Opinion

Center for Auto Safety v. Elizabeth H. Dole, Secretary, Department of Transportation

  • 828 F.2d 799
  • 95 A.L.R. Fed. 735
  • 264 U.S. App. D.C. 219
  • 1987 U.S. App. LEXIS 12127
Court
Court of Appeals for the D.C. Circuit
Filed
Sep 8, 1987
Status
Published
Author
Bork
On the bench
Wald, Mikva, Bork
Cited by
22 cases
Authority
More cited than 25.5%

Vacated on other grounds by Center for Auto Safety v. Elizabeth H. Dole, Secretary, Department of Transportation, 846 F.2d 1532 (1988)

finding “judicially manageable” a National Highway Transportation Safety Administration regulation providing that the “agency will grant a petition if the agency finds that there is a ‘reasonable possibility’ of a safety-related defect in the manufacturers’ cars”

How later courts described this case

  • finding “judicially manageable” a National Highway Transportation Safety Administration regulation providing that the “agency will grant a petition if the agency finds that there is a ‘reasonable possibility’ of a safety-related defect in the manufacturers’ cars”
  • holding that “[t]he APA does not permit a reviewing court to decide which side of the factual dispute it thinks “the preponderance of the evidence” in the administrative record falls on”
  • "the court must accord the agency the appropriate degree of deference on such a record-based finding involving scientific and technological data"

Written by the judges who cited it.

The opinion

BORK, Circuit Judge,

dissenting:

In 1977, the National Highway Traffic Safety Administration (“NHTSA”), an arm of the Department of Transportation, opened a lengthy investigation of possible safety defects in certain Ford Motor Company automobiles and trucks. The Secretary of Transportation settled the matter with Ford in 1980 and NHTSA closed its investigation in 1981. This court upheld the settlement agreement. Center for Auto Safety v. Lewis, 685 F.2d 656 (D.C.Cir.1982). A petition to reopen the investigation was denied in 1981 and a petition to reconsider the denial, submitted by the Center for Auto Safety, was denied in 1982. In 1985, some of the same parties that challenged the 1980 settlement and petitioned for reconsideration in 1982 again petitioned NHTSA to open a new investigation on the ground that accidents involving these same vehicles continued to occur. The agency obtained additional comprehensive information from Ford and the other three major American automobile manufacturers. The Administrator of NHTSA then denied the petition and set out her reasons.

Appellants brought suit in the district court asking that the Administrator’s denial of the petition be set aside and that the Secretary and the Administrator be ordered to initiate the investigation sought. The district court granted defendants’ motion for summary judgment on alternative grounds: that the denial of the petition was not judicially reviewable or that review must be limited to the agency’s statement of reasons, which were upheld as a “reasoned analysis” and not contrary to law.

*816 A majority of this court reverses the judgment below on both grounds, holding that judicial review is available and that the district court should review the Administrator’s decision not on her statement of reasons but on the underlying facts. I would affirm. I think there are strong indications that Congress intended to preclude review of denials of citizen petitions, to avoid imposing the significant burden on the agency of having to litigate such claims whenever they might arise. The point is arguable, however, and I do not rest on it because, even assuming review is not precluded, I think Congress intended to limit the scope of judicial review to an examination of the statement of reasons for denying the petition.

I.

It seems to me likely that no judicial review may be had of the Administrator’s decision not to grant appellant’s petition for an investigation. Since the point may be important if this case receives further review or a similar case arises in the future, I will set out the argument.

Congress intended that there be no review and expressed its intent to preclude review in the language, structure, and legislative history of the Act. 1 Thus the NHTSA regulations said to limit the Secretary’s discretion cannot provide review. The decision not to grant a petition is, moreover, an enforcement decision of the sort that is presumptively not reviewable.

A.

The Administrative Procedure Act declares that a person “adversely affected or aggrieved by agency action within the meaning of a relevant statute, is entitled to judicial review thereof,” 5 U.S.C. § 702 (1982), unless “statutes preclude judicial review” or “agency action is committed to agency discretion by law.” Id. § 701(a)(1) & (2). This language raises a presumption of reviewability that is defeated if a court discerns either that Congress intended to preclude review or that Congress provided no “law to apply” in assessing the legality of the agency action. See, e.g., Block v. Community Nutrition Inst., 467 U.S. 340 , 104 S.Ct. 2450 , 81 L.Ed.2d 270 (1984); Heckler v. Chaney, 470 U.S. 821 , 105 S.Ct. 1649 , 84 L.Ed.2d 714 (1985). Under the first branch of the inquiry — whether the statute precludes judicial review — the Supreme Court has said that in order to defeat the APA’s presumption of reviewability, Congress’ intent to foreclose review must be shown by “clear and convincing evidence,” Abbott Laboratories v. Gardner, 387 U.S. 136, 141 , 87 S.Ct. 1507, 1511 , 18 L.Ed.2d 681 (1967), or at least must be “ ‘fairly discernible in the statutory scheme.’ ” Community Nutrition Inst., 467 U.S. at 351 , 104 S.Ct. at 2456 (quoting Data Processing Serv. v. Camp, 397 U.S. 150, 157 , 90 S.Ct. 827, 831 , 25 L.Ed.2d 184 (1970)). Congress need not explicitly state that review is precluded; instead, the evidence of Congress’ intent can be found by examining the language, structure, and background of the statute itself. See, e.g., Community Nutrition Inst.; Southern Ry. v. Seaboard Allied Milling Corp., 442 U.S. 444 , 99 S.Ct. 2388 , 60 L.Ed.2d 1017 (1979); Morris v. Gressette, 432 U.S. 491 , 97 S.Ct. 2411 , 53 L.Ed.2d 506 (1977); see also Doe v. Casey, 796 F.2d 1508, 1514-15 (D.C.Cir.1986), cert. granted sub nom. Webster v. Doe, — U.S. -, 107 S.Ct. 3182 , 96 L.Ed.2d 671 (1987).

