State Farm Mutual Automobile Insurance Company v. Elizabeth Dole, Secretary of the Department of Transportation, American Insurance Association v. Elizabeth Dole, Secretary of the Department of Transportation the Department of Transportation Diane Steed, Administrator of the National Highway Traffic Safety Administration and the National Highway Traffic Safety Administration, Nationwide Mutual Insurance Company v. Elizabeth Dole, Secretary of the Department of Transportation the Department of Transportation Diane Steed, Administrator of the National Highway Traffic Safety Administration and the National Highway Traffic Safety Administration, National Association of Insurance Commissioners v. Elizabeth Dole, Secretary of the Department of Transportation the Department of Transportation Diane Steed, Administrator of the National Highway Traffic Safety Administration and the National Highway Traffic Safety Administration, the State of New York, Robert Abrams, Attorney General of the State of New York, James P. Corcoran, Superintendent of Insurance of the State of New York and the New York State Department of Insurance v. Elizabeth Dole, Secretary of the Department of Transportation

802 F.2d 474

• Court: Court of Appeals for the D.C. Circuit
• Decided: September 18, 1986
• Docket: 84-1301
• Status: Published

Opinion

802 F.2d 474

255 U.S.App.D.C. 398, 55 USLW 2186

STATE FARM MUTUAL AUTOMOBILE INSURANCE COMPANY, et al., Petitioners,
v.
Elizabeth DOLE, Secretary of the Department of
Transportation, et al., Respondents.
AMERICAN INSURANCE ASSOCIATION, Petitioner,
v.
Elizabeth DOLE, Secretary of the Department of
Transportation; the Department of Transportation; Diane
Steed, Administrator of the National Highway Traffic Safety
Administration; and the National Highway Traffic Safety
Administration, Respondents.
NATIONWIDE MUTUAL INSURANCE COMPANY, Petitioner,
v.
Elizabeth DOLE, Secretary of the Department of
Transportation; the Department of Transportation; Diane
Steed, Administrator of the National Highway Traffic Safety
Administration; and the National Highway Traffic Safety
Administration, Respondents.
NATIONAL ASSOCIATION OF INSURANCE COMMISSIONERS, Petitioner,
v.
Elizabeth DOLE, Secretary of the Department of
Transportation; the Department of Transportation; Diane
Steed, Administrator of the National Highway Traffic Safety
Administration; and the National Highway Traffic Safety
Administration, Respondents.
The STATE OF NEW YORK, Robert Abrams, Attorney General of
the State of New York, James P. Corcoran, Superintendent of
Insurance of the State of New York and the New York State
Department of Insurance, Petitioners,
v.
Elizabeth DOLE, Secretary of the Department of
Transportation, et al., Respondents.

Nos. 84-1301, 84-1459 to 84-1462.

United States Court of Appeals,
District of Columbia Circuit.

Argued March 11, 1986.
Decided Sept. 18, 1986.

James F. Fitzpatrick, with whom Michael N. Sohn, John M. Quinn, Merrick B. Garland and Charles A. Taylor, III, Washington, D.C., were on the brief, for petitioners, State Farm Mut. Auto. Ins. Co., et al. in Nos. 84-1301, 84-1459 and 84-1460.

Melvin Goldberg, with whom Robert Abrams, Peter Bienstock, Paul M. Glickman and Daniel D. Kaplan, New York City, were on the brief, for petitioners, The State of N.Y., et al. in No. 84-1462.

Robert H. Myers, Jr., Washington, D.C., was on the brief, for petitioner, Nat. Ass'n of Ins. Com'rs in No. 84-1461.

Douglas Letter, Atty., Dept. of Justice, with whom Richard K. Willard, Acting Asst. Atty. Gen., Dept. of Justice, Kenneth N. Weinstein, Deputy Asst. Gen. Counsel, Dept. of Transp., Frank Berndt, Chief Counsel, Nat. Highway Traffic Safety Admin., Paul Blankenstein, Atty., Dept. of Justice and Stephen P. Wood, Atty., Nat. Highway Traffic Safety Admin., Washington, D.C., were on the brief, for respondents in Nos. 84-1301, 84-1459, 84-1460, 84-1461 and 84-1462.

Christopher D. Coppin, Asst. Atty. Gen., State of N.M., Albuquerque, N.M., was on the brief, for amicus curiae, The State of N.M. urging reversal in Nos. 84-1301, 84-1459, 84-1460 and 84-1461.

Philip R. Collins, Washington, D.C., was on the brief, for amicus curiae, Automotive Occupant Protection Ass'n urging reversal in Nos. 84-1301, 84-1459, 84-1460 and 84-1461.

Debbie M. Zuckerman was on the brief, for amicus curiae, Epilepsy Foundation of America, urging reversal in Nos. 84-1301, 84-1459, 84-1460 and 84-1461.

Dennis J. Barbour, Roanoke, Va., was on the brief, for amicus curiae, American Academy of Pediatrics, et al. urging reversal in Nos. 84-1301, 84-1459, 84-1460 and 84-1461.

Jerris Leonard, Washington, D.C., was on the brief, for amicus curiae, Conference of Ins. Legislators urging reversal in Nos. 84-1301, 84-1459, 84-1460 and 84-1461.

Before MIKVA, SCALIA and STARR, Circuit Judges.

Opinion for the Court filed by Circuit Judge STARR.

Opinion concurring in part and dissenting in part filed by Circuit Judge MIKVA.

STARR, Circuit Judge:

These consolidated cases bring us once more into the long-standing controversy over the Department of Transportation's regulations with respect to "passive restraints" in automobiles. Unlike its most recent predecessor, the rule at issue requires the phased-in installation of automatic (i.e., passive) protection devices in new cars manufactured for sale in the United States beginning September 1986.1 The rule, however, contains a provision under which the Secretary of Transportation will rescind the requirement if, by April 1, 1989, States covering two-thirds of the Nation's population enact mandatory safety belt usage laws. This provision is challenged by all petitioners2 as both contrary to the applicable statute and as arbitrary and capricious. While joining in this common attack, the State of New York mounts a separate challenge to the Secretary's decision not to adopt certain alternative standards in the final regulation.

We hold that the attack upon the rescission feature of the regulation is not ripe for judicial review. We also hold that New York's separate challenge, while ripe, fails on the merits.

* In response to high death tolls on our Nation's highways, Congress enacted the National Traffic and Motor Vehicle Safety Act of 1966 (the Safety Act), 15 U.S.C. Secs. 1381 et seq. (1982). The Safety Act was intended "to reduce traffic accidents and deaths and injuries to persons resulting from traffic accidents." Id. Sec. 1381. To that end, the statute directed the Secretary to "establish by order appropriate Federal motor vehicle safety standards" that are "practicable, [and] meet the need for motor vehicle safety...." Id. Sec. 1392(a).

Under this broad mandate, the Department in 1967 promulgated Federal Motor Vehicle Safety Standard 208, which required installation of manual safety belts in all cars. 32 Fed.Reg. 2408, 2415 (1967) (Standard 208). Two years later, however, the Department initiated consideration of automatic or passive occupant protection technology since the level of safety belt usage was quite low. As a result of this inquiry, the Department in 1972 adopted an amendment to Standard 208 requiring "complete passive protection" on automobiles manufactured after August 15, 1975. 37 Fed.Reg. 3911 (1972). Standard 208 was subsequently reconsidered and modified a number of times.3 Ultimately, it was amended to require the phasing-in of passive restraints beginning with the 1982 automobile model year. 42 Fed.Reg. 34,289 (1977). This Modified Standard 208 was upheld on review by this Court. See Pacific Legal Foundation v. Department of Transportation, 593 F.2d 1338 (D.C.Cir.), cert. denied, 444 U.S. 830, 100 S.Ct. 57, 62 L.Ed.2d 38 (1979).

In February 1981, the Department reopened the rulemaking that had produced Modified Standard 208. 46 Fed.Reg. 12,033 (1981). Two months later, it postponed the date on which phase-in of passive restraints was to begin, 46 Fed.Reg. 21,172 (1981), and proposed the possible rescission of the entire standard, id. at 21,205. Following a comment period and public hearings, the Department concluded that a reliable basis no longer existed on which to conclude that the passive restraint requirements would have significant safety benefits. In view of the substantial costs of implementing the requirement and the lack of a viable alternative, the agency simply rescinded the standard. 46 Fed.Reg. 53,419 (1981). This action was overturned by this court in State Farm Mutual Automobile Insurance Co. v. Department of Transportation, 680 F.2d 206 (D.C.Cir.1982). The Supreme Court vacated our judgment, but agreed that the agency's action was arbitrary and capricious. The matter was remanded to the Secretary for further consideration. See Motor Vehicle Manufacturers Association v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29, 57, 103 S.Ct. 2856, 2873, 77 L.Ed.2d 443 (1983) (State Farm ).

It is the Secretary's determination upon reconsideration that is now before us. After suspending the effective date of Modified Standard 208 for one year, 48 Fed.Reg. 39,908 (1983), the Department issued a notice of proposed rulemaking and requested comments regarding what action it should take with respect to passive restraints, 48 Fed.Reg. 48,622 (1983).4 Following the close of the comment period, the Department issued a supplemental notice of proposed rulemaking to gather additional comments. 49 Fed.Reg. 20,460 (1984).5 On July 17, 1984, the Department published its final rule amending Modified Standard 208 to require passive restraints. 49 Fed.Reg. 28,962 (1984) (codified at 49 C.F.R. Sec. 571.208 (1984)) (Final Rule). The Final Rule can be satisfied in several ways, including airbags, enhanced padding of the automobile interior, and [ ] either detachable or non-detachable automatic belts. During the first few years after the standard takes effect, the Final Rule creates special incentives for installation of passive protection systems other than automatic belts. Specifically, for each car in which an airbag or passive interior system is installed, the manufacturer will be given credit for an extra one-half automobile toward its percentage requirement. Id. at 29,000.

