Motor Vehicle Manufacturers Ass'n of the United States v. New York State Department of Environmental Conservation , 17 F.3d 521 ( 1994 )

CARDAMONE, Circuit Judge:

The plaintiffs associations of American and foreign automobile manufacturers contest on this appeal the legality of defendant New York State’s newly adopted rules regulating automobile tailpipe, emissions. The most controversial issues to be resolved are whether New York State, like California, will in the coming years have a fixed percentage of zero emission automobiles, most likely, electric cars, and whether New York State, unlike California, can adopt auto emission requirements without adopting clean fuel requirements.

The invention and proliferation of the automobile has been a mixed blessing: its advantages are obvious and need no chronicling; its disadvantages, most notably as a source of air pollution that threatens human health and well-being, have become more and more apparent. Automobiles are the primary agents of ground level ozone and carbon monoxide. Ground level ozone, a major component of the now familiar phenomenon of urban smog, inhibits the human immune system and damages otherwise healthy lung tissue. The State of New York estimates that ten million New Yorkers live in ozone nonattainment areas. Carbon monoxide, a killer poison that interferes with the transfer of oxygen to the blood stream, adversely affects the functions of the heart and brain.

In response to the serious public health problems caused by ozone and carbon monoxide and the enormous task of cleaning up the air we breathe, the federal government enacted the Clean Air Act, and New York State under that Act adopted California regulations that mandate low emission vehicles. New York’s regulations fall heaviest on members of plaintiffs association, General Motors, Ford and Chrysler, and numerous international auto manufacturers, such as Honda, Nissan, Toyota, Volkswagen and Volvo. These manufacturers challenge on this appeal New York’s adoption of California’s standards.

Plaintiffs, Motor Vehicle Manufacturers Association of the United States, Inc. and Association of International Automobile Manufacturers, Inc. (collectively manufacturers) originally brought this action in the United States District Court for the Northern District of New York (McAvoy, C.J.) on July 9, 1992 against the New York State Department of Environmental Conservation (DEC) and its Commissioner, Thomas C. Jorling. They sought declaratory and injunctive relief against the Commissioner’s implementation and enforcement of regulations pertaining to automotive tailpipe emission standards adopted by the DEC in 6 NYCRR Part 218 (1992) (Part 218 Regulations).

After the commencement of this action, the Environmental Defense Fund, Inc. and New York State Electric & Gas Corp. moved to intervene as defendants on all counts pursuant to Fed.R.Civ.P. 24; the American Petroleum Institute moved to intervene as a plaintiff on one count, and as a defendant on two counts of the complaint. The Defense Fund’s and State Electric & Gas’ motions, were granted, as was the Petroleum Institute’s insofar as it sought intervention as a defendant. On appeal, helpful amicus briefs have been filed on behalf of the United States; the Attorneys General of New Jersey and Massachusetts; the City of New York; Congressman John D. Dingell; the American Lung Association, the Natural Resources Defense Council and the League of Women Voters of New York; and the Long Island Lighting Company.

BACKGROUND

Developments Leading up to the Clean Air Act

The Clean Air Act (Act), 42 U.S.C. §§ 7401-7671q (1988 & Supp. Ill 1991), is one of the most comprehensive pieces of legislation in our nation’s history. In order to better understand the issues it is helpful to trace briefly the development of that Act.

The original Clean Air Act, enacted by Congress in 1955, was aimed primarily at increasing federal research and assistance in air pollution prevention. It made no provision for federal motor vehicle emission standards. See Air Pollution Control-Research and Technical Assistance Act of 1955, Pub.L. No. 84-159, 69 Stat. 322. Because the several states had begun to adopt their own motor vehicle emission standards, the Senate Committee on Public Works, after noting that California was the leader in regulating automotive pollutant emissions, decided that national standards were to be preferred over having each state go its own way, “which could result in chaos insofar as manufacturers, dealers, and users are concerned.” S.Rep. No. 192, 89th Cong., 1st Sess. 5-6 (1965). As a result, the Committee proposed and Congress enacted in 1965 emission standards for new motor vehicle engines. See Motor Vehicle Air Pollution Control Act of 1965, Pub.L. No. 89-272, § 202(a), 79 Stat. 992.

A number of states, in addition to California, nonetheless continued to develop separate emission, programs. Congress thereupon promptly amended the Clean Air Act in 1967 to impose federal preemption over motor vehicle emission standards. See Air Quality Act of 1967, Pub.L. No. 90-148, § 208, 81 Stat. 485. Over the adamant objection of the auto industry, which sought a single national standard to avoid undue economic strain for manufacturers, California was excepted from preemption as the only state regulating auto emissions “prior to March 30, 1966”. Id. § 208(b). The reason for this lone exception was because the Senate Committee on Public Works was persuaded by California’s then Senator Murphy that his state’s “unique problems and pioneering efforts” warranted a waiver from preemption. S.Rep. No. 403, 90th Cong., 1st Sess. 33 (1967).

Comprehensive revisions made to the Act in 1970 established national ambient air quality standards (NAAQS) and required even more stringent uniform emission standards for new motor vehicles. See Clean Air Amendments of 1970, Pub.L. No. 91-604, §§ 4, 6, 84 Stat. 1676. In further amendments to the Act in 1977, § 209 (formerly § 208) was amended to require the U.S. Environmental Protection Agency (EPA) to consider California’s standards as a package, so that California could seek a waiver from preemption if its standards “in the aggregate” protected public health at least as well as federal standards. See Clean Air Act Amendments of 1977, Pub.L. No. 95-95, § 207, 91 Stat. 685.

Significantly, for the issues on this appeal, the Clean Air Act Amendments of 1977 added § 177, which permitted other states to “piggyback” onto California’s standards, if the state’s standards “are identical to the California standards for which a waiver has been granted for such model year.” Pub.L. No. 95-95, § 129(b), 91 Stat. 685, 750. In order for another state, here New York, to use California standards in a given model year, the House Committee on Interstate and Foreign Commerce made clear that California must adopt its standards two years in advance of such year, California must receive a waiver for its standards, and the adopting state must adopt California standards at least two years before the model year. See H.R.Rep. No. 294, 95th Cong., 1st Sess. 310 (1977), U.S.Code Cong. & Admin.News 1977, pp. 1077, 1389.

Clean Air Act Amendments of 1990

With this statutory evolution as background, we reach the Clean Air Act Amendments of 1990 and the current statute that lies at the heart of this litigation. The ami-cus brief of the United States describes the Clean Air Act as an extremely complex law and tells us that the “enormity of the 1990 amendments beggars description.” Congress, amicus informs us, took what was ■widely perceived as an “unapproachable piece of legislation” and tripled the Act’s length and “geometrically increased its complexity.” After reviewing the Clean Air Act and the voluminous record pertaining to the Act submitted on this appeal, we have no reason to doubt the validity of the government’s description of it.

