Updating Deference: The Court's 2001-2002 Term Sows More Confusion About Chevron

William S. Jordan III

The author is a C. Blake McDowell Professor of Law, University of Akron School of Law. He received his J.D. from the University of Michigan Law School and his B.S. from Stanford University.

[32 ELR 11459]

The U.S. Supreme Court's Chevron, U.S.A., Inc. v Natural Resources Defense Council, Inc.¹ decision has dominated administrative case law and scholarship for nearly 20 years.² In 1999, this dominance prompted the participants in the bi-annual Administrative Law Discussion Forum to plead that the next Forum topic be "anything other than Chevron!"³ At the 2001 Forum, however, Chevron again took center stage,⁴ driven to the fore by the Court's then-recent decisions in Christensen v. Harris County⁵ and United States v. Mead Corp.⁶

One would think that after all this time and discussion we would know what Chevron means. Far from it. In 2001, Thomas Merrill's and Kristin Hickman's influential article, Chevron's Domain,⁷ identified 14 unanswered questions concerning the application of Chevron deference to agency statutory interpretation. Mead represented an opportunity to provide clear answers to at least one of these questions, and perhaps two others.⁸ Unfortunately, as discussed below, Mead only confirmed the uncertainty as to when Chevron deference is available to agency statutory interpretations.

The Court's 2001 Term offers little hope that the discussion of Chevron will subside any time soon. The Court issued 12 decisions citing Chevron⁹ and two others involving the application of Chevron principles.¹⁰ Of those, seven involve the question of when Chevron applies to an agency interpretation.¹¹ If anything, the Court has increased the confusion over how to decide whether Chevron applies to agency statutory interpretations. At least two decisions raise questions about either the nature of the judicial role in reviewing interpretations to which Chevron does not apply or the continued viability of Chevron deference.¹² The remaining decisions involve disputes over whether the statute is ambiguous (Step One of Chevron analysis), whether, if the statute is ambiguous, the agency adopted a reasonable interpretation (Step Two of Chevron analysis),¹³ or whether the [32 ELR 11460] agency should prevail under Skidmore v. Swift & Co.¹⁴ deference.¹⁵ These topics are beyond the scope of this Article. It should be noted however that one decision, Verizon Communications, Inc. v. Federal Communications Commission,¹⁶ bears on the debate over the analytical method to be used in determining whether an interpretation is "reasonable" under Step Two of Chevron.¹⁷

I. Background

Chevron may have become the dominant decision in the area of review of agency statutory interpretations, but it was far from the first.¹⁸ In particular, well before Chevron there were two strands of deference analysis.¹⁹ The first is represented by National Labor Relations Board v. Hearst Publications, Inc.,²⁰ in which the Court reviewed an interpretation that the agency had reached in the course of deciding a formal adjudication. According to Hearst, in such circumstances, the courts must accept an agency's interpretation if there was "'warrant in the record' and a reasonable basis in law."²¹ It is inherent in the Hearst formulation that the primary interpretive power lies with the agency. The court, as a reviewing body, must accept a reasonable agency interpretation even if the court would reach a different result on its own authority.

The second strand of pre-Chevron deference analysis is represented by Skidmore, in which the Court reviewed an interpretation reached by a lower court in deciding a private lawsuit. In Skidmore, the agency's interpretation entered the picture by way of statements in an Interpretive Bulletin and in an amicus curiae brief filed with the lower court.²² Thus, the Court did not review the agency's exercise of decisionmaking authority. Rather, it reviewed the lower court's statutory interpretation. One of the questions in that review was the significance of the agency's informal statement of its interpretation. Noting that the agency's informal statement was "not controlling on the courts by reason of their authority," the Court held that such statements "do constitute a body of experience and informed judgment to which courts and litigants may properly resort for guidance. The weight of such a judgment in a particular case will depend upon the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control."²³ Known today as "Skidmore deference," this formulation leaves the decisionmaking power with the court, which must give respectful consideration to the views of the agency in light of the presence or absence of the relevant factors.

For the purpose of this discussion, the important point is that the choice between these strands of deference could make a significant difference to the outcomes of interpretive disputes. Under Hearst-type deference, the court theoretically must accept a reasonable interpretation that it would not choose on its own authority, while under Skidmore-type deference, the court may adopt its own interpretation, no matter how reasonable the agency's interpretation may seem.²⁴

In 1984, Chevron took the deference world by storm. Reviewing an agency interpretation embodied in a statutorily authorized informal rulemaking, the Court said that

when a court reviews an agency's construction of the statute which it administers, it is confronted with two questions. First, always, is the question whether the U.S. Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress. If, however, the court determines Congress has not directly addressed the precise question at issue, the court does not simply impose its own construction on the statute, as would be necessary in the absence of an administrative interpretation. Rather, if the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency's answer is based on a permissible construction of the statute.²⁵

This quickly became known as the Chevron two-step.²⁶ The first step involves deciding whether the statute is ambiguous on the point at issue. Although it may be that the agency is entitled to some respect at this stage,²⁷ the term "Chevron deference" refers to the second step, in which the court must accept an agency's interpretation of an ambiguous statute as long as it is reasonable.²⁸

[32 ELR 11461]

The deference of Chevron Step 2 sounds remarkably like Hearst deference. Indeed, some have argued that Chevron did not represent a significant break from previous treatments of agency interpretations.²⁹ Nonetheless, Chevron provides the framework within which the courts now examine agency statutory interpretations.³⁰

Early commentary focused on the nature of Chevron deference, including particularly the question of whether it usurped the role of the courts.³¹ It was only later that attention focused on the issue raised by the first sentence from the Court's statement quoted above. Chevron deference is available only "when a court reviews an agency's construction of the statute which [the agency] administers . . . ."³² What does that mean? When, for the purposes of Chevron deference, does an agency "administer" a statute?³³