The Final Rule requires the phasing-in of passive occupant protection in all passenger cars beginning September 1, 1986. Id. at 28,963. But there is another feature of the Final Rule which has drawn the petitioners' attack. The passive restraint requirements embodied in the new standard will be rescinded if by April 1, 1989, two-thirds of the population of the United States is covered by mandatory usage laws (MULs) which meet certain specified conditions. Id.6 It is this "trap door" provision, as petitioners colorfully put it, that provides the focal point of the various challenges before us. To exacerbate matters, as petitioners see it, the Final Rule further provides that the Secretary will consider waiving the minimum requirements for States that had "substantially complying" MULs in place prior to August 1, 1984. Id. at 28,999. The automatic occupant protection requirement will be rescinded immediately upon the Secretary's determination that the requisite two-thirds population level is reached.7

II

We first address the ripeness issues raised by the Secretary. After setting forth some general principles to guide our analysis, we then examine each issue raised by petitioners to determine whether it is ripe for review.

* The ripeness doctrine limits the power of federal courts in adjudicating disputes. Its roots are found in both the Article III requirement of "case or controversy" and prudential considerations favoring the orderly conduct of the administrative and judicial processes. See Regional Rail Reorganization Act Cases, 419 U.S. 102, 138, 95 S.Ct. 335, 356, 42 L.Ed.2d 320 (1974); Eagle-Picher Industries v. EPA, 759 F.2d 905, 912 (D.C.Cir.1985).8 According to the leading Supreme Court case on the subject, Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967), the ripeness doctrine is designed to protect both Article II and Article III interests. First, it is intended "to prevent the courts, through avoidance of premature adjudication, from entangling themselves in abstract disagreements over administrative policies." 387 U.S. at 148, 87 S.Ct. at 1515. Second, the doctrine is intended "to protect the agencies from judicial interference until an administrative decision has been formalized and its effects felt in a concrete way by the challenging parties." Id.

Abbott Laboratories set forth a now familiar two-part test for deciding whether an agency action is ripe for review, focusing on both (1) "the fitness of the issues for judicial decision," and (2) "the hardship to the parties of withholding court consideration." Id. at 149, 87 S.Ct. at 1515. Like other legal inquiries, application of this test is by no means an exact science; nor is it to be a matter of weaving "complicated legal distinctions" divorced from reality. See Midwestern Gas Transmission Co. v. FERC, 589 F.2d 603, 618 (D.C.Cir.1978); Continental Air Lines, Inc. v. CAB, 522 F.2d 107, 128 (D.C.Cir.1974). It requires, rather, the exercise of "practical common sense," faithful to the considerable body of law which guides us. Id. at 124.

Under the first prong of the Abbott Labs test, the court considers any institutional interests that either the court or the agency may have for postponing review. See Eagle-Picher, 759 F.2d at 915. Under this branch of our analysis, we consider such matters as whether the agency's action is final and whether the issue is a purely legal one. See, e.g., Abbott Laboratories, 387 U.S. at 149, 87 S.Ct. at 1515; Continental Air Lines, 522 F.2d at 126. But even when agency action is final and the issues presented are purely legal, a court may nonetheless properly deem a matter unfit for resolution if postponing review would provide for a more efficient examination and disposition of the issues. See Toilet Goods Association v. Gardner, 387 U.S. 158, 163-64, 87 S.Ct. 1520, 1524-25, 18 L.Ed.2d 697 (1967); Alascom, Inc. v. FCC, 727 F.2d 1212, 1217 (D.C.Cir.1984); Midwestern Gas, 589 F.2d at 620. The court, for example, might determine that "further administrative action is needed to clarify the agency's position," Action Alliance of Senior Citizens v. Heckler, 789 F.2d 931, 940 (D.C.Cir.1986), or that the court's deliberations might benefit from letting the question arise "in some more concrete and final form," Eagle-Picher, 759 F.2d at 915 (quoting Continental Air Lines, 522 F.2d at 125); or that resolution of the dispute is likely to prove unnecessary, see id. For all these reasons, courts and agencies have a legitimate interest in avoiding adjudication of speculative controversies.

The second prong of the Abbott Labs test requires consideration of the countervailing interests of the challenging parties in obtaining a prompt resolution of their dispute. See id. at 915. It is well settled that for an institutional interest in deferral to be outweighed, postponing review must impose a hardship on the complaining party that is immediate, direct, and significant. See Abbott Laboratories, 387 U.S. at 152-53, 87 S.Ct. at 1517-18; Action Alliance, at 940. Thus, as in Abbott Labs itself, an agency enforcement policy may impose the requisite hardship even before the policy is implemented if, for example, it would reasonably prompt a regulated industry, unwilling to risk substantial penalties by defying the policy, to undertake costly compliance measures. See, e.g., Abbott Laboratories, 387 U.S. at 152-53, 87 S.Ct. at 1517-18. On the other hand, a party's allegation of hardship will be found wanting if there are too many "ifs" in the asserted causal chain linking the agency's action to the alleged hardship, see, e.g., Tennessee Gas Pipeline Co. v. FERC, 736 F.2d 747, 750 (D.C.Cir.1984), or if the asserted hardship is not sufficiently concrete, see, e.g., Abbott Laboratories, 387 U.S. at 148, 87 S.Ct. at 1515. "The mere potential for future injury," moreover, is not enough. Alascom, 727 F.2d at 1217 (emphasis in original).

We recognize that even though "the courts might prefer to resolve a particular question at another time and place, they should have a very good reason for indulging that preference, if in doing so they are refusing a petitioner's request to be relieved of an onerous legal uncertainty." Continental Air Lines, 522 F.2d at 128. But, if the interests of the court and agency in postponing review outweigh the interests of those seeking relief, settled principles of ripeness squarely call for adjudication to be postponed.

B

With one exception, we conclude that in the case before us the uncertainties occasioned by postponing review are not sufficiently onerous--if indeed they are onerous at all--to overcome the institutional interests in postponement. The exception, as we will explain later, is New York's challenge to the Secretary's decision rejecting certain alternatives in her formulation of the passive restraint standard.

* All of the petitioners contend that, for various reasons, the automatic rescission feature of the Secretary's rule is arbitrary and capricious.9 This issue, we are persuaded, is not ripe for review under the Abbott Labs analysis.

To be sure, the Secretary's Final Rule constitutes final agency action. And, the question whether the rescission provision is arbitrary and capricious requires no further factual development inasmuch as this question must be resolved on the basis of the administrative record before the Secretary at the time of her decision.10

Nonetheless, the institutional interest in avoiding speculative controversies is powerfully present here because the evidence before us indicates that the possible rescission about which petitioners are vexed will likely never occur. According to data supplied by one petitioner, State Farm Insurance Co., twenty States, covering sixty percent of the U.S. population, have now passed mandatory usage laws. None of these laws, however, apparently complies with the Secretary's specific requirements. See State Farm Supplemental Brief at 7-10, 1a.11 Even if all remaining States were to pass complying MULs (and assuming that New York's pre-Final Rule MUL is counted under the Secretary's waiver provision), some States with MULs now on the books would have to amend their statutes to comply with the Final Rule in order for the two-thirds population requirement to be met.12 On the record before us, it appears singularly unlikely that the passive restraint standard will be rescinded by 1989.13 Failure to rescind the standard at that time would, of course, render the assault on the provision moot. Since it appears unlikely that the "trap door" will ever be opened, a decision on this issue may very well prove unnecessary.

The institutional interest in postponing review has not been counterbalanced by the requisite showing of hardship. Petitioners allege hardship resulting from (1) deaths that will likely occur if the Secretary's rule is rescinded; (2) the effects of the rescission provision on the technological development of passive restraint systems; and (3) the effect of the provision on the States and, in turn, on the insurance companies' and insurance commissioners' lobbying efforts. We examine each of these in turn.

The first concern is clearly misplaced and need not detain us. The evil feared by petitioners will never eventuate if rescission is never effected. If, on the other hand, the "trap door" appeared imminently ready to open, then petitioners could avail themselves of further judicial proceedings--including seeking a stay--to prevent that event from taking place. See Tennessee Gas, 736 F.2d at 751; Air New Zealand v. CAB, 726 F.2d 832, 837 (D.C.Cir.1984). There will be ample opportunity to challenge rescission if it appears imminent at any time during the Secretary's three-year window of opportunity (or, as petitioners see it, a window of vulnerability).

Petitioners' second concern is that the rescission provision may be having an immediate, adverse effect on automakers' incentives to develop passive restraint systems. They fear that automakers will delay development of such systems, with the result that the passive restraints which are eventually installed will be inferior and undertested. This, presumably, would affect the insurance companies' ultimate liabilities and, hence, their rates.

But this argument does not withstand analysis. First, it fails to take adequately into account the practical reality that the automobile industry must begin manufacturing cars with passive restraints in September of this year.14 See supra note 7. Second, and more fundamentally, the challenge is rife with speculation about what may occur in the automobile industry. Lacking any evidence to support their view, petitioners would have us gaze into a crystal ball to determine what Detroit may or may not do. On such musings judicial review cannot properly be grounded.15

The third and, according to petitioners, most serious hardship spawned by postponing review is to "weaken state efforts to enact tough MULs." State Farm Reply Brief at 9; NAIC Reply Brief at 5-6. Petitioners reason that state legislatures are forced by the Final Rule to choose between a federal passive restraint standard and strong MULs.

The threshold and, in our view, fatal difficulty with this argument is that the petitioners seek to assert the hardships of States that are not before the court. Abbott Laboratories requires that we look at the "hardship to the parties." 387 U.S. at 149, 87 S.Ct. at 1515. New York is the only State to have petitioned for review of the Secretary's Final Rule, and New York had already passed an MUL before the Final Rule was promulgated. By virtue of its pre-MUL being firmly in place, New York is left to contend only that "if New York soon determines that its MUL needs to be strengthened, the State will face a serious dilemma." New York Reply Brief at 6 (emphasis added).16 This is, again, a speculative scenario, not an indication of immediate or concrete hardship of the kind demanded under the cases.17

Petitioners also assert that the rescission provision produces hardship for States not contemplating MULs. According to petitioners, such States--already in want of a MUL--could suffer the further deprivation of a protective federal passive restraint standard if as few as sixteen other States pass qualifying MULs. See NAIC Reply Brief at 5-6. The State of New Mexico, one of the amici here, also raises this argument. See Amicus Curiae Brief of New Mexico at 3. As we have already established, however, this asserted hardship is not cognizable because it is not being asserted by any of the parties before the court. Assuming arguendo that this argument could properly be advanced by petitioners (or an amicus), this alleged hardship is still illusory. The state of affairs petitioners fear will come about, again, only if the Secretary in fact rescinds the passive restraint standard. As we have by now recounted a bit tediously, judicial review will be available should that time in fact appear to be at hand. It is not at hand, and perhaps will never arrive. Postponing review until that contingency materializes will impose no hardship on those States.