Here I think there are strong indications of Congress’ intent to preclude review. The National Traffic and Motor Vehicle Safety Act of 1966, 15 U.S.C. §§ 1381-1431 (1982), provides the Secretary and, through her, the Administrator, with complete and untrammeled authority to decide whether or not to grant a petition seeking an investigation. Section 1410a(a) states that “[a]ny interested person may file with the Secretary a petition requesting [her]” to commence an investigation (emphasis add *817 ed). Section 1410a(c) states that “[t]he Secretary may hold a public hearing or may conduct such investigation or proceeding as [s]he deems appropriate in order to determine whether or not such petition should be granted” (emphasis added). Absolutely no criterion or standard is specified to control or even to influence the Secretary’s determination. Section 1410a(d) states that within 120 days, “the Secretary shall either grant or deny the petition.” No statutory language could be written that more completely leaves the decision whether to investigate to the Secretary. And, since the form and scope of any preliminary inquiry concerning the admissibility of a defect investigation is left to the Secretary, she has complete authority to determine what evidence to consider, whether to create an administrative record, and, if so, what sort. Her sole obligation is to “publish in the Federal Register [Her] reasons for such denial.” 15 U.S.C. § 1410a(d) (1982). If this were all, we might be obliged to conclude that Congress merely conferred a discretion the Secretary could remove by regulation, providing “law to apply,” and thus herself providing judicial review. The clear language of section 1410 is by no means all, however.

The entire structure of the Act also suggests Congress’ intent to foreclose review. Judicial review was explicitly provided for stages in the regulatory process both before and after the petition stage. Thus, when the Secretary initially establishes a motor vehicle safety standard pursuant to 15 U.S.C. § 1392 (1982), judicial review of that order is provided by section 1394. Similarly, when the Secretary wishes to enforce a compliance order directed to a motor vehicle manufacturer, she must apply to a court pursuant to section 1399(a). Significantly, no judicial review is provided for the Secretary’s decision to grant or deny citizen petitions to establish a standard or issue a compliance order. Moreover, a variety of judicial remedies were provided to enforce the Act’s provisions, including the grant of injunctive relief, id. § 1399(a), the creation of private civil actions, id. §§ 1400(b) & 1415(a), and the imposition of civil penalties, id. § 1398(a), but none whatever was created to compel the Secretary to grant specific citizen petitions. The structure of the Act thus shows that Congress carefully provided judicial review in specified circumstances, but deliberately withheld review in the circumstances of this case. Congress made available an array of judicial remedies for various purposes, but none for the purpose asserted in this case, and Congress established enforcement processes by explicitly differentiating between mandatory actions and completely discretionary actions. This case presents us with a completely discretionary action that is not made subject to judicial review. This statutory structure is weighty evidence of Congress’ intent to preclude review here.

The Supreme Court’s approach to preclusion in Community Nutrition Institute emphatically demonstrates the propriety of inferring Congress’ intent to preclude review from the structure of the governing statute. There the Court considered whether Congress intended consumers to be able to challenge certain milk-market orders. Although the statutory language did not affirmatively foreclose review, the Court decided, unanimously, that no review was permitted. “Whether and to what extent a particular statute precludes judicial review is determined not only from its express language, but also from the structure of the statutory scheme, its objectives, its legislative history, and the nature of the administrative action involved.” 467 U.S. at 345 , 104 S.Ct. at 2453-54 . The Court based its holding on its detailed analysis of the structure of administrative and judicial remedies that were both provided and omitted in the statutes governing the marketing of agricultural products. Id. at 345-48 , 104 S.Ct. at 2453-55 . As we have seen, the same analysis indicates that Congress intended to preclude review in this case.