Chevron itself was clear enough. It involved a statutorily authorized informal rulemaking proceeding. By the same token, Hearst is clear enough as a statutorily authorized formal adjudication. Both involved specific delegations of authority to "administer" the statute in making decisions that would be binding on the public. The U.S. Environmental Protection Agency rule at issue in Chevron by its nature is binding on all to whom it applies. The Hearst adjudicatory decision would be binding on the parties to the proceeding and effectively bind others by alerting them to the position taken by the agency in a formal proceeding.³⁴

But what about the Skidmore-type situation? There, the agency had various responsibilities to develop expertise in the substantive area and to file lawsuits to enforce the Fair Labor Standards Act,³⁵ but it had no authority to issue substantive rules or to decide formal adjudications that bound anyone. In the language of the Administrative Procedure Act (APA), its interpretive statements were "rules" because the APA definition of "rule" is so broad that it encompasses practically any agency statement about the future.³⁶ But they were rules with no foundation in authority to issue rules or to decide individual adjudications.

It is easy to see how the Skidmore-type statement might not be entitled to Chevron deference. After all, Chevron involved a statement based in rulemaking authority, while Skidmore involved a statement whose basis lay only in expertise and responsibility, not in authority. The Court adopted this distinction in Christensen.³⁷

Christensen was a virtual clone of Skidmore, involving the same agency making a similar type of informal statement interpreting the same statute. Once again, the interpretive issue arose in a private lawsuit, not in the context of review of a statutorily authorized rule or formal adjudication. Rejecting an argument that such informal statements are entitled to Chevron deference, the Court said that "interpretations such as those in opinion letters—like interpretations contained in policy statements, agency manuals, and enforcement guidelines, all of which lack the force of law—do not warrant Chevron-style deference."³⁸

With this statement, Christensen seemed to give a fairly clear answer to the question of whether certain nonlegislative rules are entitled to Chevron deference.³⁹ The types of documents referred to by the Court—opinion letters, policy statements, agency manuals, and enforcement guidelines—may be issued by an agency at any time without any particular process. They provide guidance to the public about the agency's views. They may bind agency employees to positions taken by the agency leadership, but they do not bind the agency head. They do not, to use the Court's words, have the "force of law,"⁴⁰ apparently because they have not gone through the rulemaking process or otherwise been issued in "the exercise of the Secretary's delegated lawmaking powers."⁴¹

Thus, Christensen appears to identify the test by which we will determine whether Chevron deference applies to a given agency statement. The statement must have the "force of law." Mere informal issuances such as opinion letters do not meet that test.

[32 ELR 11462]

Christensen also confirmed the continuing viability of Skidmore deference where Chevron deference is not available. Opinion letters, policy statements, agency manuals, and the like are entitled to respect in light of "those factors which give [them] power to persuade, if lacking power to control."⁴²

Just a year later, however, Mead considerably muddied the waters by indicating that the availability of Chevron deference is a matter of legislative intent, to be determined, as Justice Antonin Scalia put it in dissent, by considering "the totality of the circumstances."⁴³ Mead, in particular, generated much discussion at the 2001 Administrative Law Forum.⁴⁴

Mead involved review of a U.S. Customs decision that certain imported day planners were "diaries" under the applicable statute and thus subject to payment of a tariff.⁴⁵ The Customs decision was reached through a "ruling letter" that was not subjected to notice and comment and was not the result of a formal adjudication. Although the ruling letter was specifically authorized by statute, the Court held that it was not entitled to Chevron deference. The Court apparently reached this conclusion because the ruling letter was officially binding only on the parties to the decision with respect to the particular merchandise, the agency did not treat such rulings as binding on others, and the statute provided that such rulings were subject to independent review in the Court of International Trade.⁴⁶

In reaching this conclusion, the Court held that Chevron deference is available only where (1) the statute has delegated to the agency the authority to make rules with the force of law, and (2) the agency has issued its interpretation in the exercise of that authority.⁴⁷ Accordingly, the threshold question is how to determine whether the statute has delegated lawmaking authority to the agency.

The Court initially answered this question in general terms: "Delegation of such authority may be shown in a variety of ways, as by an agency's power to engage in adjudication or notice-and-comment rulemaking, or by some other indication of a comparable congressional intent."⁴⁸ The Court's later discussion confirmed that Chevron deference is available where the agency is authorized to act through legislative rulemaking or through formal adjudication. In the Court's words, each of these would constitute a "relatively formal administrative procedure tending to foster the fairness and deliberation that should underlie a pronouncement of such force."⁴⁹

It is important to note, however, that the availability to the agency of regulatory or adjudicatory power does not, by itself, establish that the agency is entitled to Chevron deference. Rather, the availability of either of these powers is an important indicator of congressional intent to delegate "force of law" interpretive authority to the agency. Thus, "the want of that procedure . . . does not decide the case."⁵⁰ Citing the example of personal authority delegated to the Comptroller of the Currency,⁵¹ the Court held that Chevron deference may be available even in the absence of relatively formal procedures. The question in such cases would be whether, under all the circumstances, Congress meant to delegate to the agency the authority to issue interpretations with the force of law.⁵²

Christensen and Mead have been criticized on several grounds.⁵³ Some commentators argue, for example, that the availability of Chevron deference should not be based upon a determination of whether Congress intended to delegate such authority, but on other considerations, such as whether the agency is implementing rather than merely interpreting the statute,⁵⁴ or "whether the agency possesses relevant expertise . . . and . . . [has] brought that expertise to bear" on the matter,⁵⁵ Others argue that Mead would remove too many decisions from Chevron's purview.⁵⁶ Similarly, perhaps most important to the immediate discussion, some commentators have argued that the intent test of Mead is inappropriate, either because it seeks an intent that does not exist,⁵⁷ or because it is unworkable.⁵⁸