The direct hardship asserted by the insurance companies and commissioners is likewise inadequate. The principal effect of the rescission provision on them, as they see it, is to hamper them in their lobbying efforts to secure enactment of stringent MULs in the several States by virtue of the risk that success on that front would insidiously undermine the federal passive restraint standard. See, e.g., NAIC Reply Brief at 6. These petitioners, in a word, want both federal passive restraint requirements and tough MULs. They emphatically do not want to trade off one for the other. But uncertainty in lobbying strategy scarcely rises to the level of concrete hardship. It seems similar, upon analysis, to the "planning uncertainty" which our prior cases have rejected as insufficient. See, e.g., Tennessee Gas, 736 F.2d at 749-50 (rejecting as insufficient the assertion that the agency's interpretation affected current business planning); Diamond Shamrock Corp. v. Costle, 580 F.2d 670, 673 (D.C.Cir.1978) (rejecting as insufficient the contention by dischargers that EPA's new effluent permit regulations placed them in "acute dilemma" that affected their business planning, even before seeking permits under the new regulations). And, in view of the unlikelihood that rescission will ever occur, petitioners' uncertainty seems minimal at best.182

The second broad issue raised by petitioners is whether the Secretary's automatic rescission is contrary to the Safety Act. Specifically, petitioners contend that (1) by abandoning the passive restraint standard in favor of MULs passed by some--but not all--of the States, the Secretary violates her statutory duty to ensure that safety standards be uniform; (2) by abandoning the standard in favor of State MULs, the Secretary violates her statutory duty to put in place federal standards; and (3) the automatic rescission provision is unlawful because the statute does not confer upon the Secretary authority to seek to influence safety legislation in the States. We find these contentions, upon analysis, similarly unripe.

The first two contentions depend, of course, upon rescission of the Final Rule; unless rescission comes, the evils identified by petitioners in these two respects will never come to pass. The harm allegedly flowing from the replacement of a uniform federal standard by a patchwork of state legislation will occur only when (if ever) the uniform federal standard is rescinded. No one contends, nor could they reasonably, that the substantive (i.e., nonrescission) provisions of the Final Rule are in any wise non-uniform. Since the Secretary's substantive standards are at present national in scope and uniform in nature (and since, as already discussed, the rescission provision will likely never take effect), a substantial likelihood exists that resolution of these two issues will prove unnecessary. For the same reasons that petitioners' "arbitrary and capricious" attack on the rescission provision is unripe, these two arguments must likewise be deemed unripe.

The third statutory challenge, advanced only by New York, is also unripe, but requires a somewhat different analysis. New York claims, in essence, that the Secretary is attempting through the rescission provision to exert a here-and-now influence on state legislation. This, New York contends, the Secretary has no statutory power to do. New York further maintains that the court has no interest in waiting to see whether the Secretary's contemplated action ever takes place, for the relevant action has already occurred and indeed continues to occur. This issue is therefore, New York argues, more "fit" for resolution than the issues which we have already discussed.

We are unpersuaded that this issue is ripe. Although we agree that the Secretary's alleged violation (as New York sees it) of the Safety Act has already occurred, this issue is still unfit for review because it has not yet arisen in a sufficiently concrete setting. As we have seen, New York, the only party raising this issue, has not alleged the existence of any proposed or pending legislation in that State on which the Secretary's asserted violation is having any purported effect. As to New York itself, then, the Secretary's alleged "coercion" of the States is obviously speculation upon speculation. Moreover, neither New Mexico (which, as we noted before, filed an amicus brief) nor any other State is before us contending that its legislative process is presently being affected by the Secretary's alleged coercion. Although the issue New York raises appears to be a legal one, it would be helpful to the court to see how this alleged coercion actually operates in practice. See, e.g., Toilet Goods Association, 387 U.S. at 164, 87 S.Ct. at 1524 (although issues raised are legal, judicial appraisal of FDA regulations would "stand on a much surer footing in the context of a specific application"). In addition, postponing review until such a situation allegedly arises would help "assure that concrete adverseness which sharpens the presentation of issues." Baker v. Carr, 369 U.S. 186, 204, 82 S.Ct. 691, 703, 7 L.Ed.2d 663 (1962) (discussing rationale for standing requirement in setting of constitutional adjudication).19 Turning to the second prong of the Abbott Labs test, neither New York nor any other State alleges, as we have seen, any direct, present hardship arising from the Secretary's alleged violation of the Safety Act in this respect. Under these circumstances, the alleged hardship created by postponing review of this issue is insufficient to outweigh the institutional interest in postponement.

In sum, we find that all the issues raised by petitioners' attacks on the Secretary's rescission provision are not yet ripe for review. That provision has not been implemented. It may well never be implemented. The hardships alleged by petitioners are either not their own or rest upon multiple layers of speculation. Were we to entertain these contentions, "we would venture away from the domain of judicial review into a realm more accurately described as judicial preview." Tennessee Gas, 736 F.2d at 751. We have not been commissioned with a "roving preview function," see id., and we decline the invitation to assume that role ourselves.20

Of the parties before us, only the State of New York challenges the Secretary's decision not to require either airbags or non-detachable automatic belts as the sole mechanisms for satisfying the passive restraint requirement. New York contends that this decision is "arbitrary and capricious" under the Administrative Procedure Act, 5 U.S.C. Sec. 706(2)(A) (1982). The issue is ripe for review, for reasons we shall now describe.

First, the issue fully satisfies the fitness prong of the Abbott Labs inquiry since neither the court nor the agency has anything to gain by postponing review. The agency's decision is final. Resolution of the issue requires no more factual development than that already contained in the administrative record. No further administrative action is needed to clarify the agency's position. Indeed, it appears that the Department intends to take no further action on this issue; in consequence, no future agency action or proceeding looms on the horizon that would permit us to test the effect of the agency's decision in a more concrete setting.

Moving to Abbott Labs' second prong, New York has alleged substantial hardships resulting from postponing review. The phasing in of passive occupant protection mandated by the Secretary's rule is to begin straight away, effective September 1, 1986. If New York is correct in asserting that the Final Rule is more lenient (and therefore less promotive of automobile safety) than is justified by the record, then many people, including New York citizens, may be adversely affected in the most direct way by the Secretary's failure to require greater protection. In the absence of countervailing institutional interests, we are satisfied that New York has alleged in this respect a sufficient hardship so as to make its contention ripe for review.21III

We turn then to the merits of the one ripe issue--New York's challenge to the Secretary's failure to implement certain alternatives suggested during the course of the rulemaking. In order to prevail, New York bears the burden of establishing that the agency's action was "arbitrary or capricious." See, e.g., National Association of Regulatory Utility Commissioners v. FCC, 746 F.2d 1492, 1502 (D.C.Cir.1984).22 It hardly bears repeating that this is a narrow standard of review, one which forbids us from substituting our judgment for that of the agency. Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 823, 28 L.Ed.2d 136 (1971). We must, of course, engage in a "searching and careful" review of the agency's reasoning, see id., and avoid becoming a "rubberstamp" for the agency, see Bureau of Alcohol, Tobacco and Firearms v. FLRA, 464 U.S. 89, 97, 104 S.Ct. 439, 444, 78 L.Ed.2d 195 (1983) (quoting NLRB v. Brown, 380 U.S. 278, 291-92, 85 S.Ct. 980, 988-89, 13 L.Ed.2d 839 (1965)), or letting deference to the agency's judgments slip into "judicial inertia," see id. (quoting American Shipbuilding Co. v. NLRB, 380 U.S. 300, 318, 85 S.Ct. 955, 967, 13 L.Ed.2d 855 (1965)). But we may nonetheless overturn agency action only where a "clear error of judgment" has occurred. Bowman Transportation, Inc. v. Arkansas-Best Freight System, Inc., 419 U.S. 281, 285, 95 S.Ct. 438, 441, 42 L.Ed.2d 447 (1974); Overton Park, 401 U.S. at 416, 91 S.Ct. at 823. The agency must be upheld as long as it has articulated a satisfactory explanation for its action, including a " 'rational connection between the facts found and the choice made.' " State Farm, 463 U.S. at 43, 103 S.Ct. at 2866 (1983) (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962)).

As is by now common ground in such cases, the agency may be overturned if, for example, it "has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise." Id. Although we may not make up for deficiencies in the agency's analysis, see SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 1577, 91 L.Ed. 1995 (1947), we will "uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned[,]" Bowman Transportation, 419 U.S. at 286, 95 S.Ct. at 442.

New York contends that the passive restraint standard contained in the Final Rule fails the "arbitrary and capricious" test in three respects. First, New York asserts that the Department should have excluded the option of meeting the passive restraint requirement with detachable automatic belts, see New York Brief at 29-32;23 second, it maintains that the Secretary should have required airbags in all cars, see id. at 33-36; and finally, it contends that the Secretary failed altogether to consider the option of requiring both airbags and non-detachable belts, see id. at 32.

* The Secretary fully considered the suggestion that, as among automatic belts, only non-detachable rather than detachable belts be deemed to meet the federal standard. Her refusal to embrace the idea was based primarily upon two factors. First, since a non-detachable belt is, according to the Secretary, "the most coercive type of automatic restraint," 49 Fed.Reg. at 28,993, imposing that particular requirement would create a "serious adverse public reaction," id. at 29,002. New York contends that this conclusion "is admitted to be pure speculation and belied by surveys of consumer attitudes in the record." New York Brief at 30. New York fails, however, to support this contention with any citations to the record or to otherwise buttress its position. See id. The Secretary, by contrast, cites surveys finding that 10 to 20 percent of the public would be likely to cut non-detachable belts, thereby defeating the system. See 49 Fed.Reg. at 28,993. This sort of consideration is entirely appropriate to weigh in the balance; as we have previously held in the very context of passive restraint standards, the Department "cannot fulfill its statutory responsibility unless it considers popular reaction." Pacific Legal Foundation, 593 F.2d at 1345.

