L and another 
article that ostensibly made a similar point.
    (Response) The 2001 NAS study concluded that arsenic in drinking 
water increases the risk of bladder or lung cancer at concentrations at 
least as low as 0.003 mg[sol]L (Ref. 2). Although the study noted that 
a threshold was theoretically possible, it noted that there was no 
experimental data to identify a threshold and concluded that any 
threshold was likely to occur below concentrations that are relevant to 
the U.S. population (Ref. 3). The study did note that there was 
insufficient mode-of-action data on arsenic to provide a biological 
basis for using either a linear or nonlinear extrapolation to estimate 
low dose health risks and that the choice of extrapolation method was, 
in part, a policy decision (Ref. 4). However, the study supported the 
use of a linear extrapolation in conjunction with a discussion of the 
uncertainties associated with that approach (Ref. 5). EPA acknowledged 
uncertainty about the impact of reducing arsenic to the levels under 
consideration in this rule and carried out a sensitivity analysis to 
reflect that uncertainty. The range of potential benefits that we 
estimated in the proposed rule reflects that uncertainty. However, EPA 
could not have considered the two articles cited in the comment that 
were published after the publication of the regulation. One of these 
articles found no increase in the risk of bladder cancer from arsenic 
in drinking water at levels up to 0.1 mg[sol]L (Ref. 6). The other 
article found no increase in the risk of death from bladder cancer from 
arsenic in drinking water at concentrations between 0.003 mg[sol]L and 
0.06 mg[sol]L (Ref. 7). We based our benefit estimates on reductions of 
bladder cancer and lung cancer. EPA's estimated health benefit from 
reductions in bladder cancer were approximately 30 percent of the total 
health benefits from reductions in both bladder and lung cancer. 
Therefore, we have reduced the lower bound of our estimated range of 
benefits to reflect the possibility that none of the options under 
consideration reduce the risk of bladder cancer.
4. Option One--Re-establish a Quality Standard for Arsenic in Bottled 
Water That Maintains the Current Allowable Level of 0.05 Mg/L
    We used this option as the baseline in the analysis of the proposed 
rule. We did not receive any comments on the use of this option as the 
baseline. We did receive one comment that noted that an allowable level 
of 0.05 mg[sol]L might not lead to the same health benefits as an 
allowable level of 0.01 mg[sol]L. This observation is consistent with 
the analysis of the proposed rule in which we attributed health 
benefits to moving from the baseline to an allowable level of 0.01 
mg[sol]L.
5. Option Two--Take No Action
    Benefits of option two.
    (Comment 3) One comment from a trade group that stated it 
represented 270 bottled water establishments argued that we may have 
overestimated the benefits of taking no action and allowing EPA's 
regulations governing arsenic in drinking water to become applicable to 
bottled water. The trade group that submitted this comment stated that 
it has required its members to meet a maximum arsenic level of 0.010 
mg[sol]L (10 ppb) since 2002 as a condition of membership. The comment 
also noted that California's Department of Health Services established 
a standard of quality specifying an allowable level of 0.01 mg[sol]L 
(10 ppb) for arsenic in bottled water in 2000. This comment argued that 
most medium and

[[Page 33697]]

