CARDOZO LAW REVIEW [Vol. 20:989 1999]


Ernest Gellhorn* and Paul Verkuil**


The central operational standards of our written Constitution — that executive, legislative, and judicial powers shall be separated and that each shall be controlled by one (or more) of the three branches of government — are based on two precepts: first, that governmental powers are limited to those granted by the Constitution or approved by Congress and the President; and second, that discretion in any agency or official is constrained by the Constitution as well as by laws duly enacted pursuant to constitutional requirements. The first restraint on government lies with Congress, under its authority to “make all laws”[1] and thus to make them specific and limited. It has full authority to ensure that an agency stays within its designated jurisdiction by clear drafting and specific assignments. To this end, a revived delegation doctrine, which requires legislation to include “intelligible principle[s]”[2] for measuring the scope and not just the goals of legislation,[3] could play a critical role in confining agency discretion and ensuring agency accountability. In this way, limits on vague and standardless delegations should be a first principle of constitutional government.[4] Strong policy grounds continue to support wider application of the doctrine prohibiting standardless delegations.[5]

Properly pursued, the delegation doctrine would ensure that major policy decisions are made by an elected Congress and President, and not an appointive bureaucracy.[6] It also would give agencies and courts greater guidance in understanding the scope (and limits) of an agency’s mandate.

Despite the attractiveness of a revived delegation doctrine in constitutional law, it also must be acknowledged that the United States Supreme Court continues to show “no interest in reviving the delegation doctrine as a check on administrative discretion.”[7] In upholding wide discretion to the United States Sentencing Commission to craft mandatory sentencing guidelines in Mistretta v. United States,[8] Justice Blackmun went so far as to say that the Court’s “approval of ... broad delegations”[9] without objection “has been driven by a practical understanding that in our increasingly complex society, ... Congress simply cannot do its job absent an ability to delegate power under broad general directives.”[10] Even Justice Scalia’s dissent in Mistretta agreed with this assessment, although he derisively characterized the Commission’s authority as a “junior-varsity Congress.”[11] He reasoned that the Court has not “felt qualified to second-guess Congress regarding the permissible degree of policy judgment that can be left to those executing or applying the law.”[12]

However, the problems of unconfined agency discretion and, more particularly, of excessive delegations of authority to agencies are not limited to the scope of congressional delegations or to the standards applied to test the necessary precision of regulatory laws. Indeed, the larger question, at least in terms of practical impact and unregulated agency overreaching, is how to control overly expansive readings by agencies of their mandates. Justices Blackmun and Scalia may be right that Congress cannot be expected to do a better job of drafting legislation that sets primary policies, although we will never really know unless it is expected or commanded by the courts to do so. However, it is surely the case that, even if the Court required Congress to adopt more precise and clearer statutes when assigning agencies responsibility for clean air, a safe and sound banking system, reasonable rules for election campaigns, or fair competition, difficult, unanticipated issues not answered directly by the terms of the statutes or their legislative histories would continue. Arguments over the scope of an agency’s authority cannot be eliminated and the standards applied to decide such issues are critical in limiting the discretion of administrative officials.

Designing more effective controls on agency interpretations of their authority involves two questions: (1) how agencies should read their mandates, and (2) how courts should review such agency interpretations. Consistent with separation of powers principles, we believe that agencies should read their enabling authority restrictively. Assertion of new-found powers should occur only after receiving a clear statement of intent from Congress.[13] Such a self-imposed standard would help ensure the legitimacy of agency action outside the traditional scope of its authority.[14]

But the tendency has been to favor a more expansive approach. Regulators are expected to be proactive. The media interprets this to mean that regulators should continually press for an expansion of their authority. Perhaps this is inevitable, although the more careful analysis today holds that regulatory programs should be confined to market failures, should be limited to solutions that match specific needs, and should apply market-based standards where possible.[15] The argument for agency restraint in interpreting its legislative authority reflects a more basic principle of legitimacy and authority. Until Congress delegates specific powers to an agency with a reasonable degree of clarity, there is no basis in law for an agency to claim such power. The checks and balances central to our constitutional scheme instruct that laws are passed only after compliance with both the bicameralism and presentment requirements, not agency preferences or interests.[16]

Such a rule of interpretation is more consistent with constitutional and prudential norms, and we urge that agencies should adopt it. But we are skeptical that this recommendation will carry the day. History clearly shows that, except in highly unusual circumstances,[17] agencies read their authority expansively and often pursue agendas far beyond that envisioned when the agencies were created.[18] These many causes include: pressure from the President or congressional committees; bureaucratic imperatives; and public (i.e., interest group) demands.[19] Neither the President nor Congress is likely to narrow agency discretion to limit such tendencies when it is in their self-interest not to do so.[20] In addition, we do not believe that courts are likely to expand the clear statement doctrine beyond its current limited scope where an agency interpretation raises serious constitutional questions.

The second approach for constraining agency assertions of authority, and the one we develop more fully in this Article, would restore the independent judicial determination of jurisdictional issues outside the agency’s primary mandate or subject matter, even if contrary to an agency’s plausible interpretation (and some readings of Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc.[21]). This approach is not without its problems, of course, but we view it as far more promising than the alternatives. Admittedly, the United States Supreme Court cases have limited the application of Chevron more often than they have applied it and the cases often are difficult to reconcile.[22] But we believe that some members of the present Court may be receptive to limiting Chevron deference to narrower questions of legislative ambiguity where it is more likely that Congress both expected agencies to fill in the gaps and courts to defer to such agency determinations.[23] In any event, the application of Chevron deference to peripheral jurisdictional questions has been and remains unsettled.[24]

Adoption of a rule denying deference to expansive agency readings of their jurisdictional claims would not require overruling any precedent. Moreover, a requirement of independent judicial evaluation of peripheral jurisdictional issues is consistent with the terms and rationale of Chevron. The prudential rule of Chevron (and Skidmore v. Swift & Co.[25]), requiring judicial deference to reasonable agency interpretations of their enabling acts, was first applied in limited contexts where the specific scope of the agency’s authority to apply well-recognized regulatory programs was in doubt. Its purpose was to take advantage of agency experience and expertise in areas where Congress had delegated general authority to the agency, but the precise contours of that power were not clear or its applications to recent developments were uncertain.

This Article presents the case for restricting the application of Chevron deference to filling in the interstices of its core legislation, thereby leaving peripheral jurisdictional issues for independent judicial determination. This is not to say that agencies cannot argue the case for greater jurisdiction, including an explanation of why their interpretation should be given special attention because of the impact of the decision on their regulatory programs. Nor does it depend upon either support of or hostility to Chevron.[26] Taking the agency’s experience and special insights into account because of the merit of its analysis is one thing; but taking agency interpretations of basic statutory authority outside traditional areas of agency regulation as controlling is another.

When agency self-interest is directly implicated, such as when it must decide whether an area previously unregulated by the agency should now come within its jurisdiction, the justifications for deference fade. Its experience and expertise in the new subject matter is limited; it is at most a fiction that Congress intended that the agency would exercise jurisdiction if it had addressed the question. It is here that concern about agency aggrandizement is at its highest. Of course, peripheral jurisdictional issues are not always readily distinguishable from other more traditional questions, and this is a major criticism of our approach. But we are satisfied that even in those circumstances, courts should be cautious in applying Chevron where there is significant distance between the agency’s traditional regulation and that being asserted. The need to ensure accountability and limit agency discretion weighs heavily on the side of not expanding Chevron beyond its original and limited scope. For our purposes, it is Chevron deference that is the problem, not Chevron itself. Moreover, if our jurisdictional concerns can be resolved in step one of Chevron, where both the agency and court are bound by Congress’ legislative judgment, then we have reached our goal without adjusting the deference standard.


The Constitution[27] and the Administrative Procedure Act[28] (“APA”) require that agency action be based on authority granted to the agency by Congress. Unlike common law courts with a recognized power to create their own authority, as well as to fill in and apply the law, it is contrary to the constitutional scheme for agencies to regulate areas beyond those which Congress authorized. These propositions are not unchallenged: They have been questioned by those who see agencies as recipients of regulatory power subject to “dynamic statutory interpretation.”[29] Indeed Professor Sunstein also postulates that “[i]n the modern era, administrative agencies have become America’s common law courts.”[30] While this argument has appeal, we necessarily reject it in this Article.[31]

Determining whether an agency’s asserted authority is within a specifically delegated assignment is subject to numerous rules and principles. Most prominent among them is, of course, step two of the Chevron doctrine that agency interpretations which are reasonable shall be upheld where Congress has not spoken on the precise question.[32] While agencies often interpret their regulatory authority expansively, such assertions generally are limited to situations where the agency clearly has jurisdiction over the practices involved. For example, when the Federal Trade Commission (“FTC”) pushed its substantive and procedural authority to the boundaries in cases such as FTC v. Sperry & Hutchinson Co.[33] and National Petroleum Refiners Ass’n v. FTC,[34] serious challenges were not raised about the FTC’s authority to prohibit antitrust and “unfair trade practice” violations. That is, the trading stamp practices at issue in Sperry & Hutchinson raised tying issues closely related to traditional antitrust prohibitions, and the use of substantive rule authority in National Petroleum involved an interpretation of the FTC’s rulemaking authority under the FTC Act. Thus, the primary question in cases such as Sperry & Hutchinson was whether, as in Chevron itself, the agency had the authority to attack the problem in the manner asserted. Did the FTC’s rulemaking authority cover substantive rules? Did its assignment of prohibiting anticompetitive and unfair practices include the lock-in competitive advantage obtained by trading stamps.