[32 ELR 11463]

II. Chevron During the Court's 2001 Term

One of the attractions of Chevron is the fact that it recognized that statutory ambiguity creates a situation in which some entity—agency or court—must make a policy decision.⁵⁹ Chevron itself assigned that policy decision to the agency, both because the ambiguity constituted a congressional delegation to the agency and because the agency is more politically accountable and thus the more appropriate forum for making policy.⁶⁰ The difficulties in determining when Chevron applies may complicate things, but they should not undermine the proposition that when Chevron deference applies, the agency is the primary decisionmaker. By contrast, when Chevron does not apply, the court is responsible for performing its traditional function—determining "what the law is."⁶¹

Unfortunately, the Chevron cases from the Court's 2001 Term have compounded the confusion over Chevron's applicability. They may also have begun to undermine Chevron's sharp distinction between the administrative and judicial roles in interpreting statutes in the administrative state.

A. Determining When Chevron Deference Applies

In the days before Chevron, the Court appeared to take a multifactor approach to when deference should be given to agency interpretations. With its focus on ambiguity, Chevron seemed to bring an end to that analytical approach, replacing it with a crisp distinction between deciding whether deference applies to a particular interpretation (Step One), and deciding whether to sustain the agency's position under the deferential reasonableness standard (Step Two).

Four decisions from the Court's 2001 Term may significantly influence the resolution of the Step One question of when deference applies. Barnhart v. Walton⁶² seems to have revived multifactor analysis, albeit in a somewhat different context. Utah v. Evans,⁶³ on the other hand, may have provided some guidance concerning when agencies meet the Christensen/Mead test of acting with the "force of law." Edelman v. Lynchburg College⁶⁴ may have answered the question of whether Chevron deference is due to a Chevron-qualified agency that adopts a procedural rule without going through notice and comment, but the decision is hardly clear on the point. And Hoffman Plastic Compounds, Inc. v. National Labor Relations Board⁶⁵ may resolve the question of whether Chevron deference is due when an agency interprets its own statute where that statute is in tension with a statute not assigned to that agency.⁶⁶

1. Barnhart—A Return to Multifactor Analysis?

Barnhart involved a claim for disability insurance benefits and supplemental security income based upon a mental illness that rendered the claimant unfit for substantial employment for at least several months. As a result of his mental condition, the claimant lost his teaching job on October 31, 1994. By mid-1995, he was able to return to work as a cashier part time, and by December 1995, he held a full-time position. The Agency determined that by the end of September 1995, he had earned enough income to reach a level of "substantial gainful activity."⁶⁷

Walton's entitlement to disability benefits depended upon whether his condition qualified as a "disability" under the following definition: "inability to engage in any substantial gainful activity by reason of any medically determinable physical or mental impairment which can be expected to result in death or which has lasted or can be expected to last for a continuous period of not less than 12 months."⁶⁸ In Walton's case, there appears to have been no doubt that his mental condition (his "impairment") could "be expected to last" for 12 months, but he had returned to "substantial gainful activity" within 12 months of losing his job due to his condition.

Walton's situation raised two questions. The first is whether it is the "inability" to work or the underlying "impairment" that must last or be expected to last for at least 12 months. This Article will refer to the first question as the "inability" issue. Assuming that the "inability" must last for the required time, the second question is whether the "inability" would qualify as a disability if it would have been "expected to last" for 12 months, even though it actually lasted for less than 12 months.⁶⁹ This Article will refer to the second question as the "expected to last" issue.

The Agency ruled against Walton on both counts. In so doing, it relied upon two regulations issued "pursuant to statutory rulemaking authority."⁷⁰ As to the "inability" issue, the Agency had issued a rule stating "that a claimant is not disabled 'regardless of [his] medical condition,' if he is doing 'substantial gainful activity.'"⁷¹ No one disputed the Agency's interpretation of this regulation to the effect that an impairment did not give rise to a disability if the claimant [32 ELR 11464] was able to return to "substantial gainful activity" within 12 months of the onset of the impairment.⁷² As to the "expected to last" issue, the Agency had issued a rule under which the date of the actual return to work, if known at the time the claim is decided, would be determinative for the purpose of establishing whether the claimant had met the 12-month requirement.⁷³

The U.S. Court of Appeals for the Fourth Circuit rejected both of the Agency's interpretations. It held that the statutory language was clear in dictating that the 12-month period applied only to "impairment" and that an impairment qualified as long as it would have been "expected" to last 12 months, apparently regardless of how long it had actually lasted.⁷⁴

The Supreme Court reversed. Applying the Chevron framework, the Court found the statute ambiguous in both areas.⁷⁵ The Court then turned to the question of whether the interpretations were "permissible," or "reasonable,"⁷⁶ the second and deferential step of Chevron analysis.

The Court applied Chevron deference without initially presenting any explanation of why Chevron applied to the particular agency interpretations. This is hardly surprising. Both interpretations were embodied in statutorily authorized legislative rules that had presumably been adopted through notice and comment.⁷⁷ They were precisely the sort of agency interpretations entitled to deference under the facts Chevron itself and the later decisions in Christensen and Mead.