The second basis of the Secretary's refusal to eliminate detachable automatic belts from the passive restraint requirement was her judgment that non-detachable belts would effectively force manufacturers to eliminate the center front seat. See 49 Fed.Reg. at 28,993. According to the Secretary, even if the center seat were exempt from the automatic belt requirement, occupants of that seat would have difficulty getting past non-detachable belts to situate themselves at their front seat destination. See id. Although New York argues that this conclusion is not necessarily correct (since the Secretary could either require motorized automatic belts which come into place only when the doors are closed or, alternatively, could make an exception for cars with center seats), we cannot say that the Secretary's conclusion in this respect constitutes a "clear error of judgment," see Bowman Transportation, 419 U.S. at 285, 95 S.Ct. at 442, or is otherwise arbitrary and capricious.24

New York next attacks the Secretary's failure to mandate airbags in all cars. Consistent with the Supreme Court's admonition that she is obliged to consider this option, see State Farm, 463 U.S. at 46, 103 S.Ct. at 2868, the Secretary analyzed this possibility in considerable detail. See 49 Fed.Reg. at 28,990-92, 29,000-02.

Her ultimate decision not to require airbags, but to provide incentives for their employment, was based on two factors.25 First, despite their admitted safety benefits, airbags are costly: According to Department estimates, they would cost $320 more per car than manual belts; in addition, their replacement cost is an estimated $800, making it likely that many airbags would not be replaced once used. See 49 Fed.Reg. at 29,001.26 These cost factors were appropriately taken into account. The Supreme Court observed in State Farm that "[t]he agency is correct to look at the costs as well as the benefits of Standard 208." 463 U.S. at 54, 103 S.Ct. at 2872. Cf., e.g., American Textile Manufacturers Institute, Inc. v. Donovan, 452 U.S. 490, 509, 101 S.Ct. 2478, 2490, 69 L.Ed.2d 185 (1981) (overturning OSHA's cotton dust regulations, in part because OSHA had employed cost-benefit analysis in the face of statutory language mandating feasibility analysis). In light of these cost estimates, the Secretary concluded that the safety benefits of airbags would not be worth their high cost.27

New York vehemently challenges the agency's cost-benefit analysis in this arena of human safety. Specifically, New York complains that the standard takes into account neither the factor of pain and suffering nor the value of human life itself. See New York Brief at 33. Not so. The Final Rule devoted several pages to a discussion of the relative effectiveness of airbags, seatbelts, and passive restraints in reducing fatalities and serious injuries. See 49 Fed.Reg. at 28,984-87; 29,001.

In a related vein, New York takes the Secretary to task for relying upon the Department's own cost estimates rather than lower estimates found in the record. Again, we cannot say that the Secretary's cost estimates are arbitrary, particularly in light of other cost estimates submitted by the automobile manufacturers that run much higher than the Secretary's. In our view, such details of cost-benefit analysis are "most appropriately entrusted to the expertise of an agency," especially where, as here, the evidence runs in contrary directions. See Office of Communication of United Church of Christ v. FCC, 707 F.2d 1413, 1440 (D.C.Cir.1983).

The second basis for the Secretary's decision was public acceptability. As a threshold matter, the Secretary recognized that public acceptability would depend to a great extent on the cost of airbags to consumers, concluding that "only a small percentage appears willing to pay more than $400" for the devices. 49 Fed.Reg. at 28,988. She also took cognizance of public fears about chemicals used to deploy airbags, the possibility of inadvertent deployment of the devices, and the sense of insecurity harbored by some people at not having a belt wrapped around them. The Secretary reasoned that even though these fears are largely unfounded they must nonetheless be taken seriously. She opined that "[i]t may be easier to overcome these concerns if airbags are not the only way of complying with an automatic occupant protection requirement." 49 Fed.Reg. at 29,001. In short, the Secretary determined that these concerns could best be addressed through real-world experience in the marketplace rather than by regulatory fiat. New York's only response to this point is a single survey indicating that airbags enjoy a higher level of public acceptability than either automatic or manual belts. Particularly in light of a contrary public opinion survey and numerous public comments going in a contrary direction, see 49 Fed.Reg. at 28,988, we cannot say that the Secretary's refusal to give determinative weight to the survey championed by New York descended to the depths of arbitrary and capricious action.

New York's final attack on the Final Rule is that the Secretary failed to consider the alternative of requiring both airbags and non-detachable automatic belts. An agency, of course, is obliged to consider all practical, technologically feasible options. See State Farm, 463 U.S. at 48, 103 S.Ct. at 2869. The Government contends, however, that the Secretary did in fact consider this alternative and advanced adequate reasons for rejecting it.

We agree. Although the Final Rule is scarcely a model of clarity on this specific point, the Secretary's path may nonetheless reasonably be discerned. See Bowman Transportation, 419 U.S. at 286, 95 S.Ct. at 442. One of the subsections in the Secretary's discussion of her reasons for not adopting other alternatives was entitled "Airbags and/or Non-Detachable Seatbelts." That section discussed the weaknesses of non-detachable belts. See 49 Fed.Reg. at 29,002. It seems clear to us that this discussion was intended to respond both to the argument that detachable belts should be eliminated as a means of satisfying the passive restraint requirement and to the argument that non-detachable belts should be required in tandem with airbags. First, the perceived individual weaknesses of airbags and non-detachable belts are, standing alone, obviously germane to the desirability of requiring the combination of the two. Second, the Secretary expressly invoked her earlier discussion of various alternatives, including an outright requirement of airbags. There was obviously no need to repeat all the points on which she had already elaborated at considerable length.28 Third, had the Secretary intended in this section to address only the option of requiring airbags or non-detachable seat belts, one would think she would have chosen to entitle this subsection differently than she did.

As to the substance of the decision not to require both devices, we cannot discern any significant defects in the Secretary's reasoning. Having concluded that non-detachable automatic belts posed public-acceptability concerns sufficient to preclude requiring manufacturers to install them (rather than detachable belts), the Secretary could reasonably decide not to require that non-detachable belts be used in tandem with airbags. While not articulated with crystalline clarity, the Secretary's discussion of this issue passes muster under the "arbitrary and capricious" standard.

IV

For the foregoing reasons, the petition of the State of New York is denied insofar as it challenges the Secretary's decision not to require airbags or non-detachable seat belts. All of the other petitions, together with the remainder of New York's petition, are dismissed as unripe.

So Ordered.

MIKVA, Circuit Judge, concurring in part and dissenting in part:

I agree with my colleagues that most of the challenges to the Secretary's authority are not ripe for the review being sought. While I am troubled by the seeming absence of any statutory authority for the Secretary's stick and carrot dealings with the states, the time to confront that question frontally is not now and may never be, as far as most of the petitioners are concerned. I am not as sanguine about the New York challenges to the Secretary's actions. My colleagues agree that most of the New York complaints are timely and properly before this court. In my view, however, they have erroneously declared one claim untimely. Moreover, I believe that all the complaints are meritorious as well.

New York attacks the Secretary's Final Rule on four grounds. The majority addresses the merits of only three, upholding the agency's action as to each. Based on a misperception of the thrust of the fourth claim, the majority rules it not ripe for review. I find the agency's action in each of the four instances is ripe and fails to withstand judicial scrutiny under the arbitrary and capricious standard of review. Unlike the majority, I believe that the agency has failed to make the requisite rational connection between the facts in evidence and its judgment relating to New York's claims sufficient to pass muster under the arbitrary-and-capricious standard. Therefore, I respectfully dissent.

There is no disagreement about the standard or the evidentiary dimensions of our review. The Secretary's modification of the safety standards may be set aside if "found to be 'arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.' " See Motor Vehicle Manufacturers Association of the United States, Inc. v. State Farm Mutual Automobile Insurance Co., 463 U.S. 29, 41, 103 S.Ct. 2856, 2865, 77 L.Ed.2d 443 (1983) (citing 5 U.S.C. Sec. 706(2)(A)). An agency rule will be deemed arbitrary and capricious if "the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, or offered an explanation for its decision that runs counter to the evidence before the agency." Id. at 43, 103 S.Ct. at 2867. It is well settled that in examining the agency's actions under the arbitrary-and-capricious standard, we must confine our "searching and careful" review, see Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 824, 28 L.Ed.2d 136 (1971), to the evidence in the record. The agency record must reflect an adequate and reasonable basis for the decision, including consideration of all relevant factors. State Farm, 463 U.S. at 43, 103 S.Ct. at 2866; ILGWU v. Donovan, 722 F.2d 795, 822 (D.C.Cir.1983), cert. denied sub nom. Breen v. ILGWU, 469 U.S. 820, 105 S.Ct. 93, 83 L.Ed.2d 39 (1984). The reviewing court must judge the reasons and justifications the agency presents for its action and not supply a reasoned basis where the agency itself has provided none. SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 1577, 91 L.Ed. 1995 (1947). "[I]t is the agency's responsibility, not this [c]ourt's, to explain its decision." State Farm, 463 U.S. at 57, 103 S.Ct. at 2874.

* A

New York contends that the Secretary arbitrarily failed to adequately consider the alternative of permitting both mandatory usage laws (MULs) and a federal passive restraint standard. Maj. Op. at 480 n. 9 (citing New York Brief at 25-27). The majority incorrectly subsumes this claim into petitioners' challenge to the automatic rescission feature of the Secretary's Rule. Maj. Op. at 480 & n. 9. In so doing, the majority concludes that the claim is "not ripe for review." Maj. Op. at 480. Were I to agree that New York's claim is nothing more than an attack on the reasonableness of the "trapdoor" provision, I would concur in my colleagues' determination that the court must decline review of this issue, just as they have done when the other petitioners made this assertion. However, I believe the challenge is part of New York's contention that "the Final Rule is more lenient (and therefore less promotive of automobile safety) than is justified by the record." See Maj. Op. at 485.