large bottled water establishments of ``natural water'' belong to this 
trade group or sell water in California. Finally, the comment noted 
that approximately 25 percent of the bottled water sold in the United 
States is ``purified water'' and that most purified water is produced 
using the reverse osmosis method, which removes a substantial amount of 
any existing arsenic from the final product. The comment concluded that 
the vast majority of bottled water sold in the United States already 
meets an allowable level of 0.010 mg[sol]L and that any potential 
health benefits from revising the allowable level of arsenic from 0.05 
mg[sol]L to 0.01 mg[sol]L may have already been realized.
    (Response) In the analysis of the proposed rule, we based our 
benefit estimates on EPA's analysis of its drinking water regulations. 
EPA's analysis found that 5.3 percent of the ground water sources used 
by community water systems failed to meet a maximum arsenic level of 
0.010 mg[sol]L. We used the same percentage as the percentage of 
bottled water establishments that would fail to meet that level of 
arsenic. Thus, our benefit estimate accounted for the fact that the 
vast majority of bottled water establishments use water that already 
meets a maximum arsenic level of 0.010 mg[sol]L. The 270 establishments 
that this comment stated belonged to the trade group in question 
represent 73 percent of the 370 establishments that we identified in 
the analysis of the proposed rule. Thus, the information provided by 
the comment is consistent with the 5.3 percent estimate.
    Abatement.
    (Comment 4) One comment argued that some bottled water 
establishments might not be able to choose some of the 13 abatement 
methods that EPA discussed in their analysis. The comment noted that we 
used the average cost of these abatement methods in our analysis. 
According to this comment, establishments that bottle natural water 
containing naturally occurring arsenic may face abatement costs 
substantially higher than the average of the 13 methods discussed in 
the EPA report because of the commercial and financial restraints on 
their ability to selectively remove arsenic while maintaining the 
standard of identity for natural water. The comment also noted that 
abatement costs would depend on the initial level of arsenic found in 
the water (e.g., reductions from 0.03 mg[sol]L to 0.01 mg[sol]L is more 
expensive than reductions from 0.02 mg[sol]L to 0.01 mg[sol]L).
    (Response) We acknowledged in the analysis of the proposed rule 
that some bottled water establishments might be unable to use some of 
the 13 potential abatement methods EPA discussed in their analysis. Our 
rationale for using the average cost of those methods was that some 
establishments might be able to use the less expensive methods while 
other establishments might need to use the more expensive methods. 
Using average cost is appropriate in this context because we are 
estimating total costs rather than the costs that any particular 
facility might face. The comment did not provide sufficient information 
to revise this approach. EPA's cost estimates, on which we based our 
cost estimates, accounted for the fact that abatement costs depend on 
the initial level of arsenic in the water.
    Testing.
    (Comment 5) One comment argued that we overstated the potential 
benefit from reduced testing costs under this option and suggested that 
this option would probably not reduce testing costs at all. This 
comment noted that we estimated that adopting this option would 
eliminate between 163 and 745 tests per year. The comment said that 
such a reduction is highly unlikely because bottled water 
establishments that sell bottled water in more than one State might 
need to apply for waivers for each State in which they sell their 
product but may be unable or unwilling to pay for multiple waivers. The 
comment also noted that some States regulate bottled water as a food 
product and require annual testing for contaminants including arsenic. 
The comment said that only two of the States that regulate bottled 
water as food offer testing waivers for bottled water. In addition, the 
authors of the comment noted that they were unaware of any State 
granting a bottled water establishment a 9-year waiver for any 
contaminant. The comment claimed that adopting EPA's testing schedule 
for arsenic could result in additional tests because EPA's testing 
schedule would not coincide with States' testing schedules. Finally, 
the comment noted that the delay in testing requirements that we 
discussed in the analysis of the proposed rule would probably not 
affect bottled water establishments that operate in States that 
regulate bottled water as a food.
    (Response) In the analysis of the proposed rule, we assumed that 
between 0 and 90 percent of bottled water establishments might operate 
under a waiver in any given year. The low end of this range is 
consistent with the comment's assertion that few bottled water 
establishments would be able or willing to obtain waivers. The comment 
provided some reasons why the upper bound of 90 percent may be 
unrealistically high, but it did not provide an alternative upper bound 
estimate. The comment also did not provide sufficient information to 
estimate any additional testing that the comment claimed could be 
required under this option because of discrepancies between EPA's 
testing schedule and States' testing schedules, or the additional cost 
of tracking EPA's testing schedule if it differs from States' testing 
schedules, or the proper adjustment for the start of our testing 
requirements to account for the fact that some establishments must test 
annually because of State regulations. In the analysis of the proposed 
rule, we estimated that the change in testing costs generated by this 
option would round to $0 million per year. Attempting to further refine 
this estimate to account for these factors would have little effect on 
the overall results.
    (Comment 6) One comment argued that we failed to include some of 
the costs associated with testing requirements under this option. This 
comment noted that we previously allowed EPA regulations on maximum 
levels for nine other contaminants in drinking water to become 
effective for bottled water. The nine contaminants were antimony, 
beryllium, cyanide, nickel, thallium, diquat, endothall, glyphosate, 
and 2,3,7,8-TCDD (dioxin). The comment argued that implementing EPA 
testing requirements for these contaminants created confusion and 
inconsistencies because EPA designed their testing requirements for 
municipal water systems rather than for bottled water establishments. 
The comment suggested that this experience showed that implementing EPA 
testing requirements for arsenic would also create confusion about 
testing requirements for bottled water establishments.
    (Response) We are not addressing previous actions regarding the 
nine contaminants in this analysis. However, experiences generated by 
past actions may be relevant to this analysis. In this case, the 
comment claims that past experience suggests that adapting EPA's 
testing requirements for bottled water establishments could create some 
initial confusion. However, the comment did not provide sufficient 
information to allow us to quantify this cost. Therefore, we have added 
this cost as an unquantified cost.
    Administrative costs.
    We did not receive any significant comments on this section.