So why does Barnhart raise any question about the application of Chevron deference, even to the point of suggesting a possible return to multifactor analysis? The answer comes in the Court's discussion of whether the Agency's interpretations were reasonable. The Court makes several statements strongly suggesting Chevron's applicability can be determined by a pre-Chevron-type analysis. First, the Court relies upon the long-standing nature of the Agency's interpretation as a basis for deference. Second, the Court twice suggests that several factors bear upon the question of whether Chevron deference should apply to a particular agency interpretation.

a. The Significance of the "Long-Standing" Duration of an Agency's Interpretation

In hording that the agency's "inability" issue interpretation was "permissible," the Court noted that the "Agency's regulations reflect the Agency's own long-standing interpretation," citing three informal issuances dating back to the 1950s.⁷⁸ The reference to an interpretation of "long-standing duration" harkens back pre-Chevron days when this was one of the factors that could support judicial deference to an agency's interpretation.⁷⁹ This could be taken to mean that Barnhart signals a return to the pre-Chevron approach to deference analysis. As discussed below, however, this aspect of Barnhart can be reconciled with Chevron by assuring that long-standing agency interpretations are considered relevant in addressing the reasonableness question of Chevron Step Two, rather the applicability question that arises under Chevron Step One.

Under Chevron, as later explained more fully in Mead, the applicability of Chevron Step Two, or strong deference (the requirement to accept the agency interpretation as long as it is reasonable), depends upon a congressional delegation of interpretive authority.⁸⁰ The mere existence of a long-standing agency interpretation could have no bearing on the existence of such a delegation. Any interpretation would have to be issued after the congressional grant of interpretive authority. Congress could not known when it enacted the statute that the agency would take a position that would eventually become a long-standing interpretation. Thus, the eventual long-standing nature of an interpretation cannot support the existence of a delegation of interpretive authority.

Moreover, on the facts of this case, there was no need to go beyond the existence of the underlying legislative rule in order to establish that Chevron deference applied to the interpretation in question. Accordingly, it appears that the Court's reference to the long-standing nature of the agency's interpretation does not relate to the Step One question of whether strong deference analysis applied to the interpretation, but to the Step Two question of whether the interpretation was reasonable.⁸¹

According to Justice Scalia, the proposition "that 'particular deference' is owed 'to an agency interpretation of long-standing' duration . . . is an anachronism" dating back to the pre-Chevron era in which "there was thought to be only one 'correct' interpretation of a statutory text."⁸² Justice Scalia is surely correct in the sense that the long-standing nature of an agency interpretation has no bearing on whether Congress has delegated to the agency the authority to make choices with the force of law within the policy space provided by an ambiguous statute.⁸³ In other words, it has no bearing on the applicability of Chevron deference to a particular interpretation.

The Court's opinion to this point is most plausibly read as assuming that Chevron deference applies, and using the long-standing nature of the interpretation to support a conclusion [32 ELR 11465] that the agency's interpretation is reasonable.⁸⁴ The Court's use of the term "particular deference"⁸⁵ and its citation to a pre-Chevron decision both undermine this reading slightly.⁸⁶ But it is the best way to reconcile the Court's discussion with the traditional Chevron frame-work, and even to address Justice Scalia's argument that the antiquity of an interpretation does not bear on whether it will "qualify for deference."⁸⁷ As long as the long-standing nature of an interpretation is considered under Step Two rather than Step One, the analysis would be consistent with Chevron analysis.

b. The Significance of the Previous Informal Interpretations—Diluting Chevron?

Walton argued that the Court should disregard the Agency's formal regulations "on the ground that the Agency only recently enacted those regulations, perhaps in response to this litigation."⁸⁸ The Court initially dismissed this argument,⁸⁹ as well it should have. If Chevron deference derives from a delegation of lawmaking authority, it should not matter when the agency exercised that authority. Indeed, it would be entirely appropriate for an agency even to change a position at the last minute, as long as the agency exercised the appropriate authority.⁹⁰

Unfortunately, the Court proceeded to a confusing and utterly unnecessary discussion of the fact that the "inability" interpretation embodied in the regulation had previously been reached informally, without engaging in the rulemaking process. The Court said: "Regardless, the Agency's interpretation is one of long standing. . . . And the fact that the Agency previously reached its interpretation through means less formal than "notice-and-comment" rulemaking, . . . does not automatically deprive that interpretation of the judicial deference otherwise its due."⁹¹

The Court noted Chevron's statement that an agency empowered to administer a regulatory program must engage in the "formulation of policy."⁹² It then emphasized Mead's identification of instances in which agency interpretations were given Chevron deference despite not having been adopted through notice-and-comment rulemaking.⁹³

The Court's discussion is largely unnecessary because the interpretation at issue in Barnhart was issued through notice and comment. It would have been enough simply to note that the agency's use of legislative rulemaking definitively established the availability of Chevron deference such that the informality of any previous statements was irrelevant.

The Court's use of the word "regardless" suggests that the Court is saying that even if the Agency's use of notice-and-comment rulemaking were not enough to trigger Chevron, the interpretation would nonetheless qualify for Chevron deference. The Court has gone out of its way to emphasize that the use of the legislative rulemaking process is not necessary for Chevron deference to apply.

Moreover, the Court confuses matters by saying that the interpretation's previous informal status "does not automatically deprive that interpretation of the judicial deference otherwise its due."⁹⁴ Surely that is correct. If an inter-pretation has been incorporated into a valid legislative rule, there is no reason to think that its previous informal status would ever undermine its legitimacy or its entitlement to Chevron deference.⁹⁵ Unfortunately, the Court's use of the language "does not automatically deprive" suggests that there might be some circumstances beyond the "automatic" in which an interpretation's previous informal status might deny Chevron deference to a later interpretation issued through notice-and-comment rulemaking or formal adjudication.

Although the Court is by no means clear on the point, a careful reading of this passage suggests, as noted above, that the Court has simply taken the opportunity to emphasize that some informal statements are entitled to Chevron deference even though they were not subjected to notice-and-comment rulemaking. The Court's pointed narrowing of Christensen in this discussion suggests that this was the Court's primary point.⁹⁶ The Court's "automatically deprive" statement should not be taken to mean that that there is any situation in which prior informal status might deprive an interpretive statement of Chevron deference.

c. A Return to Multifactor Analysis to Determine When Strong Deference Is Available?