On the other hand, when an agency asserts authority far beyond its legislative assignment, as occurred in Leedom v. Kyne,[35] the courts have reviewed agency assertions of jurisdiction even before a final order is issued and without deference to the agency’s interpretation of its authority.[36] As Judge Friendly concluded in PepsiCo v. FTC,[37] Leedom allows a court to review without deference an agency complaint or other action “plainly beyond its jurisdiction as a matter of law,” even though further proceedings were contemplated, because the “proceeding ... must prove to be a nullity.”[38] However, these are pre-Chevron cases. Whether Chevron modified the Leedom rule is a question that we will seek to answer.[39]

A. Examples of Agency Action Plainly Beyond Jurisdiction

Perhaps emboldened by the role of judicial deference outlined in Chevron, agencies have increasingly ignored the boundaries of their delegated authority and made broad claims of jurisdiction into areas long thought to be outside their jurisdiction. When challenged, agencies know they can rely on the safe harbor of step two of the Chevron test by arguing that their jurisdictional conclusions are “reasonable.”[40] Illustrations of such broad claims follow. In building our model, we are aware of the difficulty of those agency actions that are marginally, rather than plainly, beyond jurisdiction. Our efforts will focus on that question in the next Part.

1. FDA Regulation of Tobacco Products[41]

Until the Food and Drug Administration (“FDA”) issued the proposed rule regulating the advertising and sale of cigarettes and smokeless tobacco in 1995,[42] other agencies exercised regulatory control of the customary marketing of tobacco products. The FDA had acknowledged that it had no jurisdiction either over controls on access to tobacco products or over advertising and marketing so long as health claims were not asserted.[43] Although tobacco was a major sector of the economy, the legislative history, debates, and the hearings leading to the Food, Drug and Cosmetic Act of 1938[44] (“FDCA”) did not mention tobacco products. The inclusion of tobacco products in the FDCA surely would have been controversial. But the Act, as passed, had the full support of tobacco-state legislators.[45] There was, in other words, no expectation in 1938 that the FDA could regulate tobacco products.

To be sure, the FDCA is to be interpreted as liberally as necessary to ensure that foods, drugs, cosmetics, and medical devices are safe for those who use them.[46] And Congress has long recognized that cigarettes and other tobacco products are dangerous to one’s health.[47]

The FDA, relying on “new evidence” that nicotine is addictive and has significant pharmacological effects and that nicotine addiction is a pediatric disease, asserted jurisdiction over the distribution, marketing, and sale of cigarettes and smokeless tobacco.[48] It emphasized the health effects of tobacco products and the dangers that marketing and advertising practices presented in enticing children to begin smoking. This need was then used to justify a broad reading of the FDCA to encompass tobacco products as combination products under the drug and device segments of the Act.[49] A federal district court accepted the FDA’s argument that the FDA had jurisdiction over tobacco products (but not their advertising and promotion) under Chevron on two grounds. First, it found that although Congress did not specifically address the issue of FDA jurisdiction over tobacco products in the FDCA, it also did not express a “clear intent to withhold from [the] FDA jurisdiction to regulate tobacco products in some place other than the text of the FDCA.”[50] Second, the court found the agency’s interpretation “reasonable”[51] and thus entitled to Chevron deference.

On appeal, the United States Court of Appeals for the Fourth Circuit reversed in a two-to-one opinion authored by Judge Widener.[52] First, the court reversed the district court’s presumption of statutory authority, calling it a “fundamental misconception.”[53] Once it took this step, it was relatively easy to retrace and reverse the jurisdictional assumptions contained in the district court’s opinion. From the Chevron perspective advocated here, the Fourth Circuit’s approach is preferable. When expanded areas of jurisdiction are under scrutiny, Chevron “deference” must itself be viewed skeptically. In this connection, the court stated: “We also note that ascertaining congressional intent is of particular importance where, as here, an agency is attempting to expand the scope of its jurisdiction.”[54]

2. Application of Civil Rights Law to Environmental Decision-Making

Decisions made by agencies such as the Environmental Protection Agency (“EPA”) and the Nuclear Regulatory Commission (“NRC”) regarding the permitting of industrial plants and the siting of nuclear facilities are required to be made in accordance with statutory criteria related to environmental pollution, safety, and efficacy. For example, the NRC, created in 1974[55] and responsible for the “[p]rincipal licensing and regulation” of nuclear facilities,[56] must consider several factors when evaluating nuclear facility sites.[57] Among the considerations previously applied to siting applications are the “intended use of the reactor,”[58] the safety features of the facility,[59] and both the “[p]opulation density and use characteristics of the site environs.”[60] The NRC reworked the criteria for siting applications submitted after January 1, 1997.[61] These factors include “the population distribution, and site-related characteristics,”[62] as well as the “[p]hysical characteristics of the site.”[63] Thus, until recently, the EPA and NRC read their mandates as not authorizing them to include civil rights criteria under Title VI of the Civil Rights Act[64] when deciding permit and site issues.[65]

However, in 1994 both the EPA and NRC had to confront Executive Order 12,898,[66] which they read as mandating a different approach. The order directed that all federal agencies identify and address “environmental justice” concerns — e.g., that minority communities not host a disproportionate number of high-emission chemical or other hazardous facilities — pursuant to Title VI of the 1964 Civil Rights Act.[67] The presidential directive specified that it was designed to sensitize regulatory officials to possible disproportionate effects of their decisions on minority communities. But, it was “intended only to improve the internal management of the executive branch” and did not “create any [enforceable] right, ... substantive or procedural,” or provide that agency compliance was judicially reviewable.[68]

A recent decision by the NRC’s Atomic Safety and Licensing Board (“Board”) illustrates how the Board has relied upon Executive Order 12,898 to expand its authority and reach results inconsistent with the NRC’s enabling statute. The case, In re Louisiana Energy Services, L.P.,[69] involved the siting of a nuclear facility, which according to NRC regulations generally requires that a nuclear plant be located away from sensitive “receptors,” such as hospitals, schools, and nursing homes.[70] Even though the Board found these criteria satisfied, it ruled that the site selected was invalid because minority communities generally are underserved by such “amenities.”[71] In particular, the Board severely criticized the staff’s unwillingness to consider siting the nuclear facility near a popular fishing lake.[72] Reliance on “quality of life considerations” was held to be improper because it favored the middle class at the expense of the poor.[73] In reaching this conclusion, the Board relied upon Executive Order 12,898 as modifying its siting criteria even though the President’s order expressly did not — because it could not — expand the NRC’s statutory authority.[74]

Another example concerns an experimental program in California whereby oil companies obtain extra emissions credits for each old, heavy-emission car that they buy and scrap.[75] By acquiring extra emissions credits, these oil companies are able to concentrate statewide emission sources at refineries located near minority communities, resulting in an alleged violation of Title VI and the EPA’s siting requirements.[76] These effects have led several community groups to file a complaint with the EPA, alleging that the “‘Old-Vehicle Scrapping Program’ results in an adverse disparate impact on regional communities of color.”[77]

Several opinions issued by the EPA further demonstrate the interplay between industrial plant permitting and Executive Order 12,898. In In re Chemical Waste Management of Indiana, Inc.,[78] In re Envotech, L.P,[79] and In re EcoElectrica, L.P.,[80] the EPA’s Environmental Appeals Board stated that the EPA could consider “environmental justice” issues, as set forth in Executive Order 12,898, in its criteria for reviewing permit applications. Despite concluding in these cases that Executive Order 12,898 did not add any substantive requirements to its permitting criteria, the EPA in effect now considers environmental justice concerns in its permitting decisionmaking. Thus, like the NRC, the EPA arguably has exceeded its statutory authority by incorporating Executive Order 12,898 into its permitting criteria.