Title I of the Act directs the Administrator of the EPA (Administrator) to develop NAAQS for pollutants the Administrator determines “cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare.” 42 U.S.C. § 7408(a)(1)(A); see id. § 7409(a). The several states are vested with the primary responsibility for attaining and maintaining the NAAQS through the development and operation of a state implementation plan (SIP). See Act § 110, 42 U.S.C. § 7410. Each state’s SIP, which is submitted to the EPA, must explain exactly how the state intends to reduce or maintain the concentration of pollution in the air to meet the NAAQS. The states have broad license to institute their own programs for the reduction of air pollution, principally through the regulation of stationary sources, such as industrial stacks and vents.

In Title II of the Act Congress endeavors to resolve the problems caused by moveable sources or vehicle emissions. The emission standards applicable to any given vehicle depend upon its weight and use classification, and its model year designation. See Act §§ 202, 207(c), 42 U.S.C. §§ 7521, 7541(c). Section 202 authorizes the Administrator to promulgate emission standards for motor vehicles sold in the United States. Motor vehicle emission standards primarily regulate emissions of carbon monoxide (CO), hydrocarbons or volatile organic compounds (VOCs) and nitrogen oxides (NOx). Hydrocarbons and nitrogen oxides are two classes of chemicals that combine in the presence of sunlight to form ozone, a major component of urban smog.

The cornerstone of Title II is Congress’ continued express preemption of state regulation of automobile emissions. See Act § 209(a), 42 U.S.C. § 7543(a) ¹. Pursuant to the Act, the great majority of states have chosen to rely on the federal emission standards as set forth in § 202 of the Act, 42 U.S.C. § 7521. Only California, because its unique Los Angeles smog problem caused it to begin regulating auto emissions “prior to March 30, 1966,” enjoys a statutory exemption allowing it to promulgate its own emission standards. See Act § 209(b), 42 U.S.C. § 7543(b)(1)².

Even so, as § 209(b) spells out, California may only adopt and enforce its own emission standards after applying to and obtaining the approval of the EPA for a waiver of preemption. The California Air Resources Board (CARB) submits an application upon determining that its proposed standards “will be, in the aggregate, at least as protective of public health and welfare as applicable Federal standards.” Id. The Administrator can deny the waiver request if she finds (1) that the protectiveness determination of California is arbitrary or capricious, (2) that California does not need separate state standards to meet “compelling and extraordinary conditions,” or (3) California’s “standards and accompanying enforcement procedures are not consistent with [§ 202(a) of the Act].” Id. Those standards are deemed not to be consistent with § 202(a) “if there is inadequate leadtime to permit the development of technology necessary to meet the proposed requirements, giving appropriate consideration to the cost of compliance within [the proposed] time frame.” U.S. Environmental Protection Agency, Waiver of Federal Preemption, California Low-Emission Vehicle Standards 57-58 (Jan. 7, 1992). Were California simply to change its standards, and such change were found to be within the scope of an existing waiver, California need not submit a new waiver application. See 46 Fed.Reg. 36742 (July 15, 1981) (setting forth standards for a change to be within the scope of a previous waiver).

The effect of the Clean Air Act is that motor vehicles manufactured for sale in the United States must be either “federal cars” — certified to meet federal vehicle emission standards as set by the EPA — or “California cars” — certified to meet that state’s standards. Both the EPA and California utilize the same two-step certification procedure. First, prototypes from each model year of a particular class of vehicles are tested prior to sale. See 42 U.S.C. § 7525; Cal.Code Regs. tit. 13 §§ 2050-2110. The test is designed to establish whether the vehicle will be in compliance with its designated standard for its “useful life”. The second and more rigorous test for emissions compliance occurs after the vehicles in a given model year have been sold to the public. See 42 U.S.C. § 7541(b); Cal.Code Regs, tit. 13 §§ 2136-2140. Samples are procured from the driving public and tested following the same procedures used during pre-sale certification. If the vehicles are not in compliance with the standards, both the EPA and CARB have the authority to order their recall at the manufacturer’s expense. See 42 U.S.C. § 7541(c); Cal.Code Regs. tit. 13 §§ 2133-2135.

The Act also restricts the states’ power to enact motor vehicle fuel requirements, though the EPA may approve state regulations if necessary to meet federal air quality standards. See Act § 211(c)(4)(C), 42 U.S.C. § 7545(c)(4)(A), (C). Again, California has a special exception from federal preemption allowing it to enact its own fuel requirements. Noteworthy is the fact that under the terms of the Act, EPA approval of California fuel regulations is not required. See. Act § 211(c)(4)(B), 42 U.S.C. § 7545(c)(4)(B).

Many states, including New York, are in danger of not meeting increasingly stringent federal air pollution limits. Congress has directed the Administrator to impose mandatory sanctions on states failing to timely submit approvable SIPs or SIP revisions. See 42 U.S.C. §§ 7410(m), 7509(a). The Administrator may either impose a 2:1 emission offset requirement to new and modified stationary sources of pollution, such as stacks and vents, or authorize the withholding of federal highway funds. See 42 U.S.C. § 7509(b). With the exception of certain safety or air quality related projects, the imposition of highway sanctions would prohibit federal approval or award of grants for highway construction or improvement projects. See id.

It was in an effort to assist those states struggling to meet federal pollution standards that Congress, as noted earlier, directed in 1977 that other states could promulgate regulations requiring vehicles sold in their state to be in compliance with California’s emission standards or to “piggyback” onto California’s preemption exemption. This opt-in authority, set forth in § 177 of the Act, 42 U.S.C. § 7507, is carefully circumscribed to avoid placing an undue burden on the automobile manufacturing industry. Specifically, (1) an opt-in state must adopt standards that are identical to California’s; (2) California must receive a waiver from the EPA for the standards; and (3) both California and the opt-in state must adopt the standards at least two years before the beginning of the automobile model year to which they apply. See Act § 177, 42 U.S.C. § 7507.

In the 1990 sweeping amendments to the Act two further restrictions were added to the last paragraph of § 177³. First, Congress added new language providing that § 177 shall not be construed as authorizing an opt-in state to limit the sale of California-certified vehicles. Second, it forbade opt-in states from taking’ any action that has the effect of creating a car different from those produced to meet either federal or California emission standards, a so-called “third vehicle.” See Clean Air Act Amendments of 1990, Pub.L. No. 101-549, § 232, 104 Stat. 2399, 2529.