What are we to make of the Court's statements in this passage of Barnhart? They are completely unnecessary to the Court's review of the interpretations at issue, which were embodied in rules issued through notice-and-comment rulemaking. To that extent, therefore, they are dicta and perhaps deserving of little weight.

[32 ELR 11466]

But if they are unnecessary, why are they there? Perhaps the Court is moving away from its earlier emphasis on the existence of formal adjudication, notice-and-comment rulemaking, or similar "relatively formal administrative procedure tending to foster the fairness and deliberation that should underlie a pronouncement of such force."⁹⁷

This passage states a new test for determining whether Chevron deference applies to a particular agency interpretation: "Whether a court should give [Chevron] deference depends in significant part upon the interpretive method used and the nature of the question at issue."⁹⁸ Moreover, the Court emphasized that Mead's discussion of why Chevron deference did not apply on those particular facts "would have been superfluous had the presence or absence of notice-and-comment rulemaking been dispositive."⁹⁹ This suggests, quite remarkably, that there may be situations in which the agency's use of statutorily authorized notice-and-comment rulemaking would not assure that the agency's interpretation qualified for Chevron deference.¹⁰⁰

When does the "interpretive method used" and the "nature of the question" support the application of Chevron deference? This is where the Court appears to be returning to a form of multifactor analysis.

In resolving the "inability" issue, the Court held that the following factors "indicate that Chevron provides the appropriate legal lens through which to view the legality of the Agency interpretation"¹⁰¹:

the interstitial nature of the legal question, the related expertise of the Agency, the importance of the question to administration of the statute, and the careful consideration the Agency has given the question over a long period of time . . . .¹⁰²

In resolving the "expected to last" issue, the Court stated that several factors led it "to read the statute as delegating to the agency considerable authority to fill in, through interpretation, matters of detail related to its administration."¹⁰³ The factors are:

the statute's complexity, the vast number of claims that it engenders, and the consequent need for agency expertise and administrative experience.¹⁰⁴

It is important to emphasize that the Court employs these factors to determine the applicability of Chevron deference at Step One. In these passages, the Court is not using these factors to evaluate the reasonableness of the agency's interpretation under Step Two.

Of these seven factors, five seem relevant to a determination of whether Congress intended to delegate "force of law" interpretive authority to the agency. The first factor, "the interstitial nature of the legal question," suggests just the sort of gap-filling that has long been a justification for deference. The second and seventh involve the existence or need for agency expertise. It is logical to think that the U.S. Congress might be more likely to delegate interpretive authority where those factors are present. It is also logical to think that Congress might be more likely to delegate authority in light of the sixth factor, the "vast number of claims" that the statute engenders. Better to have the agency deal with large numbers of issues than to have them flood the courts.¹⁰⁵ The fifth factor, the statute's complexity, might also make a congressional delegation more likely.

It is not clear how the third factor would relate to the likely existence of a delegation of lawmaking authority. This factor is "the importance of the question to administration of the statute." In Barnhart, the Court seems to think that the question is important to administration of the statute,¹⁰⁶ and therefore a reason to apply Chevron deference. But in Food & Drug Administration v. Brown & Williamson Tobacco Corp.,¹⁰⁷ the Court appeared to believe that the considerable importance of the policy question at issue in that case suggested that it had not been delegated to the agency.¹⁰⁸ The "importance of the question" might be a good reason to delegate it to the agency, but it also might be a good reason to suspect in some cases that Congress did not intend to delegate its resolution to the agency.

Finally, the fourth factor will often not be relevant to a determination of whether Congress intended a delegation to the agency. The fourth factor is "the careful consideration the Agency has given the question over a long period of time . . . ."¹⁰⁹ On its face, this is precisely the sort of factor that tended to justify deference in the pre-Chevron era, in which agency expertise was a primary touchstone of deference to administrative interpretations.¹¹⁰ Indeed, it is very close to what Skidmore identifies as indicators of a "power to persuade, if lacking power to control."¹¹¹

[32 ELR 11467]

Although this factor seems out of place when the question is whether Congress is likely to have delegated lawmaking authority to the agency, it is possible to draw a connection. Assume that the agency carefully considered an issue for a long period of time. If, knowing of this consideration, Congress reenacted or otherwise confirmed agency authority in the area, the factor might support a finding of delegation to the agency of lawmaking authority with respect to the particular issue. The Court did not explain its understanding of the factor in these terms, however, and the most likely reading is that the Court was simply referring back to its earlier emphasis that the "the Agency's interpretation [was] one of long standing."¹¹²

In light of the complexities of this multifactor analysis, Justice Scalia appears to have been justified in his concern that Mead had adopted a formless "totality of the circumstances" test to determine the availability of Chevron deference.¹¹³ Indeed, Justice Scalia may well be wondering whether there is much left of the distinctive Chevron framework.

2. Utah—A Useful Example of When Chevron Deference Applies—or More Confusion?

In carrying out the 2000 census, the U.S. Census Bureau chose to use a technique called "imputation" to help fill in information for certain properties. After making several attempts to obtain information about the residents of certain properties, the Census Bureau would "impute" to the unknown property the population or other relevant characteristics of the closest dwelling of the same type that did not return the census form.¹¹⁴ This practice apparently made enough difference for North Carolina to have one more congressional representative than it would have had if imputation had not been used, and for Utah to have one fewer.¹¹⁵

Utah challenged the Census Bureau's actions on the ground that "imputation" was a form of "sampling," which was specifically prohibited "for the determination of population for purposes of apportionment of Representatives in Congress among the several States."¹¹⁶ The Court, by a 5-4 vote, rejected Utah's challenge.¹¹⁷