[[Page 33698]]

    Public notification.
    (Comment 7) One comment noted that we said that under this option 
(i.e., if we take no action and EPA's regulations are applied to 
bottled water establishments according to section 410 of the act) EPA's 
requirement that community water systems prepare and distribute public 
notifications of water analyses might apply to bottled water 
establishments, but we were unsure how EPA would apply or adapt these 
public notification requirements to bottled water establishments. The 
comment argued that if we take no action, then EPA's public 
notification requirements for community water systems would not become 
applicable to bottled water establishments and that the only change in 
our current regulations would be that EPA's MCL for arsenic and testing 
requirements would replace the existing maximum arsenic level and 
testing requirements. In addition, the comment noted that bottled water 
would remain under our jurisdiction.
    (Response) If we take no action, then EPA's NPDWR for arsenic in 
public drinking becomes applicable to bottled water. In addition to 
MCLs and monitoring requirements, EPA's NPDWRs (40 CFR part 141) 
contain other requirements such as analytical requirements (e.g., use 
of certified labs), reporting (e.g., test results submitted to the 
states), public notification (e.g., consumer confidence reports), and 
recordkeeping (chemical test results to be kept for at least 10 years). 
As such, EPA's public notification requirements would be applicable to 
bottled water. However, we agree with the comment that bottled water 
would remain under our jurisdiction and that we would be responsible 
for enforcing EPA's public notification requirements for bottled water 
establishments.
    Total costs and benefits of option two.
    Based on the analysis of the proposed rule and the preceding 
discussion, we estimate that taking no action and allowing EPA's NPDWR 
for arsenic to become applicable to bottled water would generate 
quantified benefits of $6 to $36 million per year (revised from $9 to 
$36 million per year in the analysis of the proposed rule), quantified 
costs of $11 to $15 million in the first year and $7 to $11 million in 
every year after the first year, plus any costs associated with public 
notification requirements, any costs associated with potential 
confusion associated with adapting EPA's testing requirements and any 
loss of consumer utility associated with product identity changes. This 
option could also cause some firms that produce bottled spring water to 
lose profits and firms producing competing products to increase 
profits.
6. Option Three--Establish a Quality Standard for Arsenic in Bottled 
Water That Adopts EPA's MCL for Arsenic in Drinking Water of 0.010 Mg/L
    (Comment 8) One comment noted that one advantage of this option is 
that the vast majority of bottled water establishments would not need 
to change their current testing procedures and States could easily 
harmonize their regulations with FDA regulations.
    (Response) This option would maintain current testing requirements 
and would therefore probably not disrupt existing testing schedules or 
otherwise create confusion about monitoring requirements. We did not 
attribute these costs to this option in the analysis of the proposed 
rule.
7. Option Four--Establish a Quality Standard for Arsenic in Bottled 
Water That Sets the Allowable Level of Arsenic at 0.02 Mg/L
    We did not receive any significant comments on this section.
8. Option Five--Establish a Quality Standard for Arsenic in Bottled 
Water but That Sets the Allowable Level of Arsenic at 0.005 Mg/L
    Benefits.
    We discussed the only comment that we received on the benefits of 
this option (that some bottled water establishments may need to purify 
their water and change the identification of their products from 
``spring water'' to ``purified water'' to meet this requirement) in the 
preceding section entitled General Comments because that comment was 
relevant to all of the options.
    Cost.
    (Comment 9) One comment noted that this option would affect more 
establishments than would Option 2 because this option involves a lower 
allowable level for arsenic. The comment suggested that this would 
generate a further increase in costs that is unknown but could be 
substantial.
    (Response) We estimated the costs of this option by adjusting our 
cost estimate for Option 2 upward by 232 percent based on the change in 
EPA's estimate of overall abatement costs associated with MCLs of 0.005 
mg[sol]L and 0.01 mg[sol]L. EPA's cost estimate accounted for the fact 
that a MCL of 0.005 mg[sol]L would affect more community water systems 
than would a MCL of 0.01 mg[sol]L. Thus, our estimate already 
indirectly accounted for an increase in the number of affected 
establishments under this option.
    Summary of benefits and costs for regulatory options.
    We present a summary of our revised cost and benefit estimates in 
table 1 of this document. Option 3 (adopting EPA's allowable arsenic 
level) appears likely to generate higher net benefits than either 
maintaining the current allowable level of arsenic in bottled water of 
0.05 mg[sol]L or taking no action and allowing EPA's NPDWR for arsenic 
to become applicable to bottled water. The estimated net benefits of 
adopting an allowable level of 0.010 mg[sol]L overlap significantly 
with the estimated net benefits of adopting an allowable level of 0.005 
mg[sol]L. The lower end of the range of potential net benefits is 
substantially higher for 0.010 mg[sol]L, but the higher end of the 
range is substantially higher for 0.005 mg[sol]L.