This amendment to § 177 was not included in either the House or Senate versions of the 1990 amendments to the Act. It was added by the Conference Committee in October 1990. This came about as the Joint House-Senate Staff completed work on Title II of the Act, covering Mobile Sources. On October 10, 1990 the Joint Staff recommended to the legislative conferees that states not be allowed, when adopting California standards, to take any action that would result in a motor vehicle or engine different from a California or a federal vehicle, that is, no third vehicle could be required because of the burden such action would place on the auto industry. See H.R.Conf.Rep. No. 952, 101st Cong., 2d Sess. 337 (1990), U.S.Code Cong. & Admin.News 1990, pp. 3385, 3869. When Congress amended § 177 in 1990, it adopted these recommendations.

California’s Plan

With the evolution and scope of the Clean Air Act in mind, we turn now to examine the development of California’s emission limitation plan. In 1988 the California Legislature directed CARB to adopt the most cost-effective combination of motor vehicle controls, vehicle fuel restrictions, and in-use vehicle control requirements so as to achieve a 55 percent reduction in organic gas emissions by December 31, 2000. See Cal.Health & Safety Code §§ 43018(b), (e). CARB proceeded to establish low-emission vehicles (LEV) and clean fuels (CF) requirements, which became effective on September 28, 1991.

California’s LEV program applies to passenger cars and light-duty trucks (collectively light-duty vehicles) and medium-duty vehicles, and requires the creation of four classes of California light- and medium-duty vehicles, to be phased in over the next decade. The four categories are (1) Transitional Low-Emission Vehicles (TLEVs), (2) Low-Emission Vehicles (LEVs), (3) Ultra-Low-Emission Vehicles (ULEVs), and (4) Zero-Emission Vehicles (ZEVs). See Cal.Code Regs, tit. 13 § 1960.1(g)(2). For each category a set of more stringent emission standards for carbon monoxide, nitrogen oxides and formaldehyde applies. The average emissions from the mix of these categories of vehicles produced by a given manufacturer in a given year must meet an overall “fleet average” requirement. See id.

In addition, the fleet average for allowable exhaust emissions beginning in 1994 declines each year until 2003. See id. Automobile manufacturers, under CARB’s regulations, have the flexibility to decide how many vehicles of each type they manufacture and sell in order to meet the fleet average. Additional flexibility is provided through the establishment of a marketable credit system: manufacturers may earn credits if they sell more low-emission vehicles than needed to meet the fleet average. • Credits many be banked internally to offset future short-falls or sold to other manufacturers unable to meet their fleet average.

The only limitation on this fleet averaging approach is the sales quota established for ZEVs — vehicles with no tailpipe emissions of any pollutants throughout their lifetimes, which are presumably electric cars. By 1998, 2 percent of all vehicles certified for sale in California must be ZEVs, with the rate increasing to 5 percent in 2001 and to 10 percent in 2003. See id. The ZEV quota initially applies to all manufacturers who sell annually more than 35,000 light- and medium-duty vehicles in California. But, beginning in 2003 i':at threshold drops to manufacturers who sell only 3,000 or more vehicles in California. See id. The ZEV quota also has a marketable credit system that works just like the fleet averaging credit system. See id.

The CF component of California’s plan, see Cal.Code Regs. tit. 13 §§ 2300-2317, entails a two-step introduction of “reformulated gasoline,” which is referred to as Phase 1 and Phase 2 gasoline. These fuels will have a lower sulfur content than gas sold in the rest of the nation, such as the Venezuelan gasoline sold on the east coast, which has a high sulphur content. With limited exceptions, gasoline sold in California after January 1, 1992 had to be Phase 1. Compliance with the Phase 2 gas requirements — which are considerably more stringent than the Phase 1 requirements — commences on March 1,1996. Officials of CARB have frequently stated that “[t]he primary purpose of the specifications will be to reduce emissions from pre-1993 and off-road vehicles.”

Notwithstanding that fact, California will allow vehicle manufacturers to certify, both pre-sale and in-use, their new vehicles using a test fuel with properties of the Phase 2 reformulated gasoline. Manufacturers need not wait until the commercial introduction of these fuels; rather they can begin certifying vehicles on Phase 2 gas immediately. In fact, General Motors has already done so in an attempt to secure LEV credits for the upcoming years. Prior to this plan, California tested vehicles using Indolene, a special non-commercial certification fuel used by the EPA in the rest of the nation. The Chief Deputy Executive Officer of CARB has stated that “allowing vehicle manufacturers to certify gasoline-powered low-emission vehicles on Phase-2 reformulated gasoline certification fuel ... [is] an integral part of the [C]ARB’s new motor vehicle standards and test procedures.”

In October 1991 California submitted its plan to the EPA for approval as required by § 209(b). See 42 U.S.C. § 7643(b). While the ability to use Phase 2 gas for certification was not in the waiver application, CARB intends to ask the EPA for a waiver or assurance that the Phase 2 amendments are within the scope of the October 1991 application. The EPA approved California’s plan on January 7, 1993. Significantly, despite requests by automobile manufacturers, this approval was not conditioned on the commercial availability of Phase 2 gasoline.

New York’s Adoption of California’s Plan

According to DEC, New York currently has one of the most intractable air pollution problems in the nation due largely to automobile emissions. Mobile sources emit more than 90 percent of the carbon monoxide and nitrogen oxides that pollute its air. New York — like 39 other states and the District of Columbia — is not in compliance with the ozone NAAQS or carbon monoxide NAAQS established pursuant to the 1990 amendments to the Act. New York City suffers from serious carbon monoxide pollution and has been classified as a “moderate” nonat-tainment area. See 56 Fed.Reg. 56,694 (Nov. 6, 1991). In addition, eight counties in New York are not in compliance with the carbon monoxide NAAQS and 22 counties failed to meet the ozone requirements. See id. at 56,804. Of these 22, the New York City area has been designated a “severe” nonattainment area.

New York therefore must prepare a SIP adequate to meet the federal air quality standards for carbon monoxide by the end of 1995 and for ozone by late 2007. See 42 U.S.C. §§ 7512(a)(1), 7511(a)(2). The SIP must demonstrate to the satisfaction of the EPA that New York (1) will meet the health-based standards by the dates specified in the Act, and (2) will achieve a 15 percent reduction in volatile organic compounds from 1990 baseline emissions by 1996, and a 3 percent reduction on average each year thereafter until the federal standard is achieved. See id. §§ 7511a(b), (c). Failure to meet these obligations will trigger the sanctions discussed earlier. See id. § 7509.