These cases are similar to the FDA’s assertion of authority to regulate tobacco products where no health claims are made. In the environmental cases, the agencies’ statutory authority for identifying the criteria to be applied in permitting and siting decisions is explicit. None of these criteria, however, includes application of the civil rights laws. All have been interpreted as limiting the decisional criteria to factors directly related to the two agencies’ regulatory programs. Moreover, as with FDA’s tobacco rules, the NRC and EPA cases involve the agency’s application of an executive order that is not directly part of its regulatory authority, and thus generally not within Chevron deference.[81] Both agencies’ contrary interpretation of their authority has resulted in a substantial expansion of that authority in a manner not considered or probably even imagined by Congress. As appealing as these assertions of jurisdiction may be as social policy matters, there is no reason to defer to the agencies’ broad reading of their authority.

3. Federal Reserve Board’s Interpretation of the Glass-Steagall Act

To ensure their safety and soundness, the Bank Holding Company Act of 1956[82] and the Glass-Steagall Act[83] have long prohibited commercial banks from engaging in nonbanking activities or from owning voting shares in any company that is not a bank.[84] This insulation of commercial banks from investment banking and dealing in securities was thought necessary to avoid conflicts of interest that had resulted in breaches of fiduciary duties by banks in the collapse of the early 1930s and to restore confidence in the integrity of the commercial banking system.

However, changing times have led to calls for the elimination of the Glass-Steagall firewall to make U.S. banks more competitive. Commercial banks have long argued that they must be allowed to enter many markets, including the securities and insurance business, if they are to remain viable.[85] For example, Citicorp, a bank holding company, and Travelers Group Inc., a holding company for securities, insurance, and other financial services firms, recently sought to merge under the Bank Holding Company Act.[86] Commenters, however, argued that approval of the merger would violate the Glass-Steagall Act by allowing for the combination of a member bank with one of the largest U.S. securities firms.[87] Despite these concerns, the Federal Reserve Board (“FRB”) approved the merger finding that “Travelers has committed to conform the activities of [its domestic subsidiaries that cannot be affiliated with a bank under section 20 of the Steagall Act] to the requirements of the Glass-Steagall Act and the [FRB’s] orders and interpretations thereunder.”[88]

Traditional lending has become a much smaller market for banks because large companies can raise capital independently at a lower cost by selling bonds, commercial paper, and stocks in the capital market. In addition, U.S. banks have fallen behind large German and Japanese banks that have, among other things, acquired investment banks and engaged in the insurance business. But proposals for reform have stalled because insurance groups and smaller commercial banks (who are allowed to sell insurance under an exception to the prohibition of commercial banks from engaging in nonbanking activities[89]) fear that they would be squeezed by larger banks if the Glass-Steagall prohibitions were eliminated. As a result, proposals in Congress to increase the power of banks to deal in securities and insurance have failed because the political power of small banks and insurance agents, who reside in every congressional district, has neutralized the influence of large banks.

To overcome this legislative impasse, the FRB unilaterally “amended” section 20 of Glass-Steagall by a change in Regulation Y[90] which implements the statutory provisions separating commercial banking from nonbanking activities.[91] Regulation Y had long interpreted section 20 as prohibiting bank holding companies from acquiring a subsidiary unless it was “engaged principally” in permissible activities.[92] But under the revised regulation, the FRB has broadly expanded the range of permissible activities and allowed bank holding companies to purchase investment banks if seventy-five percent, of the new subsidiary’s activities were permissible.[93] Formerly the threshold was ninety percent, and the permissible activities category was interpreted narrowly.[94]

By allowing banks to shift permissible functions, such as underwriting of government bonds or dealing in securities for a customer to a subsidiary, the new limitations have the effect of allowing commercial banks to acquire investment banks and, perhaps, other commercial entities, without restriction. For example, the FRB has approved the purchase by Bankers Trust Co. (a commercial bank) of Alex. Brown & Sons (an investment bank) even though Alex. Brown ranked sixth in underwriting initial public offerings of stock in 1996 and was a leading underwriter of high technology stocks.[95] The practical effect of the FRB’s aggressive reading of its powers is the collapse of the statutory firewall without the benefit of legislation.

Again, the result in these cases may well be good policy. Only a few years ago, dire predictions were made that commercial banks would soon suffer the fate of the dinosaur. Investment banks, mutual funds, and other financial services were actively making loans and deposits, a business that had long been the protected preserve of commercial banks. Under amended Regulation Y, banks are repositioning themselves to respond to such competition, although newly-approved subsidiary enterprises may make it more difficult for bank auditors and regulators to measure the safety and soundness of U.S. banks. On another level, however, this dynamic interpretive response by the FRB to modernize its authority without congressional authorization seems clearly contrary to its legislative mandate and separation of powers principles. Whether the policy is wise is not at issue. Our only point is that adoption in the face of a long-standing contrary understanding by the FRB and Congress should, if challenged, eliminate any Chevron-inspired claim for judicial deference.

4. Redefining the Definition of the Term State

Most interpretations of agency authority that disregard express or implied jurisdictional limits involve narrow issues that seldom create public controversy. Thus, they are unlikely to cause recognized policy mischief. Nonetheless, the attitude that an agency may do whatever is necessary even though it is not directly authorized to do so is lawless, not merely creative. One recent example of such “creeping” assertions of jurisdiction without statutory support is the rule proposed by the Federal Housing Finance Board (“Finance Board”) to redefine the term State in its membership regulations by adding American Samoa and the Northern Mariana Islands to its list.[96] The revised definition would allow the financial institutions of both entities to be eligible for membership in the Federal Home Loan Bank. As a result, member banks in the Samoan territory and Mariana commonwealth would now be eligible for fund advances and other financial services from the Finance Board.

On the surface, this inclusion of American Samoa and Northern Mariana Islands within the scope of the Federal Home Loan Bank seems unobjectionable. It may be consistent with the broad coverage Congress intended for savings and loan institutions that are already insured entities under other regulatory programs. In contrast to the FRB’s reinterpretation of the Glass-Steagall Act, there would appear to be no opposition to including Samoan and Mariana thrifts as members of the Finance Board. Nonetheless, the route chosen by the Finance Board to expand its membership in light of a more limited statutory directive raises serious questions.

The difficulty is that the Federal Home Loan Bank Act[97] (“FHLBA”) clearly defines the “term ‘State’ [to] include[] the District of Columbia, Guam, Puerto Rico, and the Virgin Islands of the United States.”[98] The FHLBA does not include either American Somoa or the Northern Mariana Islands in this list. In addition, the defined term is specifically relied upon elsewhere in the FHLBA, and in particular, in defining the criteria for membership eligibility.[99] Thus, under traditional rules of statutory construction, including expressio unis, the omission of American Samoan and the North Mariana Islands from the statute appears to be intentional and exclusive. Nor is this omission contradicted by evidence of a strong contrary legislative intent.

The case for including the North Mariana Islands is arguably stronger because specific provisions of the Covenant Agreement[100] executed by the United States and the North Mariana Islands at the time they became a commonwealth already made the FHLBA applicable to them.[101] No such justification can be offered for American Samoa. Indeed, the Finance Board’s proposed rule openly acknowledges that “bills currently are being considered in Congress that would achieve this same result legislatively.”[102] Why administrative action was necessary or permissible under the circumstances is not explained. For our purposes, this becomes an admittedly small, but nonetheless instructive, example of why Chevron deference is not appropriate and should not be applied on judicial review, assuming of course that this becomes a step two case.

Each of these examples has appeal as substantive regulatory policy. None, however, appears to be grounded in statutory authority. All can, and probably should, be decided by a reviewing court without deference to the agency’s views. Indeed, each assertion of authority warrants special skepticism because of the agency’s obvious self-interest.[103] To the extent these decisions on review do not land in step one of Chevron, they should find no safe harbor in step two.


In their debate over the application of Chevron deference to agency jurisdiction, Justices Brennan and Scalia sharply disagreed over whether it “applies to an agency’s interpretation of a statute designed to confine its authority.”[104] Both Justices relied upon the same authorities to support their contrary positions while neither admitted any doubts or limitations about his position.

Were the question so simple, it seems to us that Justice Scalia’s position is clearly correct. Of course, Chevron allows an agency to interpret the scope of its jurisdictional authority; otherwise the rule is a nullity because there would be almost no statutory interpretations to which it could be applied. Filling statutory gaps, such as deciding how pollution is to be measured — smokestack by smokestack as the environmental-petitioners argued, or plant by plant as EPA urged and Chevron decided — ultimately is a question about the scope of the agency’s authority.