The legislative history of the Act also reveals evidence of Congress’ intent to preclude review. The original bill would have allowed citizen petitions and provided that when the Secretary denied a citizen petition, the citizen would be entitled to obtain full review of that decision in federal dis *818 trict court. See H.R. 5529, 93d Cong., 1st Sess. § 7 (1973). Although citizen petitions were kept in the statute that was enacted by Congress, this provision for judicial review of the Secretary’s denial of those petitions was deleted in committee. One of the original sponsors objected to the deletion, and stated that he would offer an amendment to assure the availability of these “citizens’ suits.” See 120 Cong. Rec. 27,-807-08 (1974) (Rep. Eckhardt). He never did so, however, and the bill as enacted did not make the denial of these petitions subject to review. In addition, a letter from the Secretary to the Chairman of the House Committee on Interstate and Foreign Commerce — the committee that deleted the review provision — presented the Department’s views on the bill after it had emerged from the committee. The Secretary stated that he was “very pleased” that the committee had adopted the “critical” change of amending the bill “to delete provisions that would have authorized use of the courts to compel, without consideration of such factors as available technology and resources, a reordering of the priorities we have established to carry out the purposes of the Act.” This letter was included in the committee report that was available to the legislators who voted on the bill. See H.R.Rep. No. 1191, 93d Cong., 2d Sess. 34, reprinted in 1974 U.S. Code Cong. & Admin. News 6046, 6079.

The majority argues that the legislative history, in this particular, shows Congress’ intent to leave citizen petitions subject to the usual forms of judicial review under the APA. But the argument is deficient in two respects. First, it fails to take account of the Secretary’s interpretation of what Congress intended when it deleted the provision for citizen suits, which supports complete preclusion of review, and which was itself incorporated by Congress into the legislative history of the Act. Second, as the majority concedes, the strongest piece of evidence to the contrary is one passage in the Conference Report, which states that the conferees “decided to take no position on whether or not pre-enforcement judicial review is available under the conference substitute. The conferees decided to leave that question to the courts.” H.R. Conf. Rep. No. 1452, 93d Cong., 2d Sess. 32, reprinted in 1974 U.S. Code Cong. & Admin. News 6084, 6095. Yet the majority takes this passage completely out of context, and mistates its import. Here the conferees were discussing 15 U.S.C. § 1415 (1982), which concerns manufacturer suits to prevent enforcement of a defect order against it, not citizen petitions under section 1410a, which are discussed in a completely different part of the Conference Report. Compare 1974 U.S. Code Cong. & Admin. News at 6093-96 with id. at 6106 . The Report makes it absolutely clear that the “pre-enforcement judicial review” discussed here refers only to manufacturer attempts to seek review “of the Secretary’s determination of defect or failure to comply [before] the enforcement stage of the administrative process.” Id. at 6093 .

Each of these increments in the legislative history is simply one element in the complete picture, but when they all are added to the very strong language and structure of the Act, I would conclude that there is “clear and convincing evidence” that Congress intended to preclude review of the Secretary’s denial of a petition submitted under section 1410.

B.

The majority does not accept the view I have just set forth, and holds that Congress did not preclude judicial review, but merely failed to provide “law to apply” to assess the legality of the agency action. The majority in this case holds that when Congress fails to provide “law to apply” to an agency action, the agency may render the action reviewable by adopting rules that provide judicially manageable standards for assessing the legality of the agency action, thus settling an issue that the Supreme Court had left open. See Chaney, 470 U.S. at 836 , 105 S.Ct. at 1658 ; see also California Human Dev. Corp. v. Brock, 762 F.2d 1044 , 1048 n. 28 (D.C.Cir.1985) (agency rules may provide the requisite “law to apply”); id. at 1053 (Scalia, J., concurring) (same). The denial of the petition in this case is thus found to be subject *819 to judicial review based on the “reasonable possibility” standard for granting or denying petitions that NHTSA has set out in its own rules. 49 C.F.R. § 552.8 (1986).

It is settled law, however, that “an agency' cannot create through its implementing regulations a right of review withheld by the underlying statute.” Harrison v. Bowen, 815 F.2d 1505, 1517 (D.C.Cir.1987); see also id. at 1517 n. 27 (collecting cases). If Congress goes further than merely omitting any “law to apply” to the agency action, and affirmatively indicates its intent to preclude judicial review altogether of certain actions, then the content of the agency’s own rules are unavailing in face of congressional intent. That is the situation before us. There is ample evidence that the Act withholds any right to obtain review of the Secretary’s decision to, deny the petition at issue here. The agency’s own regulations cannot therefore provide review, for if they did they would be invalid: they would be “in excess of statutory jurisdiction, authority, or limitations.” 5 U.S.C. § 706 (2)(C) (1982). 2 Thus, we have no authority to take review under the regulations upon which the majority relies.

C.

Even if the evidence in the Act were thought to be too weak to defeat the APA’s normal presumption of reviewability, we should not find that we are authorized to undertake judicial review in this case. The Secretary’s denial of a citizen petition is a “refusal to take requested enforcement action,” and the Supreme Court has said flatly that “in that situation we think the presumption is that judicial review is not available.” Chaney, 470 U.S. at 831 , 105 S.Ct. at 1656 . The Court found that these refusals to take action are generally unsuitable for judicial review for several reasons: the agency must balance many variables in ordering its priorities; a refusal to act is not an exercise of the state’s coercive power over individuals that would infringe on areas normally requiring the courts’ protection; and the refusal to institute proceedings is very much like an exercise of prosecutorial discretion, which has been traditionally understood to be unreviewable. Id. at 831-32 , 105 S.Ct. at 1655-56 . Indeed, after recounting these reasons, the Court stated that the normal presumption of reviewability is not only inapplicable in these situations, but it must be reversed: “an agency’s decision not to take enforcement action should be presumed immune from judicial review.” Id. at 832 , 105 S.Ct. at 1656 .