In order to combat its serious persistent pollution problem, New York, pursuant to § 177 and through regulations promulgated by the DEC, adopted California’s LEV program on May 28,1992. Although these regulations were filed on April 28, 1992, they did not become effective until May 28,1992. For purposes of this opinion the term “adoption” refers to the latter date. New York’s regulation' applies at present only to light-duty vehicles and does not apply to medium-duty vehicles as in California. New York’s adoption took place more than six months before the EPA granted California a waiver. The DEC acted under its rulemaking authority, and the regulations contain new standards for tailpipe and evaporative emissions for model year 1995. The regulations were published at 6 NYCRR Part 218.

New York did not adopt California’s CF program, but the regulations do provide that motor vehicles sold in New York will be certified, both pre-sale and in-use, under the identical procedures used in California, which currently permit manufacturers to use the reformulated gasoline — Phase 1 and Phase 2 — as noted above. See id. § 218-5. Federal reformulated gasoline — while not as “clean” as California’s Phase 2 gas — will be marketed in New York beginning January 1, 1995 in all ozone nonattainment areas classified marginal or worse. See 57 Fed.Reg. 7926, 7926-27 (Mar. 5, 1992). New York says it elected not to adopt the California CF plan because it concluded the plan was not cost-effective.'

In addition to New York’s adoption of the California plan, a letter of understanding was signed by all the states of the Northeast Ozone Transport Commission — a group composed of 12 states along the eastern seaboard from Virginia to Maine and the District of Columbia — on October 29, 1991 pledging to adopt the LEV plan. Other than New York, so far only Massachusetts has promulgated regulations to this effect. The other states of the Commission, we are advised, are awaiting the outcome of this litigation.

Prior Proceedings

We pass now to the proceedings held in the district court. The complaint alleges in six counts that DEC’s adoption of the LEV standards violates § 177 of the Act: (1) DEC’S failure to adopt the CF component of California’s regulations violates the “identically” requirement and (2) violates the “undue burdens” and “third vehicle” prohibitions of § 177; (3) DEC’S adoption of California’s regulations before California had received a waiver from the EPA violates § 177 and (4) did not comply with the two year leadtime requirement; (5) DEC’S ZEV sales quota contravenes the prohibition on all plans that limits the sales of California certified vehicles and (6) contravenes the prohibition against a “third vehicle”.

In October and November 1992 all parties moved for summary judgment. The district court’s initial decision and order was entered on January 26, 1993 and reported as Motor Vehicle Manufacturers Ass’n of the United States v. New York State Department of Environmental Conservation, 810 F.Supp. 1331 (N.D.N.Y.1993) (MVMA I). Defendants’ motion for reargument persuaded the district court to modify in part its decision on counts two and four. See Motor Vehicle Mfrs. Ass’n v. New York State Dep’t of Envtl. Conservation, 831 F.Supp. 57 (N.D.N.Y.1993) (MVMA II).

As a result of the district court’s rulings, the present posture of the six counts alleged in the complaint is as follows: On count one, the district court granted summary judgment for DEC. Because New York could adopt California’s emission standards “for which a waiver has been granted” pursuant to § 177, because California did not need a waiver for its CF requirements, and because the CF requirements were not part of the plan California submitted to the EPA, the trial court concluded that New York did not violate the “identicality” requirement of § 177. See MVMA I, 810 F.Supp. at 1343.

On count two, the manufacturers had claimed that New York’s failure to adopt California’s CF requirements will have the effect of obliging them to modify their California LEVs for sales in other states because of the higher sulfur content in non-California fuels, and thereby violate § 177’s third vehicle prohibition. The district court initially granted summary judgment to plaintiffs, see id, at 1345, but upon reconsideration it found there were genuine questions of material fact as to this issue and therefore denied summary judgment to both sides and ordered a trial, see MVMA II, 831 F.Supp. at 61. That ruling is not before us on this appeal.

Summary judgment was granted to DEC on count three. Because the Act does not prevent California from adopting regulations — it merely requires their approval prior to enforcement — the court concluded “it would be illogical to read the Act as allowing California to adopt standards without a waiver, and not allow other states to do the same.” MVMA I, 810 F.Supp. at 1347.

On count four, the trial court initially held DEC’s May 28,1992 adoption did not provide manufacturers with, the required two year leadtime for their 1995 model year vehicles. See id. at 1348. Upon defendant’s motion for reconsideration, it modified this holding to include a finding that only those manufacturers who could demonstrate that they commenced production of model year 1995 vehicles prior to May 28, 1994 would be exempt from the plan for 1995. See MVMA II, 831 F.Supp. at 64. This decision is challenged on appeal by plaintiffs and defendants. Plaintiffs argue that “model year” is an industry-wide uniform period for all manufacturers, and because model year 1995 commences prior to May 28,1994, a delay of implementation of DEC’s regulations until model year 1996 must be granted to all manufacturers. In response, DEC maintains that each manufacturer should be required to demonstrate commencement of production on an engine-family-by-engine-family or model-by-model basis.

On count five, the trial court concluded that because the ZEV sales quota would require the sale of a specified number of ZEVs, it would have the effect of limiting the sales of other California-certified vehicles. See MVMA I, 810 F.Supp. at 1346. Summary judgment was granted to the manufacturers on this count, as well as on count six. According to the district court, because ZEVs in New York State would require distinct features — principally due to climatic and market differences — the ZEV quota would force manufacturers to design the prohibited-by-statute third vehicle. See id. at 1347.

Final judgment on counts one, three, four, five and six was entered on August 23, 1993 pursuant to Fed.R.Civ.P. 54(b). Defendants filed a notice of appeal on September 7, 1993 and plaintiffs cross-appealed on September 14, 1993. Plaintiffs thereafter successfully moved for the appeal to be expedited, at which time plaintiffs were redesignated as the appellants.

DISCUSSION

As the background just related makes plain, our primary task is to divine the meaning of § 177 and to determine whether that meaning is transgressed by New York’s Part 218 Regulations. .The familiar rules of statutory construction look first to “the language of the statute itself because, when looking at its language, a court should presume that the statute says what it means.” Aslanidis v. United States Lines, Inc., 7 F.3d 1067, 1072 (2d Cir.1993). “In the usual case, where the language of the. statute is clear, that is the end of the analysis.” Lewis v. Grinker, 965 F.2d 1206, 1215 (2d Cir.1992); see also Aslanidis, 7 F.3d at 1073 (“If the words of a statute are unambiguous, judicial inquiry should end, and the law interpreted according to the plain meaning of 'its words.”). Even when the meaning of language seems apparent, it must be remembered that we are attempting to ascertain Congress’ purpose; where ambiguity resides in a statute, legislative history and other tools of interpretation may be employed to determine legislative purpose more perfectly. See Aslanidis, 7 F.3d at 1073. Because some of the language in § 177 admits of several readings we will from time to time in the discussion that follows examine its legislative history.

