Opinion for the Court filed by Circuit Judge STARR.
Opinion concurring in part and dissenting in part filed by Circuit Judge MIKYA.
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 chal*401lenged 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.
I
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. §§ 1381 et seq. (1982). The Safety Act was intended “to reduce traffic accidents and deaths and injuries to persons resulting from traffic accidents.” Id. § 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. § 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 *402a 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. § 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. M6 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 *403forth some general principles to guide our analysis, we then examine each issue raised by petitioners to determine whether it is ripe for review.
A
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.” 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 *404deferral 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 indeed they are onerous at 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.
1
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 *405these laws, however, apparently complies with the Secretary’s specific requirements. See State Farm Supplemental Brief at 7-10. la.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 proceedseeking a 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 un-dertested. 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 chal*406lenge 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
*407Petitioners also assert that the rescission provision produces hardship for States not contemplating MULs. According to petitioners, such in want of a 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.18
*4082
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 not 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 *409the 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
3
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. § 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.21
*410III
We turn then to the merits of the one ripe 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 sec*411ond, 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 nondetachable belts, see id. at 32.
1
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
2
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 De*412partment 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 *413point 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.
3
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 Seat-belts.” 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.
. "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 interi- or 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.
. 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.
. 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).
. 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.
. 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.
. 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. § 571.208.S4.1.5.2 (1984).
.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.
. 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.
. 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.
. 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. § 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.
. 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.
. 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.
. 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.
. Moreover, it is questionable whether the short-run incentives for development of non-belt passive restraints created by the Secretary’s *406rule (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.
. 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.
. 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.
.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 their labels and 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 *407affected. 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 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.
. 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.
. 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 *409immediately evident to us, but we need not and do not opine one way or the other on the subject.
. 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.
. 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 *410time, 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.
. The standard of review in this case is determined by the Safety Act. Section 103(b) of the Act, 15 U.S.C. § 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. § 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. § 706(2)(A).
. 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. §§ 1391(2), 1392(a). "Motor vehicle safety,” moreover, is defined as protection against "unreasonable risk.” Id. § 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 *411New 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. § 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.
. 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.
. 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.
. 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.
. 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.
. 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.