The majority claims that Chaney cannot be read to suggest, as a general matter, that under section 701(a) a presumption of unreviewability applies to an agency’s decision not to take enforcement action. The majority points out that Chaney was decided under section 701(a)(2) rather than section 701(a)(1), and rightly notes the Supreme Court's statement in Chaney that its prior case law “clearly separates the exception provided by § (a)(1) from the § (a)(2) exception.” 470 U.S. at 830 , 105 S.Ct. at 1655 . The Court said this, however, simply in order to explain how it parsed the language of section 701(a) as a whole. It did *820 not even begin to suggest that the two sections have nothing in common.

One of the things that the two sections have in common is the general presumption under the APA, as at the common law, that agency action is made subject to judicial review. See, e.g., Abbott Laboratories, 387 U.S. at 140-41 , 87 S.Ct. at 1510-11 . Nobody has ever thought that this presumption is particular to specific sub-parts of the APA and not to others; it is “a general presumption that all agency decisions are reviewable under the APA.” Chaney, 470 U.S. at 826 , 105 S.Ct. at 1653 (emphasis added). Thus, although the Court explicitly held in Chaney that “an agency’s decision not to take enforcement action should be presumed immune from judicial review under § 701(a)(2)," id. at 832, 105 S.Ct. at 1656 (emphasis added), which was after all the only part of section 701(a) under direct consideration in the case, there is simply no reason why the same position would not obtain under section 701(a)(1). Indeed, the Court spoke generally: in addition to the several factors the Court enumerated that I have already mentioned, which all bear on reviewability of agency action in a general manner, it said that an agency’s decision not to take enforcement action “has traditionally been ‘committed to agency discretion,’ and we believe that the Congress enacting the APA did not intend to alter that tradition. Cf. 5 Davis § 28:5 (APA did not significantly alter the ‘common law’ of judicial review of agency action).” Chaney, 470 U.S. at 832 , 105 S.Ct. at 1656 . In thus referring to “tradition,” “the common law,” and “the APA” as a whole, the Court did not indicate that its approach to reviewing an agency’s decision not to take enforcement action would be properly confined only to one particular subpart of the APA.

Without the normal presumption of reviewability in this case, indeed with the presumption of nonreviewability of refusals to take enforcement actions, the arguments for judicial review are greatly weakened. 3 When the other factors discussed are considered as well, the arguments for judicial review virtually disappear.

II.

The majority also reverses the district court’s conclusion that if review is had in this case, it should be limited to consideration of the Secretary’s statement of reasons for denying the petition and should not extend to the underlying administrative record. I agree with the district court that, if review is available, Congress intended to define the reviewable “record” in this case as the statement of reasons for denying the petition. Even if I thought the majority to be clearly correct in assuming the power of review here, I think we would be obliged to confine our review to the statement of reasons. In the typical case under the Administrative Procedure Act the *821 court’s review extends to the full administrative record. See 5 U.S.C. § 706 (1982). In the typical case, however, the agency’s governing statute establishes the basis for judicial review. Indeed, the fact that in this case there is a question whether Congress provided for review at all makes it all the more likely that Congress did not intend a broad scope of review in any event. In the typical case, also, the court is not reviewing an agency’s refusal to initiate an investigation, certainly not a renewed investigation of a subject the agency has previously extensively investigated. This case, far from being typical of this court’s regular fare, is instead like Dunlop v. Bachowski, 421 U.S. 560 , 95 S.Ct. 1851 , 44 L.Ed.2d 377 (1975).

In Dunlop , the Supreme Court allowed review of an agency’s decision not to initiate an enforcement proceeding against a particular party, holding that the Secretary of Labor’s decision not to bring suit to set aside a union election was reviewable under the APA. But because “the statute relies upon the special knowledge and discretion of the Secretary for the determination of both the probable violation and the probable effect, clearly the reviewing court is not authorized to substitute its judgment for the decision of the Secretary not to bring suit.” 421 U.S. at 571 , 95 S.Ct. at 1860 . The Court then found that “[t]he necessity that the reviewing court refrain from substitution of its judgment for that of the Secretary thus helps define the permissible scope of review. Except in what must be the rare case, the court’s review should be confined to examination of the ‘reasons’ statement, and the determination whether the statement, without more, evinces that the Secretary’s decision is so irrational as to constitute the decision arbitrary and capricious.” Id. at 572-73 , 95 S.Ct. at 1860 . The Court was emphatic on this point: “Thus, review may not extend to cognizance ... of a complaining member’s challenges to the factual bases for the Secretary’s conclusion____ ‘If ... the court concludes ... there is a rational and defensible basis [stated in the reasons statement] for [the Secretary’s] determination, then that should be an end of this matter, for it is not the function of the court to determine whether or not the case should be brought or what its outcome would be.’ ” Id. at 573 , 95 S.Ct. at 1860 (quoting DeVito v. Shultz, 72 L.R.R.M. (BNA) 2682 , 2683 (D.D.C.1969)).