          Table 1.--Summary of Costs and Benefits ($ millions)
------------------------------------------------------------------------
   Option            Cost              Benefit           Net Benefit
------------------------------------------------------------------------
Option 1-     Baseline           Baseline            Baseline
 Maintain
 0.05
 mg[sol]L
------------------------------------------------------------------------

[[Page 33699]]

 
Option 2-     $11 to $15 in      $6 to $36 plus      -$9 to $25 plus
 Take no       first year, $7     unquantified        unquantified
 action        to $11 every       health benefits     benefits minus
               year after first                       unquantified costs
               year, plus                             in first year, -$5
               public                                 to $29 plus
               notification                           unquantified
               costs, any costs                       benefits minus
               associated with                        unquantified costs
               potential                              in subsequent
               confusion                              years
               associated with
               adapting EPA
               testing
               requirements,
               and any loss of
               consumer utility
               associated with
               product identity
               changes
------------------------------------------------------------------------
Option 3-     $7 to $11, plus    $6 to $36 plus      -$5 to $29 plus
 Adopt 0.010   any loss of        unquantified        unquantified
 mg[sol]L      consumer utility   health benefits     benefits minus
               associated with                        unquantified costs
               product identity
               changes
------------------------------------------------------------------------
Option 4-     $3 to $4, plus     $3 to $14 plus      -$1 to $10 plus
 Adopt 0.02    any loss of        unquantified        unquantified
 mg[sol]L      consumer utility   benefits            benefits minus
               associated with                        unquantified costs
               product identity
               changes
------------------------------------------------------------------------
Option 5-     $17 to $26, plus   $9 to $64 plus      -$17 to $47 plus
 Adopt 0.005   any loss of        unquantified        unquantified
 mg[sol]L      consumer utility   benefits            benefits minus
               associated with                        unquantified costs
               product identity
               changes
------------------------------------------------------------------------