On the other hand, Justice Scalia’s further argument that it is both “necessary and appropriate” that such deference encompass all reasonable “administrative interpretation[s] of [an agency’s] statutory jurisdiction or authority”[105] lumps jurisdictional issues into a single category as if all cases are alike. His argument assumes that all jurisdictional issues are identical and asserts, without further analysis, that there is no discernible line between an agency’s authority and an agency’s “authorized application” of its authority.[106] But as Justice Breyer has explained, “there are too many different types of circumstances, including different statutes, different kinds of application, different substantive regulatory or administrative problems, and different legal postures in which cases arrive, to allow ‘proper’ judicial attitudes about questions of law to be reduced to any single simple verbal formula.”[107]

We therefore explore a multifaceted approach toward developing a peripheral jurisdictional limitation on the Chevron rule. It relies on several factors that we believe should be applied in deciding whether to defer to an agency’s interpretation of its authority. Included among these factors are: whether Congress intended that courts defer to the agency under the circumstances; whether the jurisdictional claim is within the zone of interests regulated by the agency, or alternatively, whether it is clearly outside the agency’s regular authority; and whether the statutory terms and structure are consistent with the expanded jurisdictional claim.

A. Likely Congressional Intent

By its own terms Chevron is more limited than Justice Scalia’s argument assumes. Its rule of deference “comes into play ... only as a consequence of statutory ambiguity, and then only if the reviewing court finds an implicit delegation of authority to the agency.”[108] Thus, the first question to be decided by the reviewing court, once it has determined that Congress did not address the issue itself, is whether “Congress has explicitly left a gap for the agency to fill [and whether] there is an express delegation of authority to the agency to elucidate a specific provision of the statute by regulation.”[109] When that express delegation is made, the agency’s regulation is upheld unless it is arbitrary and capricious.[110] Even where the “legislative delegation to an agency on a particular question is implicit rather than explicit,”[111] the court may find that Congress “implicitly decided” that agency experience and expertise were likely to be critical in determining whether the agency’s authority encompasses the proposed action. In that circumstance, the “court may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of the agency.”[112]

Where, however, no implicit delegation of law-interpreting authority was granted to the agency, deference to the agency’s judgment on jurisdictional issues cannot be traced to congressional intent. Without such support, Chevron teaches that basic coverage issues on the scope of the agency’s regulatory authority are for Congress to decide because there is no basis for presuming that it intended otherwise. In other words, deference to an agency’s interpretation of its primary jurisdictional coverage requires that a distinction be made between those interpretations that are within the agency’s experience and expertise — and thus likely to have been intended by Congress for agency interpretation — and those that are outside this framework, and therefore subject to independent judicial review.

The more significant the question and the greater the impact that expansion of the agency’s jurisdiction is likely to have, the greater the likelihood that Congress did not intend implicitly to delegate that determination to an agency. Nothing is more important to an agency than the scope of its regulatory authority. But agencies are established to implement regulatory programs authorized by Congress and not to decide whether those programs should be extended into new areas. In addition, the need for outside review and increased judicial scrutiny of agency-initiated expansion of its legislative authority outweighs any interest in agency sovereignty. Simply put, agencies are not common law courts. The underlying concern, as Chevron’s architecture recognizes when it requires courts to make the first determination of whether Congress answered the precise question at issue, is that agencies have no comparative advantage in reading statutes and that agency self-interest may cloud its judgment. The principle of deference under Chevron’s step two is limited to reasonable agency interpretations precisely because of a concern that “the decision to regulate may be motivated by designs for agency aggrandizement rather than by a disinterested assessment of statutory authority and appropriate policy.”[113]

Justice Breyer’s analysis of how Chevron should be applied has focused closely on assessing what a legislator supporting the statute at issue would have expected. He argues that the principal question in deciding whether step two should be invoked is whether the statute’s legislative supporters would have expected questions of subject matter coverage to be decided by the agency or a reviewing court. Several factors are likely to decide this question. If the jurisdictional question at issue is an important one, likely to have a major impact on the regulatory program and those being regulated, it seems probable that members of Congress voting for the bill would not have wanted the courts to defer to the agency’s views. For example, whether the FRB could breach the firewall between commercial and investment banking established by the Glass-Steagall Act is not something a legislator voting for the measure probably would have intended. Similarly, it seems unlikely that the Congress approving the FDCA expected it to cover a major segment of the economy not previously regulated by the FDA and not discussed in any legislative consideration.

On the other hand, if the question is interstitial involving an application of expected authority within the agency’s primary mission — e.g., does the FRB’s authority under Glass-Steagall empower it to define further what constitutes “banking” and “non-banking” activities; or does the FDA’s authority over food and drugs encompass requiring warning labels to protect consumers with specific allergies — it is more likely that the supporting Congress would have wanted courts to defer to the agency’s reasonable views. Even here, of course, Congress may have by other actions — e.g., by expressing serious concerns with any change in the separation between banking and nonbanking activities or by requiring warning labels on various alcoholic beverages posing dangers to public health[114] — given strong indication that the issue is reserved expressly for Congress to decide and therefore cannot be implied.

This examination of whether Congress meant to make an implicit delegation of jurisdiction to the agency is only the first measure for deciding whether the agency’s interpretation of its authority warrants deference. Agency experience and expertise can be described in many ways to encompass almost any question. Similarly, the interpretation of a statute is invariably important, at least to the challenging parties (as was true of the plant versus smoke-stack issue in Chevron itself). Thus, a more structured framework identifying additional criteria to measure likely congressional intent is needed. One suggestion, implied in Justice Brennan’s opinion in Mississippi Power where he stated that no deference should be given to an agency interpretation of “a statute designed to confine the scope of its jurisdiction,”[115] is that Chevron deference is not appropriate when an agency is asserting authority outside its core powers.

B. Peripheral Jurisdictional Issues

In Leedom v. Kyne,[116] the Supreme Court upheld a district court ruling that the lower court had jurisdiction to overturn an agency decision that clearly “exceeded its statutory power” and which was “contrary to a specific prohibition” in its enabling authority.[117] Although its holding only specified that the final order requirement did not prevent the trial court from reviewing interim agency findings otherwise not subject to judicial review, the Court’s decision in Leedom has relevance for the application of Chevron deference to agency statutory interpretations. At bottom, Leedom is predicated on the theory that agency actions clearly outside its regular authority deserve immediate nondeferential judicial review. To ensure that self-serving agency interests do not influence the decision of the scope of agency authority, review is warranted, indeed necessary. Leedom’s significance is limited by the procedural posture of the case before the Court where, because of its jurisdictional argument, the agency did not dispute the plaintiff’s claim that the agency had acted in excess of its jurisdiction. Nonetheless, the Court’s rationale that there was a strong inference that Congress intended that the statutory provisions governing the general jurisdiction of the courts was controlling supports the view that jurisdictional issues outside an agency’s core powers are different.

Under Leedom and other cases, reviewing courts independently have examined agency claims of regulatory authority to take action clearly beyond the scope of the agency’s traditional area of responsibility. In Leedom itself, the Court found that where the agency had sought the claimed regulatory authority from Congress and been denied, and where the agency’s exercise of the claimed regulatory power would raise serious constitutional questions, jurisdiction to review the agency’s actions was warranted even though no final order had been issued.[118] This approach is similar to that followed under the clear statement doctrine which is relied upon by courts to read agency authority narrowly where constitutional rights would otherwise be implicated.[119] However, none of these cases directly examines the scope of review applicable to challenged agency jurisdictional claims.

1. The Agency’s Primary Regulatory Assignment

The first criterion for determining whether Chevron deference should apply is whether the questioned jurisdiction is within the agency’s core regulatory assignment. It is presumptively unlikely that Congress intended to allow an agency to decide for itself whether an area of regulation is within its regulatory jurisdiction where the enabling legislation is silent on that authority. An agency only has those powers delegated to it by Congress because any other presumption would mean that enabling legislation was always an open book to which the agency could add pages and change the plot line. Otherwise the enabling act would not significantly confine agency authority because finding “some rationale” to support expansion of agency authority would seldom be difficult. To defer to the agency interpretation in this circumstance, even if that reading is reasonable, removes the independent check that courts exercise over agency actions.[120] Indeed such deference flies in the face of the APA itself, which in section 558(b) denies an agency the authority to impose sanctions on issues, rules, or orders “except within jurisdiction delegated to the agency.”[121]

For example, the safety of foods, drugs, and devices is the FDA’s primary authority; the truthfulness of advertising claims is the FTC’s principal responsibility; and the safety of consumer products not involving foods, drugs, and devices is within the usual regulatory domain of the Consumer Product Safety Commission. If the asserted jurisdiction appears to fit within the role specifically assigned to the agency and the interpretive question is of the scope of its application — such as the most appropriate means for measuring effluent sources for applying pollution control requirements set forth in the Clean Air Act (as in Chevron) — the presumption of deference that Chevron commands is applicable unless Congress has clearly indicated otherwise.