As the report of the House Committee on Interstate and Foreign Commerce demonstrates, § 177 was inserted into the Act in 1977 so that states attempting' to combat their own pollution problems could adopt California’s more stringent emission controls. See H.R.Rep. No. 294, 95th Cong., 1st Sess. 309-10 (1977). An equally significant purpose evident in the statute and the legislative history is Congress’ desire not to burden manufacturers unduly with myriad state emission regulations. See id. at 309. We endeavor in the discussion that follows to harmonize these sometimes disparate goals.

I Identicality

The first issue addressed is whether the DEC’s failure to adopt California’s CF plan when it promulgated the LEV standards violates § 177’s requirement that State emission standards be “identical to the California standards for which a waiver has been granted.” 42 U.S.C. § 7507. Exactly what “California standards” New York must adopt to meet the identicality requirement is found in the plain language of the statute.

Section 177 refers to “standards relating to control of emissions ... for which a waiver has been granted.” Id. (emphasis added). In enacting § 209(b), which establishes California’s preemption exception, Congress uses the same words as it did when it allowed California to set its own “standards .... for the control of emissions,” provided the EPA approves a waiver application. Id. § 7543(b)(1). Hence, the most logical reading of § 177 is that New York may adopt only those standards that, pursuant to § 209(b), California included in its waiver application to the EPA

California was not required to obtain an EPA waiver for its CF plan because Congress treated the two initiatives to clean up the air separately — one for engine emission and one for fuel — as shown by the fact that it granted California a fuels exemption under a separate provision of the Act, namely § 211(c)(4)(B). Moreover, the Chief Deputy Officer of CARB stated that the CF plan was not included in the waiver application submitted by California in October 1991. California adopted its LEV standards over a year before it promulgated its final rule for Phase 2 reformulated gasoline, and the EPA’s approval of the LEV plan, as already observed, was not conditioned on the commercial availability of the gasoline. Only the regulations allowing manufacturers to certify LEVs on reformulated gasoline are within the scope of the § 209(a) preemption provision, and therefore require a § 209(b) waiver, though California has yet to receive such a formal waiver. But because New York has adopted these certification standards by reference, see 6 NYCRR § 218 — 5(b), manufacturers are not deprived of the ability to certify on reformulated gasoline the vehicles they sell in New York.

In sum, the plain language of § 177 not only provides that New York need not adopt California’s CF plan, it actually precludes New York’s adoption of such plan under this provision, as the plan was not part of the waiver application. This view is entirely consistent with the reasoning of Motor & Equip. Mfrs. Ass’n Inc. v. EPA, which held “[t]he plain meaning of [§ 209] indicates that Congress intended to make the waiver power coextensive with the preemption provision.” 627 F.2d 1095, 1107 (D.C.Cir.1979), cert. denied, 446 U.S. 952, 100 S.Ct. 2917, 64 L.Ed.2d 808 (1980). Since the preemption provision of § 209 covers only automobile emission standards, the waiver provision of § 209 must only apply to automobile emission standards. Consequently,' California could not have obtained a preemption waiver for its CF plan under § 209(b). Rather, California’s exception from preemption for this program could only come from § 211, which has no federal waiver review requirements analogous to those contained in § 209.

At first glance this construction might appear to be at odds with the Act’s overall emphasis on more stringent standards for improving the quality of our nation’s air. In fact, the manufacturers ask why, when promulgating Part 218, the DEC did not impose regulations on the fuel burned as well as the emission control hardware, since tailpipe exhaust is a product of both. New York answers this query by saying regulation of fuel was not cost effective. In any event, whether to impose such regulation is a matter of policy confided to New York. The Part 218 Regulations do not preclude New York from adopting the CF plan and seeking a CF preemption waiver from the EPA under § 211(c)(4)(C) of the Act, 42 U.S.C. § 7545(c)(4)(C), should it choose to do so later. Thus, Congress’ aim of improving air quality is not undermined by reading the CF plan out of the “identicality” requirement. Were we to hold otherwise, we would undermine Congress’ scheme of treating emissions and fuel standards separately, in effect ruling in a manner outside our authority and substituting our judgment for that of the state regulatory body.

Plaintiffs launch a two-pronged attack on this reading of the Act. First they insist the term “waiver” in § 177 should be read broadly, and while the EPA did not grant a preemption waiver for California’s CF plan, California nonetheless received an “automatic waiver” under § 211(c)(4)(B), 42 U.S.C. § 7545(c)(4)(B). This argument is flawed. The phrase “for which a waiver has been granted” in § 177 on its face precludes its application to an “automatic waiver” because “granted” denotes an affirmative act of bestowal. In contrast, § 211(c)(4)(B) is self-executing. It does not provide for any “grant” of a preemption “waiver” for California fuel standards. Instead it excludes such standards altogether from the scope of the preemption provision.

On a second front, the manufacturers also maintain that § 302(k) of the Act, 42 U.S.C. § 7602(k), sets forth a definition of “emission standard” that encompasses fuel regulations such as the CF plan.⁴ Even assuming the CF plan is within the scope of this definition, § 302(k) is inapplicable to the issues before us. The precise term “emission standard” is not found in the language of either section relevant to the instant action. More importantly, § 302(k)’s definition was enacted ten years after § 209(b) and was intended to apply only in the context of regulating emissions from stationary sources, not moveable sources like automobiles. See Motor & Equip. Mfrs. Ass’n Inc., 627 F.2d at 1112 n. 35.

Further, allowing the DEC to adopt only the emission standards plan does not defeat the two goals of § 177. Our reading of § 177 permits New York to adopt emission standards more stringent than those it would be left -with under current federal law, and the LEV plan standing alone does not force auto manufacturers to do something more than they already have to do, that is, develop a plan to meet the fleet average requirement. After all, the manufacturers will presumably design and build the same cars for California as will be sold in New York. Most importantly, the auto industry will be allowed to certify vehicles as meeting emissions requirements using the same fuels that'California uses for its certification. In this respect, whether New York’s failure to adopt the CF plan somehow violates the “third vehicle” prohibition is not a question we need to reach on this appeal. That question is embodied in count two of the complaint, and it is proceeding to trial.

II Timing

A. DEC’S Adoption Prior to EPA’s Grant of California’s Waiver

The next issue for analysis is whether DEC violated § 177 by adopting the Part 218 Regulations more than six months prior to EPA’s having granted California a waiver pursuant to § 209(b). Again, the plain language of § 177, coupled with common sense, suggests how this issue should be resolved.