New York is not questioning the safety effectiveness of MULs per se--an issue which need not be reached unless and until the trapdoor opens. Rather, it is challenging the Secretary's design of a passive restraint requirement upon consideration of only an either/or alternative, without also examining a combination approach. New York argues that the evidence in the record suggests that the combination of MULs plus passive restraints would provide greater safety benefits than the MUL-or-passive restraint option. That the Secretary effectuated her design through the automatic rescission provision does not change the import of the claim. Nor, in my estimation, does it affect the claim's timeliness. If the argument proves true, "then many people, including New York citizens, may be adversely affected in the most direct way by the Secretary's failure to require greater protection." See Maj. Op. at 485. Thus, even under the majority's reasoning, this issue is ripe for review. See id. Furthermore, upon examination of the record, I conclude that the agency's decision to reject this alternative is devoid of any meaningful support.

During the most recent round of rulemaking, numerous commentators argued that passive restraints should be viewed in conjunction with and not as an alternative to state laws requiring seatbelt use. See 49 Fed.Reg. 28,999 (1984). These comments were given passing address. In the fifty-page record, the Secretary responded with a one sentence indirect retort:

This argument ignores both the public acceptability concerns set forth above and the incentive for passage of such laws--to the extent there is significant consumer resistance to automatic protection devices--created by the department's approach.

Id. Examination of this option is nil. Instead, the Secretary steadfastly maintained that either automatic occupant protection or MULs covering two-thirds of the population would meet the "standards of the Act" and "carry out the objective and purpose of the statute," so long as the MULs met the Department's criteria. Id. Playing these alternatives against one another rather than in tandem led the Secretary to adopt the MUL rescission provision. She based her action on the Department's conclusion that "coverage of a large percentage of the American people by seatbelt laws that are enforced would largely negate the incremental increase in safety to be expected from an automatic protection requirement." Id. at 28,997. There are two problems with this conclusion.

First, it is completely unsupported, if not contradicted, by the relevant data. Nowhere in the record does the Department explain how it concluded that the incremental safety benefits of passive restraints, as compared to MULs alone, would not be worth the cost. To the contrary, the Department's regulatory impact analysis strongly suggests that passive restraints could well be cost-beneficial even if states passed MULs; it concedes that the combination could maximize both short-term and long-term safety benefits. See 1984 Final Regulatory Impast Analysis at VI-29 to -30 (J.A. 250-52); see also 49 Fed.Reg. at 28,991 (noting the advantages of passive restraints combined with seatbelt use). It stands to reason that only a combination of the two methods would ensure that the safety benefits are afforded the citizens and travellers in those states that did not enact MULs. Thus, contrary to the Department's conclusion, the evidence indicates that safety would be enhanced by permitting the two regulatory schemes to coexist.

Second, the Department's finding does not support the automatic rescission of the passive restraint requirement upon passage of MULs covering two-thirds of the states. Indeed, it cannot be supported in advance of any opportunity to assess the actual effect of state MULs in operation. The Department noted, "in order for it to accept MULs as an alternative to requiring automatic crash protection, MULs must provide a level of safety equivalent to that which would be expected to be provided ... by the automatic system." 49 Fed.Reg. at 28,999 (emphasis added). The Department's conclusion that MULs can be as effective as passive restraints standing alone, although supported by an extraordinarily thin record, must be respected. But this again formulates the inquiry only in terms of an either/or approach. Particularly given the fact that at the time of reporting only one American state had ever required seatbelt usage, it seems surpassingly capricious for the Department to decide that MULs covering only two-thirds of the population would render passive restraints superfluous. If MULs actually prove to be workable and so safety-effective that passive restraints are unnecessary, the Department can always rescind the passive restraint requirement at some later date--when there is hard evidence to prove what is now a most doubtful prediction.

In its second justification for dismissing a passive restraint-plus-MUL option, the Department noted tersely that the passive restraint-or-MUL approach both addresses the "significant consumer resistance" to automatic protective devices and offers states an "incentive" to pass MULs. The Secretary apparently believed that promulgating a passive restraint along with promotion of state MULs would both meet with intense public resistance to passive restraints and prevent her from effectively encouraging MUL enactment. Hence, neither element of the scheme would be accomplished. The Secretary's reasoning is far from comprehensible. To the extent it can be pieced together, it lacks record support.

In defending her treatment of the two safety protection methods as alternatives rather than as complements, the Secretary emphasized her belief in the importance of providing some local option in the decision-making. The "option" open to states is to express their preference for MULs over automatic occupant protection by passing such laws. The Department "believe[d] that offering this 'option' should lessen any public resistance to an automatic occupant protection requirement. Having some ability to choose one alternative over the other should make both alternatives more acceptable." Id. at 28,99 9. The evidence simply can not bear this strained reasoning.

In the same rulemaking preamble, the Department concluded that there was no reason to expect the passive restraint requirement would create significant "public acceptability concerns" or "consumer resistance." Id. at 29,002-01. Nor did it expect any negative reaction to passive restraints to seriously impede their efficacy. Id. at 28,989. Consequently, the Secretary had no basis for concluding that the MUL-only alternative was necessary to sweeten the passive restraint pill, or that state legislatures would be encouraged to pass MULs by the prospect of avoiding passive restraints. In fact, it is unclear why the Department thought the states needed to be encouraged at all to pass MULs. According to the record, the Department was impressed by evidence that public support for highway safety laws was growing steadily and dramatically. See id. at 28,994.

The evidence appears to run counter to the Secretary's incentive objective. New York contends that states would pass more stringent MULs but for the Secretary's trapdoor. Indeed, events since the Rule's promulgation indicate that New York is correct. The states that have passed MULs have designed weaker, less comprehensive statutes so as to avoid compliance with the Rule's criteria for consideration in the two-third's trapdoor figure. Other states have gone even further and provided for automatic repeal of their MULs should the Department count their populations toward the automatic rescission provision. New York Supp. Brief at 6.

Regardless of the reasonableness of the incentive device, the either/or scheme arrived at by the Secretary raises a serious statutory problem. The Secretary's overt intent in implementing the provision was to affect state legislation, specifically state legislation regulating driver behavior. That is a regulatory design specifically withheld from the Secretary by the statute under which she acted.

It is axiomatic that in implementing legislation the Secretary must perform in accordance with Congress' purposes in enacting the legislation. See Pacific Legal Foundation v. Department of Transportation, 593 F.2d 1338, 1343 (D.C.Cir.), cert. denied, 444 U.S. 830, 100 S.Ct. 57, 62 L.Ed.2d 38 (1979); SEC v. Chenery Corp., 332 U.S. 194, 67 S.Ct. 1575, 91 L.Ed. 1995 (1947). Accordingly, in reviewing the Department's decision we must ensure that the agency has "remain[ed] within the bounds of [its] delegated authority." See Office of Communication of United Church of Christ v. FCC, 707 F.2d 1413, 1422-23 (D.C.Cir.1983). Neither the ingenuity nor the efficacy of a regulatory scheme can save it from challenge as unauthorized action.

Both the language and legislative history of the National Traffic and Motor Safety Act of 1966 (the Act), 15 U.S.C. Secs. 1381 et seq. (1982), posit that Congress has authorized the Department to adopt automobile safety standards which mandate performance characteristics of equipment and vehicles--exclusively. The Act directs the Secretary to establish "motor vehicle safety standards," id. Sec. 1392(a), defined as "a minimum standard for motor vehicle performance, or motor vehicle equipment performance ...," id. Sec. 1391(2). " 'Motor vehicle safety' means the performance of motor vehicles or motor vehicle equipment...." Id. Sec. 1391(1). Behavior of drivers is not within the permitted scope of safety standards under the Act. Nor did the statute contemplate that the Secretary would delegate any of her automotive safety standards authority to the states. Nor did Congress intend such results.

When Congress has sought to permit the Secretary to influence driver behavior it has specifically authorized her to structure state programs. For example, pursuant to the Highway Safety Act of 1982, 23 U.S.C. Secs. 401 et seq. (1982), states must submit to the Secretary for approval driver education programs designed to reduce traffic accidents and resultant losses. Congress explicitly provided that "[s]uch programs shall be in accordance with uniform standards promulgated by the Secretary." It then went on to outline the performance criteria which she is to employ in designing the standards. Id. Sec. 402(a). Similarly, Congress itself has set up incentive schemes when it intended the Secretary to implement federal safety standards by delegating her authority to the states. The Surface Transportation Assistance Act of 1978, 23 U.S.C. Secs. 101 et seq. (1982), offers the best example. There Congress provided that the Secretary may not approve federal funds for highway construction and repairs in any state which has a maximum speed limit above fifty-five miles per hour. Id. Sec. 154(a). Congress adopted a more direct incentive in its approach to the drunk driving problem. It directed the Secretary to make grants to "states which adopt and implement effective programs to reduce traffic safety programs resulting from persons driving while [intoxicated or under the influence of drugs]." Id. Sec. 408. Congress then spelled out the minimum statutory provisions that the state must adopt in order to be eligible.

When Congress intends to authorize such action by the Secretary it says so in no uncertain terms. The authority cannot be implied. See Calvert Cliffs' Coordinating Committee, Inc. v. Atomic Energy Commission, 449 F.2d 1109, 1122-27 (D.C.Cir.1971). There are no parallel provisions in the National Traffic and Motor Vehicle Safety Act.

With the Act, Congress meant to shift the focus of federal automobile safety regulation away from the prior concern over the driver's actions and capacity and toward the "role of the car itself." S.Rep. No. 1301, 89th Cong., 2d Sess. 6 (1966), reprinted in 1966 U.S.Code Cong. & Ad.News 2710. The Supreme Court, in referring to the Department's power to adopt safety standards under the Act, observed that "Congress decided that at least part of the answer [to the problem of highway deaths and injuries] lies in improving the design and safety features of the vehicle itself." State Farm, 463 U.S. at 33, 103 S.Ct. at 2861. Since safety features would become effective automatically, regardless of human action or inaction, it was hoped that the benefits would be more readily felt. Regulation of driver behavior was left to the states.