On the other hand, if the agency has not previously regulated the product or service, or asserted the power to do so, then there seems to be little basis for assuming that Congress would have wanted courts to defer to agency interpretations. Indeed, to be consistent with the Chevron reliance on congressional intent, the presumption in that circumstance should be that interpretive deference is not appropriate unless Congress has expressly said so. For example, the FDA has long regulated foods, drugs and devices if unsafe (adulterated)[122] or if marketed in a misleading way (misbranded).[123] But it has not heretofore regulated or restricted the marketing of lawful products where the message is truthful and not targeted at unlawful sales. In this situation, Chevron deference is not appropriate because it is outside the area that the FDA has traditionally controlled. That is, without textual support in the statute to imply a delegation to the agency, there should be no presumption of deference.

2. Statutory Structure, Operational Effect, and Legislative History

In deciding whether an agency’s core jurisdiction includes activity not previously regulated, courts should independently review the same materials they would otherwise consider in deciding whether step one of Chevron (i.e., did Congress answer the jurisdictional question directly) applied. These include the structure of the statute,[124] the operational effect of expanding the agency’s jurisdiction,[125] and the legislative history of that authority.[126] If the statutory structure, operational effect, and legislative history do not positively support the agency’s claimed authority, then Chevron deference is not appropriate where the regulation would extend beyond the agency’s primary areas of responsibility. The agency necessarily has no experience and probably no expertise in the new arena, further eliminating any justification for judicial deference to its jurisdictional conclusions. Here the courts are equally competent to figure out what Congress intended, and they do not have the inherent conflict of power aggrandizement.

Applied to nuclear facility siting issues, the statute, as well as clear precedent, have established the criteria by which such decisions are made. They do not include civil rights tests of Title VI or criteria as set forth in Executive Order 12,898. Here, of course, it is not even the agency (NRC) but the White House that suggests the expansion of jurisdiction. Similarly, the defining terms of the Federal Housing Finance Act as well as its legislative history give no indication that Congress intended for the Finance Board to expand the scope of the meaning of a state to include territories (American Samoa) or commonwealths (Northern Mariana Islands) without express authorization. The purpose of the Glass-Steagall Act, as well as the debates and reports issued when it was adopted, make clear that the wall between commercial banking and nonbanking activities was not to be breached, even in exceptional circumstances. Likewise, the FDA’s exercise of authority over the truthful marketing of tobacco products is supported neither by the structure nor the legislative history of the FDCA as amended by the Medical Devices Act.[127] Similarly, the legislative history of the FDCA contains no mention of tobacco products nor does it indicate that the FDA was given regulatory power over tobacco products as customarily marketed. Silence in this circumstance is especially telling because tobacco was a significant part of the economy, particularly in the states whose members were prominent supporters of the FDCA.[128]

The reliance on the structure and operational effect of the enabling legislation and on its legislative history to decide whether the assertion of jurisdiction warrants judicial deference is fully consistent with the rationale of Chevron. Chevron itself specifically relied upon the structure of the Clean Air Act, the effect of plant-wide effluents, and the Clean Air Act’s legislative history in determining Congress’s intent under step one.

Other traditional justifications for deference to agency interpretation — that the agency is better positioned to assess practical needs and regulatory requirements; that as the designers of the enabling legislation the agency knows more about what Congress intended and how it would have answered the question if put to it; and that deference ensures greater uniformity in interpreting the legislative authority — do not necessarily apply to agency interpretations of peripheral claims. Agencies have no comparative advantage in determining from the structure, terms, and legislative history of the enabling legislation whether it covers new or additional areas. Step one of Chevron expressly recognizes this as an area of special judicial competence. As an interested party, the agency’s views should carry no special weight with a reviewing court whose judgment is likely to be independent and measured — and thus more acceptable.

C. Additional Criteria

Additional criteria for locating the boundaries between step one and step two of Chevron are designed to further identify when deference is inappropriate because Congress probably did not intend to pass this question to the agency for decision. Three are identified, but if this framework were to be adopted, further development would be desirable.

1. Distance of Asserted Jurisdiction

The distance between an agency’s core jurisdiction and the proposed extension of its authority is obviously relevant. The closer the new regulatory program is to the agency’s primary assignment, the more likely it is that Congress would have expected the agency to make that decision and thus included it, at least implicitly, within step two deference. Correlatively, the more distant the regulation is from the agency’s primary program the less likely it is that Congress would have wanted the agency to expand its jurisdiction. In this circumstance, deference is inappropriate.

This criterion seems particularly relevant in addressing the authority of the EPA, NRC, and FDA to encompass civil rights or tobacco issues within their mandates. Neither the EPA nor NRC had previously applied civil rights criteria in making siting and permit decisions. Nor did existing criteria discriminate against any racial group. That facilities generally should not be located near recreational areas is common sense; that recreational facilities may be used less frequently by minority groups is likely to reflect imponderable issues of personal choice and resources. It is hard to see how such a decision is racially based. Here, the expansion of the agencies’ authority demonstrates a difference of kind, not just of degree.

2. Importance of Issue

In applying Chevron, the kind of statutory construction question presented often should be dispositive of the issue of deference. Is it a matter of detail in a statute whose general application is undisputed (as in Chevron) or is it a fundamental issue of the limits of administrative jurisdiction. Judge, now Justice, Breyer captures this point in his opinion in Mayburg v. Secretary of Health and Human Services[129] as follows:

The less important the question of law, the more interstitial its character, the more closely related to the everyday administration of the statute and to the agency’s (rather than the court’s) administrative or substantive expertise, the less likely it is that Congress (would have) “wished” or “expected” the courts to remain indifferent to the agency’s views. Conversely, the larger the question, the more its answer is likely to clarify or stabilize a broad area of law, the more likely Congress intended the courts to decide the question themselves.[130]

Applying this criterion to the FRB’s revision of the Glass-Steagall firewall, it seems unlikely that Congress meant to leave “so important and delicate a legal question”[131] to the agency to decide. Not only was the statute carefully considered at the time the Glass-Steagall Act was adopted, but Congress’s reconsideration in recent years has resulted in a stalemate because of intense lobbying pressure from warring industries. This is no Bob Jones situation. In these circumstances it is for Congress, not the agencies, to decide such significant “political, as well as policy, concerns.”[132] On the other hand, the designation of small islands as states, making them eligible for membership in the Federal Home Loan Bank, is far less momentous either as a political issue or for its impact on the Federal Home Loan Bank, even though Congress was currently considering the same issue for American Somoa.

3. Related Regulatory Assignments Housed in Other Agencies

Where more than one statute is relevant, the usual rule of statutory construction is to look at all statutes “together, as if they were one law.”[133] That is, a reviewing court must look at all relevant statutes on the same subject in determining Congress’s intent.[134] In addition, Chevron is limited to agency interpretations of the statutes they administer;[135] no special weight is given to agency interpretations of other, nonorganic statutes.[136]

Applying these concepts to an agency’s assertion of regulatory authority, a central question is whether another agency also has responsibility in the field, and if so, whether it is likely that Congress would have intended that more than one agency was to oversee the area. While duplicative regulatory authority is sometimes authorized by Congress — e.g., both the Department of Justice (“DOJ”) and FTC have authority to prohibit anticompetitive mergers and unreasonable restraints of trade[137] — that is the exception rather than the rule, and the usual presumption is that Congress does not intend to divide regulatory responsibility among two or more agencies.[138]

This issue of exclusive (or joint) authority is applicable to both the environmental siting/permitting cases as well as the FDA’s tobacco rule. Title VI of the Civil Rights Act is enforced by the DOJ and the Equal Employment Opportunity Commission. Neither the EPA nor the NRC have specific enforcement responsibilities under the Civil Rights Act. And the criteria authorized by Congress for making siting and permitting decisions have not included civil rights tests.[139] Equally dubious is the FDA’s attempt to regulate the marketing of tobacco products where specific legislation on warning labels and access grant enforcement authority to other agencies.[140]


The reading of Chevron proposed here — of not giving deference to agency assertions of authority beyond the core jurisdiction delegated by Congress — is consistent with the language and rationale of Chevron. Immediately after articulating a two-part test for determining when deference was appropriate, the Court explained that the power to administer a congressionally established program “necessarily requires the formulation of policy and the making of rules to fill any gap left, implicitly or explicitly, by Congress.”[141] Express delegations are tested under the arbitrary or capricious test under section 706(2)(A) of the APA. But where “the legislative delegation to an agency on a particular question is implicit,” the reviewing court “may not substitute its own construction of a statutory provision for a reasonable interpretation made by the administrator of an agency.”[142] In either case, section 558(b) of the APA admonishes the agency and the court to determine that its actions are “within [the] jurisdiction delegated to the agency.”[143]

By its terms, Chevron does not apply to extension of an agency’s jurisdiction beyond its core powers. In making such extraterritorial judgments, the agency is no longer filling gaps but annexing new territory.[144] If, after analyzing the statutory text and structure, and considering the pertinent legislative history, the court determines that the agency action is beyond its core jurisdiction, the court should decide for itself whether the action was authorized by Congress. This does not mean, of course, that such authority should not be found. But a court should not hesitate to strike agency action which is “effectively the introduction of a whole new regime of regulation ... , which may well be a better regime but is not the one that Congress established.”[145]

The absence of deference does not mean that a court will not find that the agency has jurisdiction. What it means, however, is that the court will not assume jurisdiction. Rather, the burden of proof is on the agency to demonstrate that Congress expressly or implicitly authorized it to act. If an agency’s jurisdictional claim is denied, express authorization from Congress can always be sought. That, it seems to us, is the better balance under our scheme of separation of powers as it has been interpreted by the Chevron case. It also coincides with our modern nondelegation doctrine: Congress can delegate in vague or “implicit” terms and the courts under Chevron will defer, but if Congress does not even go that far, no Chevron deference should be forthcoming.