As related, § 177 allows New York to “adopt and enforce” emission standards “identical to the California standards for which a waiver has been granted.” 42 U.S.C. § 7507. Plaintiffs insist this apparently plain language is not in fact so plain. All parties agree the EPA’s grant of a waiver to California pursuant to § 209(b) is a precondition to some type of action by the adopting state. But the manufacturers declare that New York was precluded from taking any action until the EPA actually granted California a waiver for the LEV plan — which it did on January 7, 1993 — and that New York should not have been permitted to adopt California’s LEV plan until that date.

The manufacturers wanted the DEC to withhold its action of adopting the regulations until January 7, 1993 because that date is when they believe the two year leadtime should have commenced running. They support their position by noting that this precondition was in the statute to insure that manufacturers had sufficient notice of future regulations affecting the auto industry. The DEC had of course adopted the Part 218 Regulations on May 28,1992. It responds to the manufacturer’s position by stating it was permitted to adopt the regulations prior to the EPA’s having granted California’s waiver, provided it did not enforce the regulations before January 7, 1993.

The issue is what do the above quoted words of § 177 mean, that is to say, what is the waiver a precondition to — DEC’S adoption of the LEV plan or DEC’S enforcement of the LEV plan, or both. The most sensible response, it appears to us, is that the waiver is a precondition to enforcement of the standard that has been adopted. In other words, it is sensible for DEC to adopt the standards prior to the EPA’s having granted a waiver, so long as the DEC makes no attempt to enforce the plan prior to the time when the waiver is actually obtained.

The result urged by the manufacturers ignores the fact that the need for adequate notice was addressed by Congress in the two year leadtime requirement discussed below. The manufacturers’ suggestion that more than two years is required because of possible uncertainty in the approval of California’s waiver is a bit disingenuous. Plaintiffs know that California’s waiver applications are almost always approved, in light of Congress’ decision “to permit California to blaze its own trail with a minimum of federal oversight.” Ford Motor Co. v. EPA, 606 F.2d 1293, 1297 (D.C.Cir.1979); see also Motor & Equip. Mfrs. Ass’n, 627 F.2d at 1121-22 (California’s emission standards “are presumed to satisfy the waiver requirements” and EPA may not disregard California’s determination absent “clear and compelling evidence” to the contrary).

By adopting, without enforcing, the challenged Part 218 Regulations before the EPA ruled on the waiver application, the DEC put plaintiffs on notice that New York intended to take the same path as had California. The two year leadtime of § 177 more than adequately ensures a sufficient period for manufacturers to prepare for an opt-in state’s adoption of California’s standards.

B. Two Year Leadtime

Given our holding that New York’s adoption of the Part 218 Regulations was valid even though it occurred before California was granted a waiver, a third issue is whether DEC’s adoption of California’s standards — applicable to model year 1995 vehicles — provides manufacturers the two year leadtime required by § 177. The critical date is May 28, 1992, the date when New York adopted the California standards. The district court ruled the Regulations would be unenforceable as against those manufacturers who commenced production of model year 1995 vehicles prior to May 28, 1994, because those manufacturers would have had less than two years from the 1992 adoption date. See MVMA II, 831 F.Supp. at 64. General Motors, for one, produced unrebut-ted evidence that it would commence production of all of its model year 1995 vehicles prior to May 28, 1994.

The trial court’s ruling regarding the two year leadtime requirement satisfies neither side. DEC contends the leadtime requirement turns on a specific model’s or engine family’s first production within the model year. Under this interpretation, individual manufacturers could be subject to differing certification rules for different models or engine families. The manufacturers, similarly dissatisfied with the district court’s decision, maintain that the manufacturer-by-manufacturer rule would create havoc in the automobile industry; they also reject DEC’S engine-family-by-engine-family rule as having the potential to create even greater chaos in the industry.

To determine the proper application of the two year leadtime provision, we look again to the language of § 177. It provides that states adopting California’s standards must do so “at least two years before commencement of [the] model year.” 42 U.S.C. § 7507. Congress specifically stated in § 177 that “model year” shall be “determined by regulations of the Administrator.” Id. But such regulations have never been promulgated. For the purpose of federal automobile regulations, the EPA defines a model year as a manufacturer’s annual production period. See 40 C.F.R. § 86.082-2 (1992). The EPA issued an Advisory Circular that an annual production period for any specific model within an engine family commences (1) when a vehicle or engine designated for a particular year is “first produced,” or (2) on January 2 of the calendar year prior to the model year, whichever date is later. Relying on this definition in its amicus brief, the EPA sides with the DEC, asserting that its definition means that “model year” should be applied on an engine-family-by-engine-family basis, not on a manufacturer-by-manufacturer basis. Moreover, the EPA believes its view is entitled to deference under Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843, 104 S.Ct. 2778, 2782, 81 L.Ed.2d 694 (1984). We are unwilling to accept the proffered definition. Section 177 charges the EPA with the single, narrow responsibility to issue “regulations” in order to define the commencement of a model year under § 177. The EPA Advisory Circular upon which the DEC relies is not a “regulation” for § 177 purposes and was not promulgated specifically to implement this provision.

Nor are we inclined blindly to follow the EPA’s suggestion that we defer to the interpretation of the term model year as proposed in their amicus brief. A newly announced view made by a rulemaking authority during litigation is entitled to some deference, but the circumstances of the announcement must be taken into account when considering its reasonableness and the extent of deference a court must accord it. See Director, Office of Workers’ Compensation Programs v. General Dynamics Corp., 982 F.2d 790, 795 (2d Cir.1992) (relying on Martin v. Occupational Safety & Health Review Comm'n, 499 U.S. 144, 157, 111 S.Ct. 1171, 1179, 113 L.Ed.2d 117 (1991)). While it is correct to say that the EPA’s foundation for its proposed definition of model year is not new (in other words, the federal automobile regulation and Advisory Circular are not new), EPA’s position that this definition should also apply to § 177 appears newly minted for this litigation. We consider this fact in determining whether the EPA’s position is reasonable.