With one exception, the situation here is completely parallel to that in Dunlop . The exception is that the governing statute here provides NHTSA with even more discretion than the Labor-Management Reporting and Disclosure Act of 1959 (“LMRDA”) gave the Secretary of Labor in Dunlop . The LMRDA provided that a union member could file with the Secretary a complaint concerning a statutory violation, and “[t]he Secretary shall investigate such complaint and, if he finds probable cause to believe that a violation of this subchapter has occurred and has not been remedied, he shall ... bring a civil action against the labor organization.” 29 U.S.C. § 482 (b). Thus, the statute provided the court with law to apply and the Administrative Procedure Act allowed the courts to review the Secretary’s decision that there was not probable cause to believe that violations had occurred to determine if it was arbitrary and capricious. Here, however, the Act does not confine the Secretary of Transportation’s authority at all.

The majority gives three reasons for its belief that Dunlop does not control this case. First, the precedential force of Dun-lop is claimed to be limited to the context of the specific labor statute at issue in that case. This court’s decision in Natural Resources Defense Council, Inc. v. SEC, 606 F.2d 1031 (D.C.Cir.1979), is cited as support for that view. Second, although several other courts have applied Dunlop beyond that narrow context, the majority believes that none of them has provided a rationale that would apply in this case. Third, the majority believes that the Supreme Court’s decision in Florida Power & Light Co. v. Lorion, 470 U.S. 729 , 105 S.Ct. 1598 , 84 L.Ed.2d 643 (1985), and this court’s decision on remand, see Lorion v. United States Nuclear Regulatory Comm’n, 785 F.2d 1038 (D.C.Cir.1986), compel us to review the Secretary’s decision on a wider scope *822 than that prescribed in Dunlop . I think none of these attempts to distinguish Dun-lop succeeds.

A.

The reasoning of Dunlop was clearly not confined to the context of the LMRDA or dictated by the peculiar exigencies of union elections. The Court never once indicated that its holding was limited to cases involving union elections; instead, it offered general principles about when it is appropriate to confine judicial review to the agency’s statement of reasons, and those principles obviously apply to much more than that particular statute. Neither did this court in Natural Resources Defense Council, Inc. v. SEC, distinguish Dunlop on that ground. Instead, we stated that the principles the Court relied on in Dunlop were not present in the context of the securities laws at issue in that case. The central principles from Dunlop , as we stated them, were the “special discretion” that the labor statutes afforded the agency, and “the congressional intent evident in [those statutes] to prevent undue judicial intervention into union affairs.” Id. at 1053 n. 81.

Those same principles are present in the Act that governs NHTSA. Here the statute affords “special discretion” to the agency; it does not in any way confine the agency’s discretion whether to investigate the facts alleged in the petition and what information to consider in assessing those facts. The Act does not provide for judicial review and does not offer a basis on which judicial review can be undertaken without turning away from the statute itself to the agency’s own rules. Finally, to the extent that the holding in Dunlop may have reflected a special concern to prevent courts from unduly intervening in union affairs, we also have such special concerns here. As I have indicated, this is not the agency’s first investigation of the problem. It has already made a thorough investigation and entered into a settlement that has been reviewed and upheld by this court. See Center for Auto Safety v. Lewis. What has been brought here is a petition to reopen the investigation. Subsequent petitions, alleging that there are new facts, can be brought by anyone at any time. If the Administrator seeks additional information, as she did with respect to this petition, an administrative record will be compiled which would have to be understood in the light of prior administrative records on the same subject. Full judicial examination of the administrative record every time such petitions are brought was clearly not intended in a statute that left the consideration of those petitions entirely within the discretion of the agency.

B.

The majority concedes that a number of cases have applied the rationale of Dunlop beyond the narrow context of the labor statutes at issue there. See maj. op. at 811. There are still other cases. See, e.g., Illinois v. Nuclear Regulatory Comm’n, 591 F.2d at 12 (7th Cir.1979) (applying Dunlop in the context of the Atomic Energy Act of 1954); Levin v. Connecticut Blue Cross, Inc., 487 F.Supp. 385 (N.D.Ill. 1980) (applying Dunlop in the context of the Federal Employee Health Benefits Act); cf. Singleton v. Corp, 465 F.Supp. 14 (S.D.N.Y.1978) (applying Dunlop in a different context under the labor laws). There are also cases that have rejected Dunlop’s rationale; in addition to Duquesne Light Co. v. EPA, 522 F.2d 1186 , 1193 n. 24 (3d Cir.1975), cited by the majority, there is also an extensive criticism of that rationale in Thompson v. Department of the Treasury, 533 F.Supp. 90, 92-97 (D.Utah 1981), where the district court refused to apply Dunlop in the context of the Gun Control Act of 1968.