* Professor of Law, George Mason University School of Law.

** Dean & Professor of Law, Benjamin N. Cardozo School of Law, Yeshiva University.

[1] U.S. CONST. art. 1, § 8.

[2] J.W. Hampton, Jr. & Co. v. United States, 276 U.S. 394, 409 (1928).

[3] See generally David Schoenbrod, The Delegation Doctrine: Could the Court Give It Substance?, 83 MICH. L. REV. 1223, 1249-74 (1985); David Schoenbrod, Goal Statutes or Rules Statutes: The Case of the Clean Air Act, 30 UCLA L. REV. 740, 803-26 (1983).

[4] See Ernest Gellhorn, Returning to First Principles, 36 AM. U. L. REV. 345, 352 (1987) (urging that the delegation doctrine be revived on a limited basis to prohibit “excessive delegations that are clearly used to create private goods”).

[5] See RICHARD PIERCE ET AL., ADMINISTRATIVE LAW AND PROCESS 54 (2d ed. 1992); see also Peter H. Aranson et al., A Theory of Legislative Delegation, 68 CORNELL L. REV. 1, 5 (1982) (analyzing “certain causes and consequences of the congressional delegation of legislative authority to executive-branch agencies and independent commissions”).

[6] See David Epstein & Sharyn O’Halloran, The Nondelegation Doctrine and the Separation of Powers: A Political Science Approach, 20 CARDOZO L. REV. 947, at 955-56 nn.24-26 (1999); Elizabeth Garrett, Accountability and Restraint: The Federal Budget Process and the Line Item Veto Act, 20 CARDOZO L. REV. 871 (1999); Dan M. Kahan, Democracy Schmemocracy, 20 CARDOZO L. REV. 795 (1999); David Schoenbrod, Delegation and Democracy: A Reply to My Critics, 20 CARDOZO L. REV. 731 (1999); Peter H. Schuck, Delegation and Democracy: Comments on David Schoenbrod, 20 CARDOZO L. REV. 775 (1999).

[7] PIERCE ET AL., supra note 5, at 54.

[8] 488 U.S. 361 (1989).

[9] Id. at 374.

[10] Id. at 372.

[11] Id. at 427 (Scalia, J., dissenting).

[12] Id. at 416.

[13] See National Cable Television Ass’n v. United States, 415 U.S. 336 (1974) (reading the FCC’s authority to collect fees narrowly by applying a “clear statement” standard). For the effects of a judicial command that an agency does not construe an act so broadly as to raise a delegation doctrine problem, see International Union, UAW v. OSHA, 938 F.2d 1310 (D.C. Cir. 1991).

[14] See, e.g., JAMES O. FREEDMAN, CRISIS AND LEGITIMACY: THE ADMINISTRATIVE PROCESS AND AMERICAN GOVERNMENT 93-94 (1978). But see JERRY L. MASHAW, GREED, CHAOS & GOVERNANCE: USING PUBLIC CHOICE TO IMPROVE PUBLIC LAWS 151-53 (1997) (suggesting that “vague delegations” may be a superior means of accountability and legitimacy).


[16] See, e.g., INS v. Chadha, 462 U.S. 919 (1983).

[17] The most celebrated exception is airline deregulation which was led by the Civil Aeronautics Board (“CAB”) and approved by Congress. See Airline Deregulation Act of 1978, Pub. L. No. 95-504, 92 Stat. 1705 (codified as amended at 49 U.S.C. §§ 40,101-40,120 (1994)). Airline deregulation was surely not the product of regulatory inertia or restrictiveness. Yet neither was it a function of jurisdictional aggrandizement. Alfred Kahn, the CAB’s chairman during its deregulatory phase, focused the agency on its central mission — that of preserving a healthy airline industry — by narrowing rather than expanding the agency’s rulemaking powers. See THOMAS K. MCCRAW, PROPHETS OF REGULATION 261-99 (1984) (outlining CAB chairman Alfred Kahn’s role in implementing deregulatory policies). Thus, while in many respects Chairman Kahn was no less ambitious in setting the CAB’s agenda than Dr. Kessler was in setting the FDA’s agenda, see infra text accompanying notes 48-51, Kahn clearly acted “within jurisdiction” as that term is interpreted under section 558(b) of the APA. See infra note 28.

[18] See, e.g., E.I. Du Pont de Nemours & Co. v. FTC, 729 F.2d 128 (2d Cir. 1984) (refusing to enforce an FTC order); Official Airline Guides, Inc. v. FTC, 630 F.2d 920 (2d Cir. 1980) (same); Boise Cascade Corp. v. FTC, 637 F.2d 573 (9th Cir. 1980) (same); see also Jeffrey H. Liebling, Comment, Judicial Usurpation of the F.T.C.’s Authority: A Return to the Rule of Reason, 30 J. MARSHALL L. REV. 283 (1996) (discussing the narrowing of the FTC’s authority by the federal courts).

[19] See Aranson et al., supra note 5.

[20] See Jonathan R. Macey, Organizational Design and Political Control of Administrative Agencies, 8 J.L. ECON. & ORG. 93 (1992); Kenneth Shepsle, Bureaucratic Drift, Coalitional Drift, and Time Consistence: A Comment on Macey, 8 J.L. ECON. & ORG. 111 (1992).

[21] 467 U.S. 837 (1984).


[23] See, e.g., St. Luke’s Hosp. v. Secretary of Health and Human Servs., 810 F.2d 325, 331 (1st Cir. 1987) (Breyer, J.).

[24] See Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U.S. 354, 377-91 (1988). In Mississippi Power, Justices Brennan, Marshall, and Blackmun joined in a dissenting opinion that argued, inter alia, that Chevron had no application to jurisdictional disputes, whether core or peripheral. In a concurring opinion, Justice Scalia argued that it did. In this Article, the term “peripheral” is intended to represent those jurisdictional questions that reflect territory outside the ambit of the agencies’ traditional or “core” exercises of authority.

[25] 323 U.S. 134 (1944).

[26] In fact, we count ourselves among the supporters of the Chevron rule as a necessary accommodation to the impracticality of applying the delegation doctrine. Nondeferential judicial review, after all, has almost as much potential to disrupt congressional delegations as does a revived nondelegation doctrine. But for Chevron to become a workable alternative to nondelegation, it must also have limits.

[27] U.S. CONST. art. I, § 8, cl. 18.

[28] 5 U.S.C. §§ 551-559 (1994). Note especially section 558(b) which provides that “[a] sanction may not be imposed or a substantive order issued except within jurisdiction delegated to the agency and as authorized by law.” Id. § 558(b) (emphasis added). This often overlooked provision articulates an independent APA ground for scrutinizing jurisdictional decisions. See John F. Duffy, Administrative Common Law in Judicial Review, 77 TEX. L. REV. 113 (1998). And it should be remembered that an agency is entitled to no Chevron deference in its interpretation of the APA. See infra note 81.

[29] See Cass R. Sunstein, Is Tobacco a Drug? Administrative Agencies As Common Law Courts, 47 DUKE L.J. 1013, 1020 (1998) [hereinafter Sunstein, Is Tobacco a Drug?].

[30] Id. at 1059.

[31] See also Richard A. Merrill, The FDA May Not Regulate Tobacco Products As “Drugs” or as “Medical Devices,” 47 DUKE L.J. 1071 (1998).

[32] See Chevron U.S.A., Inc. v. National Resources Defense Council, Inc., 467 U.S. 837, 843-44 (1984).

[33] 405 U.S. 233 (1972); see also FTC v. Brown Shoe Co., 384 U.S. 316, 320 (1966) (holding that the FTC had the “power to find ... an anticompetitive practice unfair, subject of course to judicial review”).