In response to New York's challenge to her statutory authority, the Secretary contends that she is permitted to take state safety laws into account in determining whether federal standards will be "appropriate." This is unquestionably true, but misses the mark. While she may consider the effect of state laws which govern driver behavior, she may not seek to affect state laws which govern driver behavior. Mandatory seatbelt use laws regulate behavior; they render failure to "buckle-up" illegal and punishable. The MUL-or-passive restraint provision not only is intended to "encourage" states to adopt MULs, 49 Fed.Reg. at 28,998, 28,999, but it also seeks to set the terms of the regulation, to define the scope of the illegality and severity of the punishment. See 49 C.F.R. Sec. 571.208.S4.1.5.2 (1984) (setting out the four conditions necessary for MUL compliance with the provision). This "incentive" is provided to fulfill the safety objectives of the Act. 49 Fed.Reg. at 28,999. Such attempts at indirect regulation of driver (not to mention passenger) behavior traverses the bounds of the Department's statutory authority.

In sum, the Secretary's disregard for the coexistence of MULs and a passive restraint requirement is arbitrary and capricious for two reasons. First, the Department failed to substantiate its conclusion that MULs covering only two-thirds of the population would be more beneficial than an automatic restraint requirement combined with promotion of MULs; it overlooked the evidence that the combination might greatly increase lives saved and injuries prevented, compared to either method alone, let alone MULs covering only two-thirds of the population. Second, the Secretary's "incentive" approach is outside the agency's mandate under the Act in that its intent, if not its effect, is to influence driver behavior at the state level and its methodology is through delegation of strictly federal authority to the states.

AND

New York urges the court to hold that the evidence can only reasonably support the requirement of a nondetachable automatic belt. The majority disparages New York's arguments. Based on the Secretary's stated justifications, it upholds her decision to allow satisfaction of the passive restraint requirement through installation of detachable automatic belts. Maj. Op. at 487. I cannot understand her reasoning and agree with New York that the evidence all points the other way.

The Secretary reasoned that non-detachable belts are the most coercive passive restraints, and would therefore cause the strongest negative public reaction. 49 Fed.Reg. at 28,993, 29,002. As the majority correctly states, the Secretary may appropriately "weigh [this consideration] in the balance." See Maj. Op. at 487. But the relevant data supports neither the Secretary's "weighing" process nor her subsequent conclusion that detachable belts should be permitted.

The Secretary indicated that 10 to 20 percent of the public might cut non-detachable belts. 49 Fed.Reg. at 28,993. Nevertheless, she concluded that a large proportion of the public would remain protected by non-detachable belts. See id. at 29,003. The Secretary does not explain why this situation is not more protective of safety than detachable belts in all cars. Indeed, she cannot rationally explain why without first finding what percentage of the population would detach detachable belts. Obviously, if more than twenty percent of the public will not use detachable belts, the safety factor is clearly in favor of the non-detachable option. The evidence seems to indicate that more people will detach detachable belts than will sever non-detachable ones. See id. at 28,984. Thus, upon a fair viewing of the record, the non-detachable scenario appears to offer greater protection.

The Secretary offered a second rationale which the majority also finds convincing. She judged that non-detachable belts would require manufacturers to eliminate the center front seat, noting that "[t]here is no commercially developed technology to provide an automatic belt for the center seat." Id. at 28,993. While the manufacturers' contention was disputed, even if it is fully accurate it does not provide a complete rationale for the rejection of a non-detachable belt mandate.

Congress charged the Secretary with the protection of safety. See 15 U.S.C. Sec. 1392; State Farm, 463 U.S. at 55, 103 S.Ct. at 2873. Although the Act permits the Secretary to consider whether the proposed standard is "reasonable, practicable and appropriate" for the automobile for which it is prescribed, "[t]he Act intended that safety standards not depend on current technology and could be 'technology-forcing' in the sense of inducing the development of superior safety design." State Farm, 463 U.S. at 49, 103 S.Ct. at 2870 (admonishing the Department for creating more lenient standards in response to auto manufacturers' unwillingness to comply with safer devices). New York and the record suggest that technology already exists to alleviate the supposed problem. The Act directs the Secretary to "consider relevant available motor vehicle safety data" in establishing her standard. 15 U.S.C. Sec. 1392(f)(1). She may not sacrifice safety by failing to weigh all the relevant factors of each viable alternative.

The Secretary never balanced the "center-seat problem" against the non-detachable belt's advantages. She stressed only their disadvantages. See 49 Fed.Reg. at 28,992-93, 29,002. Furthermore, she overlooked evidence which mitigates the detractions of the option. For example, elsewhere in the report the Secretary noted that fewer than one-third of the cars sold in 1982 had center front seats and the number has been steadily declining. The center seat is rarely used and the vast majority of its occupants are small children who are covered independently by mandatory automatic child restraint laws in all but two states. Id. at 28,996. The relative number of cars and passengers affected by any inconvenience would appear to be very small, when compared with the increase in lives saved through increased use of non-detachable belts. See id. at 28,984.

Having failed to fairly view the relevant data before her and adequately balance the evidence in reaching her conclusion, the Secretary has abdicated her statutory duty to implement laws which enhance automobile safety. Accordingly, her decision to not require non-detachable seatbelts is arbitrary and capricious.

New York also challenges the Secretary's refusal to mandate installation of airbags in all automobiles. The majority upholds the Secretary's decision, simply noting that the Secretary concluded that airbags' safety would not be worth their high cost. See Maj. Op. at 488. I believe that the Secretary's decision "runs counter to the evidence before the agency" and therefore constitutes arbitrary and capricious rulemaking. See State Farm, 463 U.S. at 43, 103 S.Ct. at 2866.

As the Department noted, "[a]irbags offer a distinct advantage over other occupant restraints in that they ensure a usage rate of nearly 100 percent for both drivers and passengers." 49 Fed.Reg. at 28,991. According to the Department's research, under any reasonable scenario airbags save more lives than any other restraint. In addition, airbags prevent more moderate and critical injuries than do seatbelts. See id. at 28,984-86 & Table 5 (comparing relative effectiveness of airbags, seatbelts and automatic restraints). Despite this evidence of the superiority of airbags, the Secretary ultimately declined to require them.

The Secretary proffered two rationales for her decision. First, she cited financial cost. Id. at 28,990-01. The Department estimated that installation of airbags in all three front seat positions of a car would cost $320 more per car than manual belts. The majority also notes that the estimated replacement cost of a deployed airbag is $800. Maj. Op. at 488. The majority credits the Secretary's conclusion that "[i]n light of these cost estimates, ... the safety benefits of airbags would not be worth their high cost." Maj. Op. at 488. The conclusion cannot be reached from the facts that are used.

The majority reiterates the Secretary's conclusion that the replacement cost of airbags "mak[es] it likely that many airbags would not be replaced once used." Maj. Op. at 488. The presumed result is that "there would be no protection for the front seat occupants of [that] automobile." 49 Fed.Reg. at 29,001. I think the argument is almost silly. As the Department itself concedes, few airbags are inadvertently deployed. 49 Fed.Reg. at 28,984. The vast majority are deployed due to sudden deceleration from moderate or high speeds. Often this deceleration will be caused by or result in frontal impact. Some of these cars will be totalled and the cost of installing new airbags is moot. In the Department's own estimation, even when not allowing for wrecked cars, few cars would be affected by the high cost of replacement. See id. at 28,984 (non-replacement together with dismantling would leave only 2 percent of all cars without bags at any one time).

As to those cars that are worth repairing after a front-end collision in which the airbags were deployed, the Department failed to recognize that it should be considering a unique sub-group of the population. In accepting the Secretary's reasoning, the majority relies upon the Department's finding "that 'only a small percentage [of the public] appears willing to pay more than $400' " for airbags. Maj. Op. at 488, quoting 49 Fed.Reg. at 28,988. Surely this opinion poll might yield different results if the "public" was comprised of those who had previously been protected in accidents by deployed bags. The record does not indicate how resistant the consumer would be to paying $800 to replace the devices that saved the car's occupants from death or serious injury. If speculation were in order, I would venture that such a consumer would find $800 a reasonable investment.

Similarly, the Secretary mistakenly characterized the installation costs of airbags as being unacceptable to the public. The Department's $320 estimate falls well within the range found acceptable in its public opinion survey. See 49 Fed.Reg. at 28,988 ("a range of approximately $150-$350"). Admittedly, as the majority points out, automobile manufacturers' cost estimates are higher, but the record notes much lower estimates as well. Id. at 28,990. The Department specifically concluded that "[t]he costs of existing automatic restraint systems [including airbags] are reasonable...." Id. at 28,996. And yet the Secretary inexplicably finds the cost of airbags prohibitive.

In sum, the agency has misapplied the cost-benefit analysis dictated by the Act, and therefore is not due the deference to its "expertise" which the majority confers. As the majority observes, Maj. Op. at 488, the Supreme Court noted in State Farm that "[t]he agency is correct to look at the costs as well as the benefits of Standard 208." 463 U.S. at 54, 103 S.Ct. at 2872. But the Court also reminded the Department that when considering the reasonableness of the costs, it "should bear in mind that Congress intended safety to be the pre-eminent factor under the Act." 463 U.S. at 55, 103 S.Ct. at 2873 (citing H.R.Rep. No. 1776, 89th Cong., 2d Sess., 16 (1966)). The Court construed the Act as mandating an effective safety measure unless its incremental costs are more than minimal. See id. at 54-55, 103 S.Ct. at 2872-73. The Secretary appears to have eschewed this advice. She focused on costs, not safety. The $320 incremental cost for airbags in context appears minimal. The agency's fears that the public may resent paying more for airbags do not reasonably override the agency's own findings regarding the superior safety-effectiveness of airbags. Cf. S.Rep. No. 1301, 89th Cong., 2d Sess. 6 (1966), reprinted in 1966 U.S.Code Cong. & Ad.News 2709, 2714 ("safety shall be the overriding consideration in the issuance of standards").

The Secretary's second rationale for rejecting an airbag requirement is a forecasted lack of public enthusiasm. The Secretary noted, and the majority recounts, public fears regarding airbags--fears which the Department admits "can be adequately addressed." 49 Fed.Reg. at 29,001. In her explanation of the Final Rule, however, the Secretary fails to note, or apparently give any weight to, the Department's finding that "[a]irbags were rated highest [among manual belts, automatic belts and airbags] on comfort, convenience and appearance and were perceived to be safer than other restraint systems by infrequent belt users." Id. at 28,988. Thus, even on the public acceptability factor alone, the evidence is at best mixed for and against airbags.