The narrow scope of review prescribed in Dunlop has apparently caused some judges discomfort. However that may be, we are bound by Dunlop , which makes the scope of review turn on the congressional intent underlying the governing statute. Whether or not the agency's rule provides the “law to apply,” the statute expresses Congress’ intent with respect to the degree *823 of judicial intervention in the agency’s enforcement discretion. 4

The Seventh Circuit’s opinion in Ward v. United States Parole Comm’n, 804 F.2d 64 (7th Cir.1986), for example, offers no support for the majority’s reading of Dun-lop. In Ward , Judge Easterbrook refused to dismiss Dunlop simply as a labor case, but stated its rule as follows:

[T]he Court concluded that if the Secretary’s decision not to file a suit against a union is reviewable, the review must be made on the basis of the Secretary’s statement of reasons. The narrow scope of review was compelled by the Secretary’s broad discretion and the absence of any statutory requirement to create an administrative record. Here, too, the cabinet officer has great discretion, and there is no statutory requirement to create a record.

Id. at 67 (emphasis added).

What was true of Dunlop and Ward is true here. This statute gives the agency broad discretion about whether to conduct investigations and what information to consider. The statute also does not require the agency to create an administrative record. In such a case, as in Dunlop , judicial review should be confined to review of the Secretary’s statement of reasons except, as Dunlop stated, “in what must be the rare case.” 421 U.S. at 572 , 95 S.Ct. at 1860 . It does not change matters that the agency has created a record in accordance with its own rules. The Secretary of Labor had created a record in Dunlop , as is shown by his detailed statement of reasons based upon the findings of an extensive investigation. If a fuller record had been regarded as desirable, the Court, or the district court upon remand, could have required the agency to create it. Yet the Court held that review was confined to the agency’s statement of reasons unless the statute itself has limited the agency’s discretion and required it to create an administrative record. 5

The majority notes that the precise boundaries of the Supreme Court’s holding in Dunlop have not been fleshed out. Nor have ,1 attempted to flesh them out here. I have simply attempted to apply Dunlop’s approach to the facts before us, and I have sought guidance in doing so from the lower courts’ reading of that case. 6 Yet the majority rejects the interpretations of Dunlop *824 rendered by this court in Natural Resources Defense Council, Inc. v. SEC, and by the Seventh Circuit in Ward , both of which have been sketched out previously. Instead, the majority offers its own view of the principle central to Dunlop: “Dunlop stands for the considerably narrower principle that courts will not conduct their usual review of the evidence in the administrative record underlying the agency’s decision if this kind of review would interfere with identifiable congressional objectives of the relevant statute.” Maj. op., supra at 813 (emphasis in original). I have two objections to this statement. First, it is not clear to me why the majority sees its “interference-with-identifiable-congressional objectives” principle as the entire holding of Dunlop rather than simply as one factor, certainly quite an important factor, that bears on Congress’ intent as it was embodied in the agency’s governing statute.

Second, I find it difficult to discern what is meant or achieved by putting so much stress on whether the courts can label a congressional objective as “identifiable.” The majority thus finds support for its holding in the fact that, unlike the statute in Dunlop , which sought to restrict outside interference with the Secretary of Labor’s decisions under the statute, the Act here “contemplates extensive citizen involvement in enforcement of the Act by providing for citizen petitions for enforcement of safety standards.” Maj. op., supra at 813. From this fact the majority infers that Congress was less concerned about the possible intrusiveness of judicial review of the kinds of decisions at issue in this case, so that Dunlop is not applicable. It seems to me, however, that precisely the opposite conclusion should be drawn. Although Congress allowed “any interested person” to petition the Administrator at any time to commence a proceeding to determine whether to issue an order aimed at remedying safety defects, 15 U.S.C. § 1410a(a) (1982), it certainly understood that responding to this extensive citizen involvement could impose a potentially tremendous burden on the agency. This concern, and the danger that defending against repeated citizen suits would hamper the agency in the discharge of its primary responsibilities, strikes me as having an important bearing on congressional objectives that are at least as “identifiable” as those the majority finds in the statute that was under consideration in Dunlop . 7

Thus Congress left the manner of the agency’s response to such petitions entirely within its discretion, requiring only that if the Administrator denied any such petition “[s]he shall publish in the Federal Register [her] reasons for such denial.” 15 U.S.C. § 1410a(d) (1982). These provisions, in my view, suggest Congress did not foresee that the Administrator’s decisions to deny these petitions would be subject to judicial review, but if somehow they did become subject to review, Congress intended that the scope of review should be confined to the only kind of “record” it required the agency to develop — the statement of reasons for denying the petition.

C.