Given the lack of EPA regulations for purposes of § 177, a practical approach to interpreting this requirement is warranted and reveals the unreasonableness of the DEC’s and EPA’s view. The crux of the LEV plan is the fleet averaging concept. Fleet averaging is designed to give manufacturers sufficient flexibility to develop varying emissions within their entire fleet to meet the overall goal. Breaking down a particular manufacturer’s fleet on an engine-family-by-engine-family or model-by-model basis would dis-serve the LEV plan itself by unduly complicating the fleet averaging plan. Moreover, the flexibility of the LEV plan allows manufacturers to buy and sell credits to meet their fleet average requirements. This feature would also be frustrated if all manufacturers were not playing by the same rules. Consequently, we hold that absent promulgated EPA regulations, “model year” is best read as applying on an industry-wide basis. Because model year 1995 commences prior to May 28,1994, as demonstrated by the uncon-troverted evidence submitted by General Motors, DEC should be enjoined from enforcing its LEV plan for model year 1995 as against all manufacturers.

Our holding accords with Congress’ express inclusion of the leadtime requirement in § 177 to protect the auto industry. This requirement must be construed for the plaintiffs’ benefit, in contrast to the deferential approach — accruing to defendant’s benefit— taken on the earlier issues addressed. The leadtime provision was specifically added to insure that manufacturers would have an ample amount of time to adjust to upcoming regulations. The district court’s manufacturer-by-manufacturer rule might well impose different emission standards on different manufacturers within the same model year depending upon the date each manufacturer starts production of its model year 1995 vehicles or engines.

We do not believe Congress contemplated a “split” model year approach, differentiating manufacturers based on when each commences production. No set of regulations has sought to implement a “split” model year. Nor has any court ever applied emission standards on a manufacturer-by-manufacturer basis. The only court to have previously touched on this subject refused to differentiate among manufacturers because of the “competitive as well as statutory considerations.” International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 640 (D.C.Cir.1973). Needless to say, imposing a manufacturer-by-manufacturer standard or, even worse, an engine-family-by-engine-family standard, would cause great confusion for the entire automobile industry, while being entirely impractical to enforce.

Ill The ZEV Sales Quota

A. Limitation on the Sale of California-Certified Vehicles

Next we discuss whether DEC’S ZEV sales quota violates § 177 because, as the manufacturers point out, it will limit sales of other classes of California-certified cars. Pursuant to § 177, New York cannot “prohibit or limit, directly or indirectly, the manufacture or sale of a [vehicle or engine] that is certified in California as meeting California standards.” 42 U.S.C. § 7507. The district court held that by requiring a certain percentage of California-certified vehicles sold in New York to be ZEVs, “DEC has effectively limited the sales of all other classes of California certified vehicles.” MVMA I, 810 F.Supp. at 1346.

The parties dispute, as an initial matter, whether the ZEV quota is a “standard[] relating to control of emissions” within the meaning of § 177. Since resolution of this dispute would not be dispositive of the ZEV mandate’s validity, it need not be addressed other than to say that regardless of whether or not the ZEV quota is a standard, it still cannot violate § 177. Leaving this squabble aside, it must be observed that the parties’ semantical arguments have muddied the waters of what appears to be a straightforward statutory provision. Analyzing the issue in simple terms will aid in reaching a sensible and reasoned decision.

The ZEV sales mandate was obviously included by California in the LEV plan to encourage the development, production, sale and use of ZEVs. The Act is concededly a “technology forcing” law. Equally obvious is the fact that ZEVs sold in New York will themselves be California-certified cars. Under the current ZEV plan, manufacturers may sell any number of California-certified vehicles other than ZEVs so long as they also sell the specified percentage of ZEVs. Hence, New York has not acted so as to limit (either directly or indirectly) the number of non-ZEV California-certified cars the manufacturers will sell. The more non-ZEV California-certified vehicles they sell, the more ZEVs they must sell. How the manufacturers will insure the sale of ZEVs is not a matter we must resolve; suffice it to say the sale of ZEVs will hinge on marketing and competitive factors, fields in which the manufacturers are experts.

Even though California and New York have not decided how they intend to penalize noneompliance, manufacturers would not be penalized per se for the sale of too many California-certified cars other than ZEVs. Rather, it would be their failure to sell the appropriate number of ZEVs (or purchase sufficient ZEV credits) that would lead to penalties. Thus, in our view the purpose of the sales limitation prohibition is to prohibit § 177 opt-in states from attempting to regulate against the sale of a particular type, not number, of California-certified cars. It must be remembered that the manufacturer has already expended resources specifically for the design and certification of emissions hardware. There is no allegation in this record that New York has overtly limited the sale of any type of California vehicle.

It would be naive not to recognize that since the market for cars is not unlimited, the ZEV quota may affect the sale of non-ZEV California-certified cars. But “affect” is not the key word of § 177, the catchword is “limit”. See 42 U.S.C. § 7507. The district court’s holding that the DEC regulation respecting a percentage sale of ZEVs limited the sale of other classes of California vehicles skews Congress’ purpose. Congress wanted the plans of opt-in states to be identical to those of California, as is evident from the identicality requirement. Ruling in effect that one portion of the plan adopted according to the specific instructions in § 177 somehow at the same time violates § 177, places New York or other potential opt-in states in a Catch-22 position. Like the third vehicle rule, the sales-limitation rule is designed to reinforce the identicality requirement. It would be incongruous for us to hold that the DEC wrongly mandated a ZEV sales percentage identical to California’s mandate.

Moreover, under the trial court’s view of this matter, the fleet averaging concept standing alone would have to be read as violating this same provision because it necessarily forces manufacturers to sell some unfixed number of lower emissions vehicles. The only difference between the fleet averaging and the ZEV quota notions is that with the ZEV quota the actual percentage of vehicles sold is fixed; while the number can fluctuate under the rest of the LEV plan, the fixed fleet average must also still be met.

In light of Congress’ recognition, when it amended the Act in 1990, of California’s LEV program, including the ZEV mandate, see, e.g., Act §§ 241(4), 243(f), 246(f)(4), 42 U.S.C. §§ 7581(4), 7583(f), 7586(f)(4), we are disinclined to pave the way for a nullification of New York’s Part 218 Regulations or to impair any other state’s ability to adopt California’s LEV plan. It would be inappropriate to construe the 1990 amendments in a manner that would effectively prohibit any state from opting into the California program since Congress so obviously planned for the several states to have that option. See Lems v. Grinker, 965 F.2d 1206, 1215 (2d Cir.1992). Because the district court may unintentionally have laid the groundwork for an abrogation of the entire LEV plan, its decision on this issue must be reversed.

B. The Third Vehicle Prohibition

The remaining issue is whether summary judgment was properly granted to the manufacturers on the rationale that DEC’S ZEV quota would inevitably cause production of a third vehicle in violation of § 177. Obviously, the disposition of this issue turns on the meaning of the third vehicle bar that was added to § 177 in 1990. Section 177 prohibits states from “tak[ing] any action of any kind to create, or have the effect of creating, a motor vehicle or motor vehicle engine different than a motor vehicle or engine certified in California under California standards (a ‘third vehicle’) or otherwise creating] such a ‘third vehicle’.” 42 U.S.C. § 7507. According to the district court, because ZEVs in New York would require distinct features to satisfy the demands of local climates, the ZEV quota violates this provision. See MVMA I, 810 F.Supp. at 1347.