The agency is entitled to consider anticipated adverse public reaction. See Pacific Legal Foundation, 593 F.2d at 1345. However, it may not extract only negative predictions and posture them as explanations for rejecting highly superior safety devices. The majority would permit the Secretary to leave the determination to the "real-world experience [of] the marketplace." Maj. Op. at 488. This solution seems facile. The Act charges the Secretary with meeting the needs of motor vehicle safety. Here, the Secretary has considered only one side of an argument that at best bears only tangentially on her decision and has given it paramount importance.

The record is overwhelming that airbags would save more lives and prevent more injuries than automatic belts or any reasonable scenario of seatbelt use. The Secretary has inadequately explained her reasons for reaching a finding which is at odds with this evidence. Neither justification she advances provides a sufficient basis for not requiring airbags in all automobiles. There is no "rational connection between the facts found and the choice made." See State Farm, 463 U.S. at 42, 103 S.Ct. at 2866 (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 246, 9 L.Ed.2d 207 (1962)). Thus, the Secretary's decision should be vacated as arbitrary and capricious.

C

Finally, New York claims that the Secretary acted arbitrarily and capriciously in failing altogether to consider, let alone adopt, the alternative of requiring both airbags and non-detachable automatic belts. The majority makes a specious argument that the Secretary did consider this alternative and then concludes that she acted appropriately in rejecting the option. I find that the Secretary "entirely failed to consider [this] important aspect of the problem." See State Farm, 463 U.S. at 43, 103 S.Ct. at 2867.

The majority states that the Secretary's discussion of the weaknesses of non-detachable belts was intended to both show that detachable belts should be a means of compliance under the Rule and respond to the argument that non-detachable belts should be required together with airbags. The majority then sketches its three reasons for this conclusion. Maj. Op. at 489.

First, the majority opines that individual weaknesses of airbags and non-detachable belts are germane to the desirability of requiring both. This ignores the possibility that the individual weaknesses might be alleviated by requiring both, in tandem. Specifically, research noted in the record has found that airbags function very well in non-catastrophic, frontal collisions up to speeds of 45 miles per hour, but are less effective in side or angle impacts, rollovers, and catastrophic frontal crashes. 49 Fed.Reg. at 28,98 6. The Department thus concluded, "the most effective system is an airbag plus a lap and shoulder belt. To obtain maximum protection in not only frontal, but also side and roll over accidents, occupants of cars with airbags and lap belts must use a lap belt to supplement the airbag." Id. In these types of collisions seatbelts would fill in for the low effectiveness of airbags. Even at the 12.5 percentage usage estimated for manual belts, the Department postulated that many more lives would be saved by a combination of airbags and belts than by any one alternative means used alone. See id. at 28,986, Table 5; page 13 supra. The record, otherwise replete with estimates, data and analysis, does not once attempt to analyze the consequences of combining airbags with non-detachable belts, where usage, according to the record, is at least 80 percent.

The second reason postulated by the majority is that the Secretary expressly invoked her earlier discussion of various alternatives, including requiring airbags for all cars. The majority erroneously believes this is responsive to the claim that both airbags and non-detachable belts should be required. The possibility that the relevant data would change when options are considered together or in groups was never examined by the Secretary.

Third, the majority claims that the section's title "Airbags and/or Non-Detachable Seatbelt," indicates an intent to address the option of requiring airbags and non-detachable seatbelts. Inferring intent from two words in the section title not only seems tenuous at best, but incorrectly assumes that intent is as good as act. If the Secretary failed to address this option, even if she intended to, she cannot be said to have considered the option. The plain words within the section belie the majority's interpretation: "The rationals [sic] provided in the preceding sections for adopting the new rule and for not retaining the old rule or amending it to require airbags in all cars essentialy [sic] provides [sic] the basis for the Department's decision not to amend the old rule to require either airbags or nondetachable belts or just nondetachable belts." 49 Fed.Reg. at 29,002 (referring to the agency's decision to give automobile manufacturers an option rather than mandate any one method). The broken grammar, spelling and syntax suggest the lack of consideration that was given to the whole subject of airbags. There is not a word about the option of airbags plus non-detachable automatic belts.

The Final Rule is not merely slightly unclear on the issue, as the majority would have us believe. It is silent. Given the Secretary's lack of consideration, the majority's enigmatic statement as to the reasonableness of her "decision" is extravagant praise for non-performance.

III

The subject of how to enhance automobile safety remains the controversial subject it has always been. Congress has wrestled with it in numerous contentious and agonizing battles. No statutory proposal has ever commanded universal enthusiasm, and the statutes on the books have been criticized for what they contain and for what they lack. It is not for the agency or the courts to settle these basic policy disputes. Rather, both branches of government have an obligation to apply and enforce the mandates that Congress has laid down. Because, as New York contends, the agency failed to do so, and because my colleagues have given these agency failures a pass, I dissent.

1 "Passive protection" refers to technologies that require no affirmative conduct by the automobile occupant. Examples are: airbags, which are deflated bags stored under the dashboard or in the steering wheel of a car that inflate in front of the occupant very rapidly when a car suddenly decelerates; passive interiors, additional padding and changes to such items as the steering column designed to make a car's interior sufficiently safe even without safety belts or airbags; and automatic safety belts, which move into place automatically when the passenger sits in a seat and closes the door. Automatic belts can be detachable, meaning that they have an easily reachable release mechanism that can be used to disconnect the belt and leave it that way permanently, or nondetachable, meaning that they can only be disconnected by disabling them completely, such as by cutting the belt itself

2 Petitioners in this case are State Farm Mutual Automobile Insurance Co.; the National Association of Independent Insurers; Nationwide Mutual Insurance Co.; the American Insurance Association; the National Association of Insurance Commissioners; the State of New York, its Department of Insurance, its Attorney General, and its Superintendent of Insurance; and two individuals, Kent Mason and Patricia Warren

3 This history is detailed in State Farm Mut. Auto. Ins. Co. v. Department of Transp., 680 F.2d 206, 210-12 (D.C.Cir.1982), vacated sub nom. Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)

4 The notice of proposed rulemaking suggested several possible courses of action, including retaining Modified Standard 208 but setting a new compliance schedule; amending Modified Standard 208 to require airbags only, or airbags or non-detachable belts only; rescinding Modified Standard 208; conducting a voluntary demonstration program by automobile manufacturers to gather more data; and seeking mandatory state safety belt usage laws. 48 Fed.Reg. 48,622

5 This notice sought comments regarding public acceptance of automatic restraints, usage rates and effectiveness of the various restraint systems, and the benefits to be derived from the various alternatives. The Department also sought comments on the following automobile occupant protection alternatives: requiring automatic restraints with a waiver for States with mandatory safety belt usage laws (MULs); requiring automatic restraints except if three-fourths of the States enact MULs; requiring a demonstration program by automobile manufacturers; and requiring airbags for the driver's side of small cars only. 49 Fed.Reg. 20,460 (1984). This comment period closed on June 13, 1984

6 These conditions are: (1) that the MUL require each out-board (non-center front seat) occupant of a passenger car that that was required by federal regulation, when manufactured, to be equipped with front seat occupant restraints to have those restraints properly fastened about their bodies at all times while the vehicle is in forward motion; (2) a prohibition of waivers from the MUL except for medical reasons; (3) an effective date of not later than September 1, 1989; and (4) an enforcement program that includes (a) a minimum penalty of $25, with a separate penalty imposed for each person violating the law, (b) a civil litigation penalty providing that the violation of a MUL may be used in mitigating any damages sought by a person injured in an accident, (c) the establishment of a program to encourage compliance with the MUL, and (d) the establishment of a MUL evaluation program by the State to provide information to the Department regarding its MUL. 49 C.F.R. Sec. 571.208.S4.1.5.2 (1984)

7 The compliance schedule in the absence of rescission is as follows: ten percent of all automobiles manufactured after September 1, 1986; twenty-five percent after September 1, 1987; forty percent after September 1, 1988; and one hundred percent after September 1, 1989. 49 Fed.Reg. at 28,963

8 This court has recently stated that "[t]he point at which constitutional constraint fades into persuasive practicalities is difficult to discern and unnecessary to identify." Action Alliance of Senior Citizens v. Heckler, 789 F.2d 931, 940 (D.C.Cir.1986). The same result obtains whether the court bases its finding of unripeness on constitutional or prudential considerations

9 Petitioners advance three grounds for this assertion: (1) the Secretary failed adequately to consider the alternative of permitting both MULs and a federal passive restraint standard, see State Farm Brief at 33-36; NAIC Brief at 11; New York Brief at 25-27; (2) the Secretary has provided insufficient justification for her view that MULs in States covering two-thirds of the U.S. population would provide safety benefits equal to or greater than a passive restraint standard, see State Farm Brief at 36-45; NAIC Brief at 11; New York Brief at 24-25; and (3) that the provision's waiver rule lacks a rational basis, see New York Brief at 27-29

10 One of the petitioners also contends that this case is fit for review because "Congress has explicitly indicated that final automotive safety standards ... be reviewed within 60 days of their promulgation. 15 U.S.C. Sec. 1394(a)(1)." NAIC Brief at 3. The petitioner, however, has misconstrued the statute. That measure does not speak to courts at all; instead it requires anyone who seeks to challenge the Secretary's standard to file a petition for review within 60 days of promulgation

11 Indeed, petitioners assert that since the Secretary promulgated the Final Rule, the trend in state legislatures is to pass MULs that deliberately fall short of the Secretary's requirements so that the population of those States will not be counted toward the Secretary's numerical requirement. See State Farm Supplemental Brief at 9

12 Aside from New York, States covering approximately 52 percent of the U.S. population have passed apparently noncomplying MULs since the Final Rule was promulgated. New York, which accounts for approximately 8 percent of the population, passed a MUL prior to August 1, 1984. New York's MUL appears to "substantially comply" with the Secretary's requirements, thus it is quite possible that New York's population will be counted towards the two-thirds population requirement under the waiver provision in the Final Rule. Assuming New York's MUL is counted, States covering at least an additional 59 percent of the U.S. population will still have to pass complying MULs between now and 1989