As a final matter, the majority’s citation of Florida Power & Light Co. v. Lorion, *825 470 U.S. 729 , 105 S.Ct. 1598 , 84 L.Ed.2d 643 (1985), is inapposite. Although the factual setting in that case was very similar to the facts before us here, the Supreme Court focused its attention on an entirely different legal issue. Lorion concerned not the proper scope of review of agency action, but whether initial subject-matter jurisdiction to review certain orders of the Nuclear Regulatory Commission was vested in the district courts or in the courts of appeals. The Court held, as a matter of congressional intent determined from the governing statute, that jurisdiction to review such orders was vested in the appellate courts. The only real concern about this conclusion stemmed from the possibility that the appellate court (which, unlike a district court, does not make factual findings) would be forced to review the agency’s decision on an insufficient factual record. The Court addressed this problem by pointing out that often a sufficient factual record will have been developed by the agency, and in other cases the court can remand to the agency for further investigation or explanation. Id. at 743-45. Similarly, the Lorion case on remand in this circuit is not a precedent in any meaningful sense of the word, since this court never even considered, let alone rejected, the applicability of Dunlop in a situation like the one we have here.

The Supreme Court’s decision in Lorion, see 470 U.S. at 743-44 , 105 S.Ct. at 1607, merely establishes two points that have no bearing on the issue in this case. First, it restates the accepted principle laid down in the APA that in the typical case judicial review is conducted on the basis of -the entire administrative record. Second, it notes that whether such a record exists does not depend on whether the agency action was “formal” (taken after a formal hearing) or “informal.” These points, however, work no change in the principles set out in Dunlop . The accepted scope of review in the typical case does not apply under the circumstances described there. This is a matter of congressional intent as set out in the governing statute, and mere restatement of the APA’s general principle is unavailing. In Lorion, for example, no party questioned that the governing statute provided for judicial review, and there was no imputation that it provided for judicial review on anything other than the administrative record. 470 U.S. at 734-36 & n. 8,105 S.Ct. at 1602-03 & n. 8 (construing 42 U.S.C. § 2239 (b) (1982) and 28 U.S.C. § 2342 (4) (1982)). This case differs on both points. In addition, there is no suggestion in Dunlop that its limited scope of judicial review always applies to informal agency action irrespective of the existence of an administrative record. The essence of the matter is not whether an administrative record exists — for one could always be developed — but whether Congress has evinced an intent in the governing statuté to limit the scope of judicial review through such signals as granting “special discretion” to the agency, not requiring in the statute itself that the agency must'compile a record, or manifesting other reason for concern about undue interference with particular agency functions.

In Dunlop , the Court was satisfied that the Secretary’s “statement of reasons supporting his-determination” would “enable the reviewing court intelligently to review the Secretary’s determination,” as long as it “informed] the court and the [complainants] of both the grounds of decision and the essential facts upon which the Secretary’s inferences are based.” 421 U.S. at 571, 573-74 , 95 S.Ct. at 1860, 1861 . Given the context of the governing statute, a fuller administrative record was' not thought to be needed for review. But we have as much here. It is not contended that the statement of reasons provided in this case is inadequate “to enable the court to determine whether the Secretary’s decision was reached for an impermissible reason or for no reason at all.” Id. at 573 , 95 S.Ct. at 1861 . On the contrary, the statement thoroughly explains the reasons why the agency denied the petition to commence new proceedings in this matter. See Appellees’ Appendix at 35.

For these reasons, I believe we should limit the scope of our review to the statement of reasons the Administrator gave for denying the petition. That statement of reasons seems to me entirely adequate to *826 survive review on an arbitrary and capricious standard.

Since I would affirm the district court’s decision, I respectfully dissent.

. The statute itself refers not to the Administrator but to the Secretary of Transportation, who is one of the appellees in this case. See, e.g., 15 U.S.C. §§ 1410a, 1412 (1982). The Secretary is authorized to carry out the Act through the Administrator, as she did here. See 49 U.S.C. § 105 (d) (1982).

. The majority places a great deal of emphasis on the proposition that the Supreme Court set down in Service v. Dulles, 354 U.S. 363 , 77 S.Ct. 1152 , 1 L.Ed.2d 1403 (1957), that judicial oversight exists to require an agency to follow its own rules. But I have absolutely no dispute with this proposition. The difficulty, instead, is that there is a tension between the above proposition and this court’s holding in Harrison, 815 F.2d at 1517 , that an agency’s own rules cannot confer judicial review over an agency decision when Congress has precluded review of that decision. The only way to reconcile Service and Harrison is to recognize that Congress may very well empower an agency to promulgate certain rules and implement them without necessarily granting the courts the power to review the everyday operation of those rules; agencies and the courts do not travel in tandem if Congress did not intend that they do so. The majority, however, takes a different point from Service : that congressional intent to preclude review cannot be inferred simply from the fact that Congress confers discretion upon an agency. I agree with this point, but I do think that such discretion is at' least one piece of evidence that forms part of the more complete picture in discerning congressional intent, which comprises all the justifiable inferences that a court is obliged to draw from the language, structure, background, and legislative history of the agency’s governing statute.