[34] 482 F.2d 672 (D.C. Cir. 1973) (upholding the FTC’s authority to implement trade regulation rules). The court concluded that “the Federal Trade Commission is authorized to promulgate rules defining the meaning of the statutory standards of the illegality the Commission is empowered to prevent.” Id. at 698.

[35] 358 U.S. 184, 188 (1958) (reviewing agency assertions of jurisdiction in a suit brought “to strike down an order of the [NLRB] made in excess of its delegated powers and contrary to a specific prohibition in its Act”).

[36] The NLRB had attempted to include in the “unit” for collective bargaining purposes “employees whom it found were not professional employees.” Id. at 188-89. The Court held that the NLRB had “attempted [an] exercise of power that had been specifically withheld.” Id. at 189; see also Rockford Redi-Mix Co. v. Zipp, 632 F.2d 30, 31-33 & n.6 (7th Cir. 1980) (holding that the NLRB’s refusal to issue an unfair labor practice complaint may be reviewable by a court if the NLRB has exceeded its delegated powers); Associated Builders & Contractors, Inc. v. Irving, 610 F.2d 1221, 1226-28 (4th Cir. 1979) (same); Bova v. Pipefitters & Plumbers Local 60, 554 F.2d 226, 228-29 (5th Cir. 1977) (same).

[37] 472 F.2d 179, 187 (2d Cir. 1972); see also Railway Labor Executives’ Ass’n v. National Mediation Bd., 29 F.3d 655 (D.C. Cir. 1994) (en banc).

[38] 472 F.2d at 187.

[39] See infra text accompanying notes 104-07 (discussing Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U.S. 354 (1988)).

[40] The magnitude of this safe harbor is reflected by the fact that the Supreme Court has never struck down an agency interpretation relying on step two. See Ronald M. Levin, The Anatomy of Chevron: Step Two Reconsidered, 72 CHI.-KENT L. REV. 1253, 1261 (1997).

[41] One of the authors, Ernest Gellhorn, has served as a consultant to the tobacco industry on issues of statutory construction and Chevron deference.

[42] See Regulations Restricting the Sale and Distribution of Cigarettes and Smokeless Tobacco Products to Protect Children and Adolescents, 60 Fed. Reg. 41,314 (1995) (to be codified at 21 C.F.R. pts. 801, 803, 804, 897) (proposed Aug. 11, 1995) (FDA proposed rule and jurisdictional analysis), published as final rule, 61 Fed. Reg. 44,396 (1996).

[43] See Brown & Williamson Tobacco Corp. v. FDA, 153 F.3d 155 (4th Cir. 1998) (denying the FDA jurisdiction over the regulation of tobacco products), petition for cert. filed, 67 U.S.L.W. 3484 (U.S. Jan. 19, 1999) (No. 98-1152).

[44] 21 U.S.C. § 301 (1994).

[45] See Brown & Williamson Tobacco Corp., 153 F.3d at 168 n.15 (stating that two main supporters of the Act were from two leading tobacco states).

[46] See United States v. An Article of Drug ... Bacto-Unidisk ...., 394 U.S. 784, 798 (1969).

[47] See 21 U.S.C. §§ 331(a), 352(j) (drug or device misbranded if “dangerous to health when used in the dosage, or with the frequency or duration prescribed, recommended, or suggested in the labeling thereof”); see also Public Health Cigarette Smoking Act of 1969, Pub. L. No. 91-222, § 4, 84 Stat. 87, 87-90 (1970) (amending 15 U.S.C. §§ 1331-1338 (1994)) (warning label requirement).

[48] See Coyne Beahm, Inc. v. FDA, 966 F. Supp. 1374, 1384 (M.D.N.C. 1997), rev’d, Brown & Williamson Tobacco Corp. v. FDA, 153 F.3d 155 (4th Cir. 1998), petition for cert. filed, 67 U.S.L.W. 3484 (U.S. Jan. 19, 1999) (No. 98-1152).

[49] See id. at 1393-97.

[50] Id. at 1380 (emphasis added).

[51] Id. at 1384.

[52] See Brown & Williamson Tobacco Corp. v. FDA, 153 F.3d at 161. In particular, the court reviewed the FDCA as a whole rather than through the prism of the definitional sections alone as the district court had done. See id. at 163 & n.11 (citing Adams Fruit Co. v. Barrett, 494 U.S. 638, 650 (1990); Federal Maritime Comm’n v. Seatrain Lines, Inc., 411 U.S. 726, 745 (1973) (warning against “bootstrap” jurisdiction)).

[53] Id. at 161.

[54] Id. at 162.

[55] See 10 C.F.R. § 1.1 (1998).

[56] 42 U.S.C. § 5843(b)(1) (1994).

[57] See 10 C.F.R. § 110.00 (1998) (setting forth the factors).

[58] Id. § 110.00(a)(1).

[59] See id. § 110.00(a)(4).

[60] Id. § 110.00(b).

[61] See id. § 110.20.

[62] Id. § 100.20(a).

[63] Id. § 100.20(c).

[64] 42 U.S.C. § 2000(d) (1994).

[65] See James H. Colopy, Note, The Road Less Traveled: Pursuing Environmental Justice Through Title VI of the Civil Rights Act of 1964, 13 STAN. ENVTL. L.J. 125, 181 & n.176 (1994) (referencing EPA statements denying that it had civil rights obligations).

[66] 3 C.F.R. 859 (1995), reprinted in 42 U.S.C. § 4321 (1994).

[67] See id.

[68] Id. at 863.

[69] 45 N.R.C. 367 (1997).

[70] See id. at 388, 396.

[71] See id. at 408.

[72] See id. at 394-97.

[73] See id. at 387.

[74] See Chrysler Corp. v. Brown, 441 U.S. 281, 308-13 (1979) (ruling that executive orders must be authorized by statute to have legal effect); Meyer v. Bush, 981 F.2d 1288, 1296 n.8 (D.C. Cir. 1993) (holding that the executive order lacks “legal significance”); New River Valley Greens v. United States Dep’t of Transp., 1996 U.S. Dist. LEXIS 16547, at *16-20 (W.D. Va. Oct. 1, 1996) (holding that executive orders do not create enforceable rights or obligations).

[75] See Branford C. Mank, The Environmental Protection Agency’s Project XL and Other Regulatory Reform Initiatives: The Need for Legislative Authorization, 25 ECOLOGY L.Q. 1, 26-31 (1998) (describing a lawsuit challenging the validity of the emissions-trading initiatives).

[76] See John Chambers, The Supreme Court Has Agreed to Take up an Issue That Has Stymied Regulators and Judges: Waste Disposal Facilities Planned for Construction in Minority Areas, NAT’L L.J., June 22, 1998, at B6.

[77] Id. The complaint filed with the EPA is captioned as Communities for a Better Environment v. South Coast Air Quality Management District, EPA File No. 10R-97-R9 (filed June 23, 1997). See id. at B7 n.23.

[78] 1995 EPA App. LEXIS 25 (June 29, 1995).

[79] 1996 EPA App. LEXIS 4 (Feb. 15, 1996).

[80] 1997 EPA App. LEXIS 5 (Apr. 8, 1997).

[81] See United States v. Florida E. Coast Ry. Co., 410 U.S. 224, 236 n.6 (1973) (holding that the ICC does not have primary responsibility for administering the APA and its interpretations “do[ ] not carry the weight, in ascertaining the intent of Congress, that an interpretation by an agency ‘charged with the responsibility’ of administering a particular Glass-statute does”); Air N. Am. v. Department of Transp., 937 F.2d 1427, 1436-37 (9th Cir. 1991) (same) (post-Chevron).

[82] Bank Holding Company Act of 1956, 12 U.S.C. § 1841.

[83] Banking Act of 1933, 12 U.S.C. § 227, 48 Stat. 162.

[84] See 12 U.S.C § 1843 (1994); see also 12 U.S.C §§ 24, 78, 221, 335, 377, 378(a).

[85] See Richard W. Steveson, Financial Services Heavyweights Try Do-It-Yourself Deregulation, N.Y. TIMES, Apr. 7, 1998, at A4 (discussing the “gambl[e]” of Citicorp (a bank) and Travelers (an insurance company) in merging, despite a ban existing on such combinations).

[86] See Order Approving Formation of a Bank Holding Company and Notice to Engage in Nonbanking Activities, 84 Fed. Res. Bull. 985, 985 (1998).

[87] Id. at 1004.

[88] Id. at 985.

[89] See Nationsbank of N.C., N.A. v. Variable Annuity Life Ins. Co., 513 U.S. 251 (1995).

[90] Bank Holding Companies and Change in Bank Control (Regulation Y), 12 C.F.R. pt. 225 (1998).