13 Petitioners also seem concerned that the Secretary might bend the requirements of the Final Rule and count noncomplying MULs, passed after promulgation of the rule and therefore not subject to the waiver provision, toward the two-thirds population requirement as long as they are in substantial compliance with the Secretary's MUL standards. Should the Secretary take such action, she would, of course, be changing the rules in the middle of the game. She has not yet indicated any intention of doing so. If she were to alter the rules in the manner feared by petitioners, however, review on that ground as well would, of course, be available at that time

14 Moreover, it is questionable whether the short-run incentives for development of non-belt passive restraints created by the Secretary's rule (giving extra credit for non-belt restraint systems installed during the phase-in period), see 49 Fed.Reg. at 29,000, will be seriously diluted by the risk that the standard will be rescinded sometime before the end of the phase-in period. As already noted, it now seems most unlikely that the Secretary will have occasion to invoke the rescission provision at all; in addition, the federal requirement of passive restraints goes into effect for ten percent of the fleet coming off the assembly line in the next few months. Rescission is plainly not imminent; what is imminent is the installation, pursuant to federal mandate, of passive restraints

15 Petitioners' position seems in this respect similar to the alleged hardship which we rejected as a basis for ripeness in Midwestern Gas. There, importers of Canadian natural gas challenged an agency's conditional authorization allowing other importers to import Canadian gas through part of the Alaska pipeline. The importers alleged that, if Canadian reserves dwindled, Canada might perceive the U.S. agency's action as a signal that this Nation "favored" the Alaska pipeline; in consequence, the theory went, the Canadian energy authorities might choose to grant import authorizations to "favored" importers whose purchases came through that pipeline. This claim was deemed insufficient to merit adjudication at that time largely because the asserted injury was grounded on speculation about future events and future behavior of third parties. See 589 F.2d at 622-23. So too here

16 Petitioners allege that, in formulating its mandatory seatbelt legislation, the State of New Jersey consciously decided to employ a $20 fine for non-compliance, rather than a higher fine, in order to avoid having its law count toward the Secretary's requirement. New Jersey, however, is not a party to this litigation

17 But even were we to overlook this threshold hurdle and allow the insurance companies and commissioners to assert the interests of the States, we would still discern no hardship of sufficient concreteness. The dilemma of the non-present States as perceived by petitioners seems to us illusory. States can seemingly avoid having their MULs "count" toward rescission of the federal standard by the expedient of including in their MULs a provision that automatically nullifies the measure as soon as it is counted toward the Secretary's two-thirds requirement. Indeed, a number of States have incorporated precisely such provisions in their MULs. See State Farm Reply Brief at 7-10 app. We of course assume the obvious, namely that States would prefer not to insert a provision that poses a risk, however remote, that a much desired state law might be rescinded. But the perceived necessity of including such a provision does not seem to us a sufficiently immediate or onerous hardship, especially when it now appears that a rescission is unlikely. And, if rescission does appear imminent, the courthouse doors are open for a renewed challenge to the "trap door" provision

Even putting to one side the availability of an automatic "sunset" provision, the States' asserted dilemma would still be quite unlike that faced by the pharmaceutical companies in Abbott Laboratories. There, as we previously alluded to, the pharmaceutical companies were, for all practical purposes, coerced into a particular course of conduct--changing their labels and advertisements--by the prospect of civil and criminal penalties that might be visited upon them. See 387 U.S. at 152-53, 87 S.Ct. at 1517-18. Primary conduct was clearly being affected. Here, by contrast, the States fear that unless they take a particular course of action--passing MULs that are less stringent than they would otherwise prefer--the agency will withdraw federal regulation which the States believe to be of benefit. That is not the concrete effect on primary conduct that ripeness doctrine demands. The remote possibility that such a benefit will be withdrawn is, in our view, a considerably less substantial sort of "hardship" than the possibility that civil and criminal penalties will be imposed.

The hardship to the States asserted by petitioners also seems much less onerous than that asserted in Pacific Gas and Elec. Co. v. State Energy Resources Conservation and Dev. Comm'n, 461 U.S. 190, 103 S.Ct. 1713, 75 L.Ed.2d 752 (1983), relied upon by NAIC. There, electric utilities contemplating construction of nuclear plants challenged as preempted by federal law a California statute imposing a moratorium on nuclear plant construction until the State determined that adequate facilities had been developed nationally for the permanent disposal of nuclear wastes. The utilities asserted that, in light of the long lead time required before constructing a nuclear facility, they needed a judicial decision immediately; otherwise, they would be forced either to abandon all nuclear energy development or run the risk that planning and development expenses would be incurred for naught. See id. at 201-02, 103 S.Ct. at 1720-21. In either case, the utilities would suffer serious losses. In view of those circumstances, the Court concluded, "[t]o require the industry to proceed without knowing whether the moratorium is valid would impose a palpable and considerable hardship on the utilities, and may ultimately work harm on the citizens of California." Id. Here, by contrast, none of the petitioners alleges that the rescission provision of the Final Rule has any immediate financial effects on them.

18 In addition, the availability of an alternative state statute that would be rescinded if "counted" for federal recission purposes detracts significantly from the force of any effects on the insurance industry's (and insurance regulators') lobbying activities in the various state capitals

19 In so holding, we do not pass upon the underlying question whether an alleged effect upon an ongoing legislative process is sufficiently concrete in nature to satisfy the considerations that ripeness doctrine sets before us. The judicial manageability of such an inquiry is not immediately evident to us, but we need not and do not opine one way or the other on the subject

20 Our decision does not create a "catch 22" for litigants seeking review of agency action under a statutory provision that requires them to petition the court within a short period after the agency's action becomes final. As we have previously stated, "it is the duty of the court to make the prudential judgment whether a challenge to agency action is ripe; it is the responsibility of petitioners to file for review within the period set by Congress." Eagle-Picher, 759 F.2d at 912

21 The dissent argues that a portion of the New York challenge found unripe by the court is in fact ripe, namely that "the Secretary arbitrarily failed to consider the alternative of permitting both mandatory usage laws (MULs) and a federal passive restraint standard." Dissent at 490. But this contention is, upon analysis, simply one part of the broader attack on the "trapdoor" provision itself as being arbitrary and capricious. The Secretary's "permitting" state-adopted MULs (plus requiring passive restraints) is just another way of describing the elimination of the Secretary's "trapdoor." A careful scrutiny of the various petitioners' claims in this respect confirms our reading. See, e.g., State Farm Brief at 29-36. This should come as no surprise, since the situation of the Secretary's "permitting" both mandatory usage laws and imposing a federal passive restraint standard is in fact the situation that now obtains. State MULs (and the passive restraint requirement) will be "permitted" until such time, if ever, that the requisite population coverage is met, a condition which we have already determined to be unlikely to occur. It is the contingent (and indeed remote) nature of this desired state of affairs coming to an end that renders this contention unripe for review

22 The standard of review in this case is determined by the Safety Act. Section 103(b) of the Act, 15 U.S.C. Sec. 1392(b) (1982), provides that "all orders establishing, amending, or revoking a Federal motor vehicle safety standard" shall be promulgated under the informal rulemaking procedures of the Administrative Procedure Act, 5 U.S.C. Sec. 553 (1982). The APA, in turn, provides that, insofar as the present bases of attack are concerned, the agency's action may be set aside by a reviewing court only if found to be "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." Id. Sec. 706(2)(A)

23 New York also contends that the Secretary's failure to eliminate detachable seatbelts as an acceptable automatic restraint violates the Safety Act. The argument appears to be as follows: The Act directs the Secretary to establish motor vehicle standards that "shall meet the need for motor vehicle safety." 15 U.S.C. Secs. 1391(2), 1392(a). "Motor vehicle safety," moreover, is defined as protection against "unreasonable risk." Id. Sec. 1391(1). Because the usage rate with non-detachable belts is higher than that for detachable automatic belts, DOT is required to eliminate the detachable belt option; to do otherwise would permit "unreasonable" risk. See New York Brief at 22-23. Even assuming the accuracy of New York's assertion about the effectiveness of non-detachable belts, we cannot agree with this argument. The Safety Act does not require the Secretary to adopt the technological alternative providing the greatest degree of safety. The Act expressly permits the Secretary to consider such factors as reasonableness and practicality in addition to safety features. See 15 U.S.C. Sec. 1392(f)(3). Both the Supreme Court and this court, moreover, have recognized the Secretary's authority to consider such factors as cost and public acceptance. See State Farm, 463 U.S. at 54, 103 S.Ct. at 2872; Pacific Legal Foundation, 593 F.2d at 1345

24 We also note that an earlier passive restraint standard permitting compliance by means of a detachable belt was upheld by this court in Pacific Legal Foundation

25 The dissent suggests that the States do not need to be encouraged to pass MULs. Dissent at 495. That view, however, is not buttressed by any facts. Indeed, it would appear that the passage of MULs has in fact come about in response to the Secretary's rule. Supp. Brief for Respondent at 5

26 The dissent argues that since airbags are rarely "inadvertently deployed," consumers will gladly pay to "replace the devices that saved the car's occupants from death or serious injury." Dissent at 496. The dissent's analysis leaves no room for the category of airbags that are deployed by virtue of sudden deceleration, without a crash, or those deployed in a minor accident for which seat belts would have sufficed. While those saved from serious injury by virtue of deployed bags may be willing to incur the replacement cost, drivers not fitting into that category may resist paying $800 to replace inadvertently deployed bags

More broadly, it should not go unnoticed that the dissent goes farther in its attack on the Secretary's rule than the insurance industry itself. Unlike the dissent, State Farm does not contend that the Secretary's substantive rule on passive restraints is unduly lenient.

27 The dissent argues that the Secretary improperly focused on costs, not safety. Dissent at 496. However, as the dissent concedes, the Secretary concluded that the replacement cost of airbags would likely deter consumers from replacing them. If the airbags are not replaced, the Secretary feared, then the occupants of those cars would be left unprotected. Dissent at 495. In our view, the crux of the Secretary's concern in this respect was safety, not cost per se

28 The dissent contends that the individual weaknesses of airbags and non-detachable belts might be alleviated by requiring both. Dissent at 497. While a dual requirement might maximize protection, it would obviously do nothing to relieve the Secretary's concerns about the primary drawbacks of the devices, namely cost and public acceptability