. The Court in Chaney did find that its presumption of unreviewability "may be rebutted where the substantive statute has provided guidelines for the agency to follow in exercising its enforcement powers.” 470 U.S. at 832-33, 105 S.Ct. at 1656-57. In this case, however, Congress provided no such guidelines. Though the majority finds "law to apply” in the agency’s rule that petitions are denied when there is not a "reasonable possibility” that an investigation will establish alleged safety defects, see 49 C.F.R. § 552.8 (1986), the Chaney Court refused to find that an agency's action was made reviewable by its own rule when that rule conflicted with another rule which indicated that the action was not understood to be reviewable. 470 U.S. at 836, 105 S.Ct. at 1658. Based on that holding, the issue here is even less difficult, since here the rule relied on by the majority conflicts with Congress’ intent to preclude review as set forth in the governing statute.

In addition, it is not obvious that the rule cited by the majority is intended to provide judicially manageable standards that would allow review. The majority finds the "reasonable possibility” standard to be a "judicially manageable” one, and so I suppose it is, but that is not quite the point. The question is not whether we can manage it, but whether we should. That in turn would depend upon the intention underlying the regulation. Although the rule is said in one place to "limit[ ] the discretion of the Administrator," 40 Fed.Reg. 42,013 (1975), much of its language seems purely descriptive. Most of the language is in the present tense and conspicuously lacks any mandatory word such as "shall" or "will.” It reads much more like a discussion of how the agency works, intended for the information of outsiders, than it does like internal law. It is at least doubtful that it cabins the Administrator’s discretion sufficiently to give the courts "law to apply.”

. The majority expresses the fear that the Dun-lop exception could become “the 'Jaws’ of administrative law, gobbling up the usual APA rule that judicial review, even of informal, discretionary decisions, is based on the administrative record." Maj. op. at 812. This concern is unfounded. Congress has stated that in the typical case under the APA judicial review, is to be had on the record, see 5 U.S.C. § 706 (1982), just as in the typical case agency action is presumed to be subject to judicial review. Dunlop merely indicates that Congress retains the authority to express its intent, in the agency’s governing statute, that specific agency actions are to be reviewed on something less than the record as a whole, just as Congress retains the authority to indicate that other specific agency actions are not reviewable at all. Thus Dunlop simply recognizes, as we must, that Congress remains master of the field of administrative law, and is completely free, subject only to the broad limits imposed by the Constitution, to establish the substantive and procedural constraints under which an agency is to administer the law in its particular field, and which the courts are bound to apply on review.

. Judge Easterbrook’s reading of Dunlop turns on whether there is "any statutory requirement to create an administrative record," 804 F.2d at 67 (emphasis added), rather than whether there is any "legal duty” to create a record. This difference is crucial because it highlights our obligation to determine, according to Dunlop , whether in a particular governing statute Congress intended to confine judicial review to the agency’s statement of reasons for its action rather than providing for the usual, more extensive review based on the entire administrative record. In Dunlop , for example, the Court could have required the Secretary of Labor to develop a fuller record rather than simply providing a statement of reasons; then the agency would have been under a legal duty to do so. Indeed, the. Court could also have read 5 U.S.C. § 706 (1982), which provides for judicial review in the typical case to be undertaken on "the whole record or those parts of it cited by a party" as itself imposing a legal duty on the agency to develop a fuller record that would facilitate review. But the Court did not read this language so expansively, or impose such a requirement itself; instead it based its conclusions about the proper scope of review simply on the language and structure that Congress set out in the governing statute that was at issue.

. I therefore disavow each of the grander and erroneous views that the majority attributes to me. I do not say that a court will be confined to reviewing agency action on a statement of reasons whenever the statute confers a “special” discretion on the agency. Maj. op., supra at *824 811-812. This phrase comes instead from this court’s reading of Dunlap in Natural Resources Defense Council, Inc. v. SEC, 606 F.2d at 1053 n.31, a decision that is binding upon us, and this phrase itself forms a subordinate part of the appropriate inquiry as it is set forth in Natural Resources Defense Council, Inc. v. SEC. Nor do I mean to imply that review of nonenforcement decisions "must automatically be limited to the agency’s statement of reasons.” Maj. op., supra at 812. When a nonenforcement decision falls within the holding in Dunlop , as some such decisions surely will, then review will be limited. But I do not claim that all such decisions are embraced by Dunlop .

. Nor did Dunlop rest at all on the unpersuasive distinction the majority draws between issues requiring the agency to draw credibility determinations and complex issues of technical fact. See maj. op., supra at 813-814. Not only might the latter issues involve credibility determinations as well, but the legal premise underlying the distinction is itself faulty: appellate courts review credibility determinations all the time, albeit very deferentially, just as they also review agency determinations on complex technical issues, also very deferentially. See, e.g., Baltimore Gas & Elec. Co. v. Natural Resources Defense Council, Inc., 462 U.S. 87, 103 , 103 S.Ct. 2246, 2255 , 76 L.Ed.2d 437 (1983).

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.

A word about cookies

We need a few to keep you signed in and the library working. The rest help us see which pages people use and where they get stuck. They stay off unless you say yes.