B. Small Entity Analysis
    We have examined the economic implications of this final rule as 
required by the Regulatory Flexibility Act (5 U.S.C. 601-612). If a 
rule has a significant economic impact on a substantial number of small 
entities, the Regulatory Flexibility Act requires us to analyze 
regulatory options that would lessen the economic effect of the rule on 
small entities. We find that this rule would have a significant 
economic impact on a substantial number of small entities.
    In the analysis of the proposed rule, we discussed allowing small 
firms to produce bottled water containing a higher level of arsenic 
than larger firms as one possible approach to reducing the burden on 
small firms.
    (Comment 10) One comment argued that such an approach would provide 
less protection to consumers and would be difficult to communicate to 
consumers. The comment suggested that we instead consider delaying the 
effective date of the rule for small businesses.
    (Response) We acknowledged the impact on benefits that would occur 
if we allowed small firms to produce bottled water containing a higher 
level of arsenic than larger firms in the analysis of the proposed 
rule. We did not discuss the difficulty of communicating the existence 
of different maximum arsenic levels to consumers. However, this cost is 
not an additional cost but an alternative to the health costs that we 
discussed in the Small Entity Analysis section of the proposed rule. If 
we successfully communicated the existence of different arsenic levels 
to consumers, then consumers would presumably choose bottled water with 
lower levels of arsenic, and we would not expect to see a decline in 
health benefits. We do not have sufficient information to evaluate the 
cost or effectiveness of educating consumers on different arsenic 
levels as an alternative or partial alternative to the health costs of 
allowing small firms to produce bottled water containing more arsenic 
than bottled water produced by larger firms. Delaying the effective 
date for small firms would delay the onset of abatement costs but would 
not otherwise reduce those costs. Delaying costs would reduce the 
present value of those costs due to discounting. However, delaying the 
effective date would also delay the onset of the corresponding health 
benefits.
    (Comment 11) One comment argued that some bottled water 
establishments may need to purify their water using reverse osmosis or 
other methods in order to meet a maximum arsenic level of 0.010 
mg[sol]L. This comment suggested that some of these methods would 
require those establishments to change the identification of their 
products from ``spring water'' to ``purified water.'' The comment noted 
that this might change how consumers value the water and could reduce 
sales for the firms producing that water. The comment noted that it was 
unable to estimate this cost. We discussed this comment in the 
preceding impact analysis of this document. However, this comment is 
also relevant to this section because it noted that any loss of profit 
was more likely to affect smaller firms than larger firms because 
smaller bottlers have more limited treatment options and distribution 
areas.
    (Response) The comment is correct that any changes in product 
identity that might take place if bottled water establishments found it 
necessary to adopt certain treatment methods might lead to changes in 
how consumers value the water and could reduce sales and profits for 
some small firms. The comment did not provide sufficient information to 
estimate this potential impact on small firms.

VI. Paperwork Reduction Act

    FDA concludes that this final rule contains no collections of 
information. Therefore, clearance by OMB under the Paperwork Reduction 
Act of 1995 is not required.

VII. Federalism

    FDA has analyzed this final rule in accordance with the principles 
set forth in Executive Order 13132. FDA has determined that the rule 
has a pre-emptive effect on State law. Section 4(a) of the Executive 
Order requires agencies to
    ``construe * * * a Federal Statute to preempt State law only 
where the statute contains an express preemption provision, or there 
is some other clear evidence that the Congress intended preemption 
of State law, or where the exercise of State authority conflicts 
with the exercise of Federal authority under the Federal statute.''
Section 403A(a)(1) provides that
    ``no State or political subdivision of a State may directly or 
indirectly establish under any authority or continue in effect as to 
any food in interstate commerce-(1) any requirement for a food which 
is the subject of a standard of identity established under section 
401 that is not identical to such

[[Page 33700]]