Writing for the majority, Justice Stephen Breyer plunged into a fairly standard statutory analysis, with no reference to deference. First, Justice Breyer used the specific statutory language to conclude that Congress intended to use the word "sampling" as a "term of art with a technical meaning."¹¹⁸ Second, he applied both the technical literature and expert testimony concerning sampling to the imputation method used by the Census Bureau. He found the two to be distinctly different, essentially because sampling seeks to use a small population to represent and provide conclusions about a larger population, while imputation seeks to fill in information about specific parts of a population.¹¹⁹ He then relied upon that difference in distinguishing "imputation" from the "sampling" methods that had been found impermissible in a previous decision, U.S. Department of Commerce v. House of Representatives.¹²⁰

To this point, from all appearances, the involvement of an administrative agency had no bearing on the majority's analysis. Under Chevron Step One, the first step "always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress."¹²¹ Thus, if Congress has clearly decided the issue, the involvement, if any, of an administrative agency may be irrelevant.¹²²

Had Justice Breyer stopped at that point, his opinion would have been somewhat unusual in failing to make explicit account of the agency's interpretation, but it would have been essentially unremarkable in terms of Chevron analysis. He continued, however, by noting that Utah had to overcome "one further legal hurdle," the Census Bureau's "own interpretation of the statute."¹²³ In discussing the significance of the Census Bureau's interpretation, Justice Breyer emphasized two points: (1) the history of the interpretation; and (2) the extent of the statutory delegation.

There are three elements to the Court's discussion of the history of the interpretation. First, the Census Bureau had recommended the statute to Congress. Second, the Census Bureau had "consistently and for many years, interpreted the statute as permitting imputation." Third, Congress was aware of the Census Bureau's interpretation and "enacted related legislation without changing the statute."¹²⁴

Having indicated that these elements of the interpretation's history are somehow relevant to determining the availability of Chevron deference, Justice Breyer then added the second point, the proposition that "the statute itself delegates to the Secretary the authority to conduct the decennial census 'in such form and content as he may determine.'"¹²⁵ Finally, he said, "although we do not rely on it here, under these circumstances we would grant legal deference to the [Census] Bureau's own legal conclusion were [32 ELR 11468] that deference to make the difference."¹²⁶ This paragraph raises a number of interesting questions.¹²⁷ For the purpose of this discussion, the important point is that the majority recognized that Chevron deference applies to Census Bureau interpretations of this sort. Justice Sandra Day O'Connor disagreed, noting that the majority does not "indicate where the [Census] Bureau has provided an interpretation of § 195 that would have the force of law on this issue."¹²⁸

It is not clear whether Justice O'Connor is arguing that the Census Bureau does not have the authority to issue an interpretation with the force of law, or whether, assuming it had such authority, it had not exercised that authority in the appropriate way. Assuming both points to be at issue, what guidance would we find in the majority's opinion?

Although Justice O'Connor cited Christensen¹²⁹ for the proposition that an agency's interpretation must have the "force of law" to be entitled to Chevron deference, Mead has more fully articulated the analytical framework. To determine whether the Census Bureau's interpretation is entitled to Chevron deference, we must ask (1) whether Congress had delegated to the Census Bureau the authority to act with the force of law, and (2) whether the Census Bureau promulgated its interpretation in the exercise of that authority.¹³⁰ In Mead, the Court left open the possibility that Chevron deference might be available in the absence of the delegation or exercise of the authority to act through legislative rulemaking or formal adjudication.¹³¹ The question would be whether Congress intended to delegate the authority to issue a "force of law" interpretation under the particular circumstances.¹³²

If the Census Bureau is to qualify for Chevron deference, it must do so under this general congressional intent aspect of Mead analysis. Although the Secretary of Commerce has been granted the authority to "issue such rules and regulations as he deems necessary to carry out his functions and duties,"¹³³ the Court does not mention this provision. Nor does the Court mention any regulation or other formal issuance by which the Census Bureau's interpretation might have been promulgated.¹³⁴ Under these circumstances, the majority's willingness to grant Chevron deference appears to derive from (1) the Secretary's statutory authority to conduct the census "in such form and content as he may determine,"¹³⁵ and (2) the combination of (a) the Census Bureau's having recommended the statute, (b) the fact that the Secretary's practice in implementing the census had, for many years, included the use of imputation, and (c) Congress' failure to make any changes despite awareness of the agency's practice.¹³⁶

The latter three elements of the Court's analysis sound remarkably like the factors used to support deference in the days before Chevron.¹³⁷ It is difficult to relate them all to the proposition that Congress intended to delegate interpretive authority to the agency. As to the first, if Congress accepted an agency's legislative recommendation, it might be more likely than not that Congress intended the delegation. But this would provide little help in distinguishing among agency statutes. As to the second, the mere fact that an agency has taken a consistent position for many years does not support a congressional delegation of authority. Presumably Congress enacted the relevant provision before the agency began taking the interpretive position, so even long-term consistency itself cannot support the existence of a congressional delegation. Only when we add the fact that Congress was aware of and apparently accepted a long-standing agency interpretation do we even reach the neighborhood of support for the existence of a delegation of lawmaking authority. It is arguable that the congressional inaction reflected congressional satisfaction with the agency's exercise of its powers, and thus an acknowledgement of a delegation of authority to the agency. It seems more likely, however, that congressional inaction would signify acceptance of the particular interpretation, not of a delegation of interpretive authority.

The difficulty of relating these elements of the Court's analysis to the issue of delegated authority suggests the Court's discussion may indicate a deviation from the clear two-step Chevron framework. The Court's language suggests that it would grant Chevron deference where an agency had been given broad authority to implement a particular program and the agency had consistently followed a particular statutory interpretation for several years in implementing the statute, perhaps with congressional acquiescence. In terms of Mead's analytical framework, the statutory grant establishes the necessary delegation of authority, while the consistent implementation constitutes the Census Bureau's "promulgation" of the interpretation.¹³⁸ If this is a correct understand of what the Court has done, the "consistency" element of pre-Chevron analysis has crept back into the analysis of whether Chevron deference is available for particular interpretations.