An enormous amount of verbiage is spent by the parties debating whether the third vehicle prohibition is simply a clarification of the identicality requirement of § 177 or something more. As mentioned earlier, the third vehicle bar was not included in either the House or Senate versions of the 1990 amendments to the Act. It was added by the Conference Committee in October 1990. The Conference Report is devoid of any substantive discussion of the additional text, other than to observe that the amendment “make[s] plain that States exercising this section 177 option may not, in such adoption and enforcement, create a ‘third vehicles’ [sic] that is not a California vehicle or a 49-state Federal vehicles [sic], because of the burden it would place on the motor vehicle manufacturers.” H.R.Conf.Rep. No. 952, 101st .Cong., 2d Sess. 337 (1990), U.S.Code Cong. & Admin.News 1990, pp. 3385, 3869.

Resorting again to the plain language of the statute, it is clear that the third vehicle prohibition is not simply surplusage, but instead was meant to guard against several forms of state action. One clear purpose of the third vehicle prohibition is to bar a state from administering and enforcing standards identical to California’s in such a burdensome way as to effectively require a third vehicle. For instance, a state that has met the identi-cality requirement cannot adopt imprecise testing procedures to cheek compliance with those standards, or require remedial action for noncompliance with identical California emission standards that have the effect of imposing an “undue burden” on the manufacturer. With respect to the ZEV quota, New York has adopted California’s standard precisely and no party contends that New York is administering this standard or remedying its noncompliance in a manner or method different than California. Hence, the DEC regulation does not violate § 177’s third vehicle prohibition in respect to administration and enforcement.

Nor does the ZEV requirement violate the third-vehicle prohibition, apart from matters of administration and enforcement. It may be that a state would never violate the third vehicle prohibition so long as it meets the identicality requirement and does not administer or enforce the California emission standards in a manner more burdensome than what occurs in California. We need not go so far on this appeal in order to reject the manufacturers’ claim. They contend that, even if New York is deemed to have met the identicality requirement by adopting the LEV standard without the CF standards, a third vehicle has been required because in New York’s colder climate it will be necessary to install, a more powerful heater than the one the manufacturers plan to use in California. The short answer is that whatever heater the manufacturers choose to install on cars sold in New York is a marketing choice of theirs and not a requirement imposed by DEC. We note in that connection that California recently amended its definition of a ZEV to permit the inclusion of a fuel fired heater without altering the zero-emission status of the vehicle. See Cal.Code Regs. tit. 13 § 1900(b)(15).

Whether we could so easily reject a third-vehicle claim as to vehicles meeting California’s requirements if some aspect of New York’s climate required alteration to the emission control features of the automobile or substantially impaired the ability of the vehicle to perform basic transportation functions are issues we need not consider on this record. In the absence of facts even putting such claims in issue, the grant of summary judgment to the manufacturers based on the notion that the ZEV quota violates § 177’s third vehicle prohibition must be reversed, and summary judgment on this count must be entered for the DEC. No doubt as a result of the technology forcing nature of the Clean Air Act, today’s automobile as we know it is passing away. But the manufacturers’ argument with respect to the difficulty of building a viable ZEV is reminiscent of the view that 100 years ago some thought the U.S. Patent Office should be closed because anything that ever could be invented had already been invented.

CONCLUSION

Accordingly, for the reasons stated, the judgment of the district court is affirmed, in part, and reversed, in part in accordance with this opinion and the district court is directed to issue an injunction staying the enforcement of the Part 218 Regulations by DEC against plaintiff manufacturers for model year 1995.

No costs.

1 . Section 209(a) provides:

No State or any political subdivision thereof shall adopt or attempt to enforce any standard relating to the control of emissions from new motor vehicles or new motor vehicle engines subject to this part. No State shall require certification, inspection, or any other approval relating to the control of emissions from any new motor vehicle or new motor vehicle engine as condition precedent to the initial retail sale, titling (if any), or registration of such motor vehicle, motor vehicle engine, or equipment.

2 . Section 209(b) provides:

(1)The Administrator shall, after notice and opportunity for public hearing, waive application of this section to any State which has adopted standards (other than crankcase emission standards) for the control of emissions from new motor vehicles or new motor vehicle engines prior to March 30, 1966, if the State determines that the State standards will be, in the aggregate, at least as protective of public health and welfare as applicable Federal standards. No such waiver shall be granted if the Administrator finds that—

(A) the determination of the State is arbitrary and capricious,

(B) such State does not need such State standards to meet compelling and extraordinary conditions, or

(C) such State standards and accompanying enforcement procedures are not consistent with section 7521(a) of this title.

(2) If each State standard is at least as stringent as the comparable applicable Federal standard, such State standard shall be deemed to be at least as protective of health and welfare as such Federal standards for purposes of paragraph (1).

(3) In the case of any new motor vehicle or new motor vehicle engine to which State standards apply pursuant to a waiver granted under paragraph (1), compliance with such State standards shall be treated as compliance with applicable Federal standards for purposes of this subchapter.

3 . Section 177 in its current form provides:

Notwithstanding section 7543(a) of this title, any State which has plan provisions approved under this part may adopt and enforce for any model year standards relating to control of emissions from new motor vehicles or new motor vehicle engines and take such other actions as are referred to in section 7543(a) of this title respecting such vehicles if—

(1) such standards are identical to the California standards for which a waiver has been granted for such model year, and

(2) California and such State adopt such standards at least two years before commencement of such model year (as determined by regulations of the Administrator).

Nothing in this section or in subchapter II of this chapter shall be construed as authorizing any such State to prohibit or limit, directly or indirectly, the manufacture or sale of a new motor vehicle or motor vehicle engine that is certified in California as meeting California standards, or to take any action of any kind to create, or have the effect of creating, a motor vehicle or motor vehicle engine different than a motor vehicle or engine certified in California under California standards (a "third vehicle") or otherwise create such a "third vehicle".

4 . Section 302(k) provides:

(k) The terms "emission limitation” and "emission standard" mean a requirement established by the State or the Administrator which limits the quantity, rate, or concentration of emissions of air pollutants on a continuous basis, including any requirement relating to the operation or maintenance of a source to assure continuous emission reduction, and any design, equipment, work practice or operational standard promulgated under this chapter.