standard of identity or that is not identical to the requirement of 
section 403(g) * * *.''
FDA has interpreted this provision to apply to standards of quality (21 
CFR 100.1(c)(4)). Although this rule has pre-emptive effect in that it 
would preclude States from issuing requirements for arsenic levels in 
bottled water that are not identical to the allowable level for arsenic 
as set forth in this rule, this pre-emptive effect is consistent with 
what Congress set forth in section 403A of the act.
    Section 4(c) of the Executive Order further requires that ``any 
regulatory preemption of State law shall be restricted to the minimum 
level necessary'' to achieve the regulatory objective. Under section 
410 of the act, not later than 180 days before the effective date of an 
NPDWR issued by EPA for a contaminant under section 1412 of the SDWA 
(42 U.S.C. 300g-1), FDA is required to issue a standard of quality 
regulation for that contaminant in bottled water or make a finding that 
such a regulation is not necessary to protect the public health because 
the contaminant is contained in water in public water systems but not 
in water used for bottled water. Further, section 410(b)(3) of the act 
requires a quality standard for a contaminant in bottled water to be no 
less stringent than EPA's MCL and no less protective of the public 
health than EPA's treatment techniques required for the same 
contaminant. FDA has determined that the MCL for arsenic that EPA 
established for public drinking water is appropriate as a standard of 
quality for bottled water, and is issuing this final regulation 
consistent with section 410 of the act.
    Further, section 4(e) of the Executive order provides that ``when 
an agency proposes to act through adjudication or rulemaking to preempt 
State law, the agency shall provide all affected State and local 
officials notice and an opportunity for appropriate participation in 
the proceedings.'' Given the statutory framework of section 410 of the 
act for bottled water, EPA's issuance of an MCL for arsenic in public 
drinking water provided notice of possible FDA action for a standard of 
quality for arsenic in bottled water. FDA did not receive any 
correspondence from State and local officials regarding an arsenic 
standard for bottled water subsequent to EPA's NPDWR on the MCL for 
arsenic or in response to FDA's proposed rule (69 FR 70082, December 2, 
2004) to adopt EPA's MCL for arsenic as an allowable level in the 
quality standard for bottled water. Moreover, FDA is not aware of any 
States that have requirements for arsenic in bottled water that would 
be affected by FDA's decision to establish a bottled water quality 
standard for arsenic that is consistent with EPA's standard for public 
drinking water. For the reasons set forth previously in this document, 
the agency believes that it has complied with all of the applicable 
requirements under the Executive order.
    In conclusion, FDA has determined that the pre-emptive effects of 
the final rule are consistent with Executive Order 13132.

VIII. References

    The following references have been placed on display in the 
Division of Dockets Management (HFA-305), Food and Drug Administration, 
5630 Fishers Lane, rm. 1061, Rockville, MD 20852, and may be seen by 
interested persons between 9 a.m. and 4 p.m., Monday through Friday.
    1. Arsenic in Drinking Water: 2001 Update, National Academy 
Press, Washington, DC, 2001, available on the Internet at http://www.nap.edu/books/0309076293/html/.
    2. Ibid., p. 13.
    3. Ibid., pp. 6-7, 11.
    4. Ibid., pp. 6, 11.
    5. Ibid., p. 11.
    6. Steinmaus, C., Yuan, Y., Bates, M.N., et al., 2003, ``Case-
Control Study of Bladder Cancer and Drinking Water Arsenic in the 
Western United States,'' American Journal of Epidemiology, 
158(12):1193-1201.
    7. Lamm, S., Engel, A., Kruse, M., et al., 2004, ``Arsenic in 
Drinking Water and Bladder Cancer Mortality in the United States: An 
Analysis Based on 133 U.S. Counties and 30 Years of Observation,'' 
Journal of Occupational and Environmental Medicine, 46(3):298-306.

List of Subjects in 21 CFR Part 165

    Beverages, Bottled water, Food grades and standards, Incorporation 
by reference.

0
Therefore, under the Federal Food, Drug, and Cosmetic Act and under 
authority delegated to the Commissioner of Food and Drugs, 21 CFR part 
165 is amended as follows:

PART 165--BEVERAGES

0
1. The authority citation for 21 CFR part 165 continues to read as 
follows:

    Authority: 21 U.S.C. 321, 341, 343, 343-1, 348, 349, 371, 379e.

0
2. Section 165.110 is amended by removing the entry for ``Arsenic'' in 
the table in paragraph (b)(4)(i)(A), by revising paragraph 
(b)(4)(iii)(A) and the introductory text of paragraph (b)(4)(iii)(E), 
and by adding paragraph (b)(4)(iii)(E)(14) as follows:


Sec.  165.110  Bottled water.