Although it is possible to deconstruct the elements of Utah's deference analysis as suggested above, perhaps the best way to understand the decision is to view it as a companion to NationsBank of North Carolina, N.A. v. Variable Annuity Life Insurance Co.¹³⁹ The Mead Court cited NationsBank in support of the proposition that "administrative formality" is not required to support the availability of [32 ELR 11469] Chevron deference.¹⁴⁰ The Mead Court relied upon the following language from NationsBank: "The Comptroller of the Currency is charged with the enforcement of banking laws to an extent that warrants the invocation of [the rule of deference] with respect to his deliberative conclusions as to the meaning of these laws."¹⁴¹ The Mead Court also noted that the Comptroller has been characterized as having been given "personal authority" under the National Bank Act.¹⁴²

The point of these passages from NationsBank appears to be that sometimes there is such a strong delegation to a specific official that Chevron deference is due to that official's interpretations despite the absence of relatively formal administrative procedures. In Utah, the Secretary's authority to "take a decennial census . . . in such form and content as he may determine"¹⁴³ may simply be such a strong delegation.

In sum, Utah suggests two possible approaches to determining the availability of Chevron deference in the absence of relatively formal administrative procedures. The first would consider the statutory delegation in connection with such factors as consistency of the agency's interpretation and congressional acquiescence. The second, which seems more likely, would consider the apparent strength of the delegation coupled with the history of the agency's implementation of the delegation.

3. Edelman-Chevron Deference to Procedural Rules Even Without Notice and Comment?

One if the continuing Chevron issues identified by Professor Merrill and Ms. Hickman is whether Chevron deference applies "when a Chevron-qualified agency renders an interpretation in a procedural rule exempt from the notice-and-comment requirements of APA [§] 553?"¹⁴⁴ Edelman¹⁴⁵ offered the Court an opportunity to give a clear answer. Not only did Justice David Souter's majority opinion fail to provide clarity on that issue, it also discussed deference in a way that raises questions about the continuing significance of Chevron deference as a distinct concept.¹⁴⁶

Edelman involved the validity of a regulation issued by the Equal Employment Opportunity Commission (EEOC) concerning the requirements for filing a timely discrimination complaint. Under Title VII of the Civil Rights Act of 1964, a complainant may not file a discrimination action in federal court unless he has first filed a "charge" with the EEOC and has sworn or affirmed that the allegations in the charge are true.¹⁴⁷ On the facts of Edelman, the statute required that the charge be filed within 300 days of the alleged illegal action.¹⁴⁸

The interpretive issue in Edelman was whether such a filing could be timely if the substance of the charge was filed within the statutory deadline, but the required affirmation was not filed until after the deadline had passed. The EEOC had directly addressed this question in a regulation providing that a "charge is sufficient" as long as it adequately identifies the parties and actions complained of. The regulation specifically provided that a later verification would relate back to the date the charge was originally filed.¹⁴⁹ Edelman's filing fell squarely within the terms of the regulation. Lynchburg College argued that the late filing of the verification rendered the entire charge out of time.

Justice Souter upheld the EEOC interpretation in a somewhat peculiar opinion. First, he analyzed the statute to the point of concluding that it was "open to interpretation," so that "the regulation addresses a legitimate question."¹⁵⁰ Second, he held that the regulation was procedural in nature and thus within the EEOC's authority to adopt "'suitable procedural regulations.'"¹⁵¹ Third, Justice Souter briefly addressed the matter of deference, which he considered "insignificant in this case" because "we so clearly agree with the EEOC [that] there is no occasion to defer."¹⁵² Having thus dismissed the need to consider deference, Justice Souter returned to statutory analysis, ultimately concluding that "the EEOC's relation-back regulation [is] an unassailable interpretation of" the statute.¹⁵³

It is odd that Justice Souter dismissed the need to consider deference after having clearly concluded that the statute was ambiguous. This is doubly odd because he later notes that his interpretation "does not mean to say that the EEOC's position is the 'only one permissible,'" and that the EEOC could have adopted a different interpretation.¹⁵⁴ Indeed, Justice O'Connor, concurring in the judgment, doubts "that it is possible to reserve [the question of Chevron deference] while simultaneously maintaining . . . that the agency is free to change its interpretation."¹⁵⁵ This exchange raises significant questions about the nature of deference and the nature of judicial interpretations of agency statutes.¹⁵⁶ For the purpose of this discussion, however, the question is whether Edelman provides any guidance concerning the question of when Chevron deference applies to agency interpretations.

Although Justice Souter's opinion disclaims any decision on deference issues, it seems to suggest that a statutorily authorized procedural rule would be entitled to Chevron deference even if it had not gone through notice-and-comment rulemaking. Justice Souter's first point in discussing deference is to agree with the government's assertion that Chevron deference "does not necessarily require an agency's exercise of express notice-and-comment rulemaking power."¹⁵⁷ In a footnote to this statement, Justice Souter recognizes [32 ELR 11470] that neither Title VII nor the APA requires the use of notice-and-comment rulemaking to issue procedural rules.¹⁵⁸ Since neither of these statements is necessary to Justice Souter's conclusions, it is not clear what purpose they serve. One possibility is that they could be indications that the Court would recognize Chevron deference for interpretations embodied in procedural rules issued without notice-and-comment rulemaking.