[91] See 12 U.S.C. § 377 (1994).

[92] Id.

[93] See Revenue Limit on Bank-Ineligible Activities of Subsidiaries of Bank Holding Companies Engaged in Underwriting and Dealing in Securities, 61 Fed. Reg. 68,750, 68,752 (1996) (effective Mar. 6, 1997).

[94] See id. at 68,751.

[95] See Applications Approved Under Bank Holding Company Act, 84 Fed. Res. Bull. 507, 511 (1998); Orders Issued Under the Bank Holding Company Act, 83 Fed. Res. Bull. 780, 780-84 (1997) (order approving notice to engage in nonbanking activities).

[96] Membership Eligibility, 62 Fed. Reg. 49,943, 49,943 (1997).

[97] 12 U.S.C. §§ 1421-1449 (1994).

[98] Id. § 1422(3).

[99] Id. § 1424(a)(1)(A).

[100] See Covenant to Establish a Commonwealth of the Northern Mariana Islands in Political Union with the United States of America §§ 502(a)(1), 502(a)(2) (1986) (cited in Membership Eligibility, 62 Fed. Reg. 49,943, 49,944 (1997)). Both the legislation approving the commonwealth and the related presidential proclamation make this clear.

[101] See id.

[102] Id. at 49,944.

[103] See National Credit Union Admin. v. First Nat’l Bank & Trust Co., 522 U.S. 479 (1998) (unanimously applying Chevron step one to reverse an agency’s interpretation of its jurisdictional authority over credit unions).

[104] Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U.S. 354, 380 (1988); see Quincy M. Crawford, Chevron Deference to Agency Interpretations That Delimit the Scope of the Agency’s Jurisdiction, 61 U. CHI. L. REV. 957 (1995).

[105] Mississippi Power & Light Co., 487 U.S. at 381.

[106] This contention by Justice Scalia, that jurisdictional cases are indistinguishable, also is contradicted by his very next sentence conceding that there is, however, a discernable line where Congress has defined such authority in “plain terms” and its intention is obvious, “in which case the agency has no discretion” to decide otherwise. Id. at 382. This of course may reflect Justice Scalia’s view that he is more comfortable deciding issues of agency authority in Chevron step one and if so we have no quarrel. See Michael Herz, Deference Running Riot: Separating Interpretation and Lawmaking Under Chevron, 6 ADMIN. L.J. AM. U. 187, 220-21 (1992) (seeking to reconcile Justice Scalia’s views by placing both substantive and jurisdictional issues in step one, rather than step two).

[107] Stephen Breyer, Judicial Review of Questions of Law and Policy, 38 ADMIN. L. REV. 363, 373 (1986) [hereinafter Breyer, Judicial Review].

[108] Sea-Land Serv., Inc. v. Department of Transp., 137 F.3d 640, 645 (D.C. Cir. 1998) (citing Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 842-44 (1984)).

[109] Chevron, 467 U.S. at 843-44.

[110] Step two of Chevron requires the reviewing court to defer to the agency if the agency’s interpretation is “reasonable.” For an analysis that this standard should be merged with the “arbitrary and capricious” test under 5 U.S.C. § 706(2)(A) (1994), see Levin, supra note 40.

[111] Chevron, 467 U.S. at 844.

[112] Id.

[113] Thomas W. Merrill, Judicial Deference to Executive Precedent, 101 YALE L.J. 969, 1024 (1992); see also Cass R. Sunstein, Constitutionalism After the New Deal, 101 HARV. L. REV. 421, 467 (1987) (referring to agency aggrandizement by stating that “foxes should not guard henhouses”). We much prefer this view of Professor Sunstein to his agencies as common law courts one. See Sunstein, Is Tobacco a Drug?, supra note 29.

[114] See 21 U.S.C. § 352(e)(1) (1994).

[115] Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U.S. 354, 387 (1988).

[116] 358 U.S. 184 (1958).

[117] Id. at 186, 188.

[118] See id. at 188-91.

[119] See, e.g., National Cable Television Ass’n v. United States, 415 U.S. 336, 342-43 (1974); Kent v. Dulles, 357 U.S. 116, 129 (1958); Ashwander v. Tennessee Valley Auth., 297 U.S. 288, 345-48 (1936) (Brandeis, J., concurring).

[120] See Breyer, Judicial Review, supra note 107, at 364-65 (“[C]urrent doctrine is anomalous. It urges courts to defer to administrative interpretations of regulatory statutes, while also urging them to review agency decisions of regulatory policy strictly.”).

[121] 5 U.S.C. § 558(b) (1994). This section adds a statutory basis for the requirement that agencies act within their jurisdiction, and, since agencies are not entitled to Chevron deference in interpreting the APA, it also adds an independent basis for judicial review. See supra note 28.

[122] See 21 U.S.C. § 342 (1994) (adulterated food), id. § 351 (adulterated drugs and devices).

[123] See id. § 343 (misbranded food); id. § 352 (misbranded drugs and devices).

[124] See United Sav. Ass’n v. Timbers of Inwood Forest Assocs., 484 U.S. 365, 371 (1988).

[125] See Dole v. United Steelworkers of Am., 494 U.S. 26, 35-41 (1990).

[126] See Atherton v. FDIC, 519 U.S. 213, 228-31 (1997).

[127] 21 U.S.C. § 360 (1994).

[128] See United States v. Rutherford, 442 U.S. 544, 554 n.10 (1979) (“[O]nce an agency’s statutory construction has been ‘fully brought to the attention of the public and Congress,’ and the latter has not sought to alter that interpretation although it has amended the statute in other respects, then presumably the legislative intent has been correctly discerned.”); see also Bob Jones Univ. v. United States, 461 U.S. 574 (1983):

Nonaction by Congress is not often a useful guide, but the nonaction here is significant.... In view of its prolonged and acute awareness of so important an issue, Congress’ failure to act on the bills proposed on this subject provides added support for concluding that Congress acquiesced in the [agency’s] rulings ....

Id. at 600-01. Sometimes a matter is so clear that no express provision is needed. See NLRB v. Bell Aerospace Co., 416 U.S. 267, 283-84 (1974).

[129] 740 F.2d 100 (1st Cir. 1984); see St. Luke’s Hosp. v. Secretary of Health and Human Servs., 810 F.2d 325, 331 (1st Cir. 1987) (Breyer, J.).

[130] 740 F.2d at 106 (citations omitted).

[131] Breyer, Judicial Review, supra note 107, at 371.

[132] Id.

[133] United States v. Stewart, 311 U.S. 60, 64 (1940) (citing United States v. Freeman, 44 U.S. (3 How.) 556, 564 (1845)).

[134] See Argentine Republic v. Amerada Hess Shipping Corp., 488 U.S. 428, 434-35 & n.3 (1989) (concluding that a later, more “comprehensive statutory scheme” prevails over earlier statute). Similarly, “when two statutes are in conflict, a specific statute closely applicable to the substance of the controversy at hand controls over a more generalized provision.” Farmer v. Employment Sec. Comm’n, 4 F.3d 1274, 1284 (4th Cir. 1993).

[135] See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843 (1984).

[136] See Adams Fruit Co. v. Barrett, 494 U.S. 638, 649-50 (1990).

[137] See 15 U.S.C. § 18 (1994) (mergers); id. § 1 (DOJ authority regarding restraints of trade); id. § 45(a)(1) (FTC authority regarding restraints of trade).

[138] See, e.g., Martin v. Occupational Safety and Health Review Comm’n, 499 U.S. 144, 151 (1991) (“Under most regulatory schemes rulemaking, enforcement, and adjudicative powers are combined in a single administrative authority.”); see also Martin v. Pav-Saver Manuf. Co., 933 F.2d 528, 530 (7th Cir. 1991) (citing Martin, 499 U.S. at 151).

[139] The application of “civil rights” criteria in the Bob Jones case is distinguishable. There the Court upheld the agency’s construction of the Internal Revenue Code as denying tax benefits under § 501(c)(3) to racially discriminatory private religious schools because they did not come within common law standards of charity, namely, they did not “serve a public purpose” and were “contrary to established public policy.” Bob Jones Univ. v. United States, 461 U.S. 574, 586 (1983).

[140] See discussion supra notes 42-51.

[141] Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843 (quoting Morton v. Ruiz, 415 U.S. 199, 231 (1974)) (emphasis added).

[142] Id. at 844.

[143] 5 U.S.C. § 558 (1994).

[144] See Chevron, 467 U.S. at 844 (“If Congress has explicitly left a gap for the agency to fill, there is an express delegation of authority to the agency to elucidate a specific provision of the statute by regulation.”).

[145] MCI Telecomms. Corp. v. AT&T, 512 U.S. 218, 234 (1994) (striking down FCC policy).

Contents | Text Version