* * * * *
    (b) * * *
    (4) * * *
    (iii) * * *
    (A) The allowable levels for inorganic substances are as follows:

----------------------------------------------------------------------------------------------------------------
                     Contaminant                        Concentration in milligrams per liter (or as specified)
----------------------------------------------------------------------------------------------------------------
Arsenic..............................................                                                     0.010
Antimony.............................................                                                      .006
Barium...............................................                                                         2
Beryllium............................................                                                     0.004
Cadmium..............................................                                                     0.005
Chromium.............................................                                                       0.1
Copper...............................................                                                       1.0
Cyanide..............................................                                                       0.2
Lead.................................................                                                     0.005
Mercury..............................................                                                     0.002
Nickel...............................................                                                       0.1
Nitrate..............................................                                          10 (as nitrogen)
Nitrite..............................................                                           1 (as nitrogen)
Total Nitrate and Nitrite............................                                          10 (as nitrogen)
Selenium.............................................                                                      0.05
Thallium.............................................                                                     0.002
----------------------------------------------------------------------------------------------------------------


[[Page 33701]]

* * * * *
    (E) Analyses to determine compliance with the requirements of 
paragraph (b)(4)(iii)(A) of this section shall be conducted in 
accordance with an applicable method and applicable revisions to the 
methods listed in paragraphs (b)(4)(iii)(E)(1) through 
(b)(4)(iii)(E)(14) of this section and described, unless otherwise 
noted, in ``Methods for Chemical Analysis of Water and Wastes,'' U.S. 
EPA Environmental Monitoring and Support Laboratory (EMSL), Cincinnati, 
OH 45258 (EPA-600/4-79-020), March 1983, which is incorporated by 
reference in accordance with 5 U.S.C. 552(a) and 1 CFR part 51. Copies 
of this publication are available from the National Technical 
Information Service (NTIS), U.S. Department of Commerce, 5825 Port 
Royal Rd., Springfield, VA 22161, or may be examined at the Center for 
Food Safety and Applied Nutrition's Library, Food and Drug 
Administration, 5100 Paint Branch Pkwy., College Park, MD 20740, or at 
the National Archives and Records Administration (NARA). For 
information on the availability of this material at NARA, call 202-741-
6030, or go to: http://www.archives.gov/federal_register/code_of_federal_regulations/ibr_locations.html.
* * * * *
    (14) Arsenic shall be measured using the following methods:
    (i) Method 200.8--``Determination of Trace Elements in Waters and 
Wastes by Inductively Coupled Plasma-Mass Spectrometry,'' Revision 5.4, 
which is incorporated by reference in accordance with 5 U.S.C. 552(a) 
and 1 CFR part 51. Method 200.8 is contained in the manual entitled 
``Methods for the Determination of Metals in Environmental Samples--
Supplement 1,'' EPA/600/R-94/111, May 1994. Copies of this publication 
are available from the National Technical Information Service (NTIS), 
PB95-125472, U.S. Department of Commerce, 5825 Port Royal Rd., 
Springfield, VA 22161, or may be examined at the Center for Food Safety 
and Applied Nutrition's Library, Food and Drug Administration, 5100 
Paint Branch Pkwy., College Park, MD 20740, or at the National Archives 
and Records Administration (NARA). For information on the availability 
of this material at NARA, call 202-741-6030, or go to: http://www.archives.gov/federal_register/code_of_federal_regulations/ibr_locations.html.
    (ii) Method 200.9--``Determination of Trace Elements by Stabilized 
Temperature Graphite Furnace Atomic Absorption,'' Revision 2.2, which 
is incorporated by reference in accordance with 5 U.S.C. 552(a) and 1 
CFR part 51. Method 200.9 is contained in the manual entitled ``Methods 
for the Determination of Metals in Environmental Samples--Supplement 
1,'' EPA/600/R-94/111, May 1994. The availability of this incorporation 
by reference is given in paragraph (b)(4)(iii)(E)(14)(i) of this 
section.
* * * * *

    Dated: May 20, 2005.
Jeffrey Shuren,
Assistant Commissioner for Policy.
[FR Doc. 05-11406 Filed 6-8-05; 8:45 am]
BILLING CODE 4160-01-S