Unlike the majority, Justice O'Connor, joined by Justice Scalia, squarely concluded that "the EEOC's regulation is entitled to Chevron deference."¹⁵⁹ In doing so, she cited Mead for the proposition that "notice and comment is 'significant . . . in pointing to Chevron authority,'" and that the vast majority of the Court's cases granting Chevron deference have involved interpretations arising from notice-and-comment rulemaking proceedings.¹⁶⁰ She noted that although the EEOC had originally issued the regulation without going through notice-and-comment rulemaking, it had ultimately "repromulgated [the regulation] pursuant to those procedures."¹⁶¹ In her view, the regulation was equal in status to those originally promulgated through notice and comment. The fact that the particular rule had been repromulgated through notice and comment allowed her to avoid the question of whether Chevron deference would be due to a procedural rule issued under the APA's exception to the notice-and-comment requirement.

It is unfortunate that the Court did not clarify the status of procedural rules. The appropriate conclusion would grant Chevron deference to statutorily authorized procedural rules regardless of the procedures used in their adoption. This would be consistent with the proposition that the fundamental question is not whether the agency has followed any particular procedures, but whether Congress has delegated lawmaking authority to the agency with respect to issue in question.¹⁶² Mead requires that two tests be met in order for an interpretation to qualify for Chevron deference. First, there must be a congressional delegation of interpretive authority. Second, the agency must have issued the interpretation in the exercise of that authority.¹⁶³ Assuming a delegation of the authority to issue procedural rules, the second test is met whenever the agency issues such a rule. In light of the APA's provision exempting procedural rules from notice and comment, Congress would presumably be aware that a delegation of procedural rulemaking authority would be a delegation to exercise any related interpretive authority without going through notice-and-comment rulemaking. Thus, in Mead's terms, an agency issuing a procedural rule without going through notice-and-comment would be "promulgating [the rule] in the exercise of [its] authority."¹⁶⁴

Moreover, it makes little sense to suggest that notice-and-comment rulemaking is "significant . . . in pointing to Chevron authority"¹⁶⁵ if, in granting interpretive authority, Congress did not know whether or not the agency would use notice and comment to issue interpretations pursuant to that authority. In the case of procedural rules, the existence of the delegation must be determined without regard to the possibility that the agency might use notice and comment to issue some of its interpretations. We must assume that Congress was aware that any procedural rules could be issued without notice and comment. Nothing in the APA or elsewhere suggests that Congress would have intended to grant deference in instances where the agency, on its own initiative, decided to employ notice and comment where it was not otherwise required. Congress either granted interpretive authority with respect to the procedural rules at issue, or it did not. If it did, all procedural rules should be entitled to Chevron deference regardless of the procedures used in their adoption.¹⁶⁶

The closest thing to clear guidance that Edelman provides on the availability of Chevron deference is Justice O'Connor's concurring statement that if a rule has been issued through notice-and-comment rulemaking, it does not lose Chevron status merely because it had previously been issued without notice and comment. Given his disclaimer of any decision about deference, Justice Souter's opinion offers only a slight hint that the Court might grant Chevron deference to interpretations embodied in procedural rules that were not subjected to notice-and-comment rulemaking.

4. Hoffman Plastic Compounds—The Role of Chevron Deference Where Statutes Conflict

One other decision from the Court's 2001 Term may provide guidance concerning the question of when Chevron deference applies to an agency interpretation. In Hoffman Plastic Compounds,¹⁶⁷ the National Labor Relations Board (NLRB) construed the scope of its remedial authority under the National Labor Relations Act (NLRA) in light of relevant provisions of the Immigration Reform and Control Act of 1986 (IRCA). In rejecting the NLRB's decision, the majority opinion suggests that Chevron deference is either unavailable or somehow limited when an agency construes its own substantive statute in light of a statute delegated to another agency.

Hoffman Plastic Compounds laid the groundwork for the Court's decision when it hired Jose Castro, an alien who appeared [32 ELR 11471] to have proper authorization to work in the United States. Several months later, a union began an organizing campaign. Castro supported that effort. The company soon fired all employees involved in the organizing campaign, including Castro. The NLRB initiated an unfair labor practice proceeding that resulted, among other things, in an order to offer "reinstatement and backpay" to four employees, including Castro.¹⁶⁸

When an administrative law judge (ALJ) held a hearing to determine the backpay owed to each employee, Castro admitted that he used fraudulent documents to obtain his position. With no record evidence to show that Castro had obtained legal authorization to work in the United States, the ALJ denied backpay or reinstatement for Castro. He did so on the ground that such an award would conflict with the IRCA, a "comprehensive scheme prohibiting the employment of illegal aliens in the United States."¹⁶⁹ The NLRB reversed, ordering backpay for Castro. The appellate court upheld the NLRB's position.¹⁷⁰

Relying upon its own prior decisions and its understanding of the purposes and requirements of the IRCA, the Court reversed.¹⁷¹ Although the Court never mentioned Chevron, its language suggests that the deference that is usually available to an agency interpreting its own statute may not be available where the interpretation relates to the provisions or policies of another statute outside the purview of the interpreting agency. Referring to its prior decisions regarding the intersection of the NLRA and immigration law, the Court said that those decisions "established that where the [NLRB's] chosen remedy trenches upon a federal statute or policy outside the [NLRB's] competence to administer, the [NLRB's] remedy may be required to yield."¹⁷²

In light of the conflict between the backpay award and the "policies underlying ICRA," the Court held that "the award lies beyond the bounds of the [NLRB's] remedial discretion." Although the Court's opinion reads as an assessment of the NLRB's remedial discretion, rather than as an assessment of a statutory interpretation, the two are inseparable in this situation. The NLRB had, in effect, interpreted the NLRA to authorize the backpay award despite the provisions of the IRCA.

In dissent, Justice Breyer argued that the backpay award did not conflict with either national immigration policy or the Court's prior decisions.¹⁷³ He then turned to the familiar territory of deference, arguing that "the law requires the Court to respect the [NLRB's] conclusion, rather than to substitute its own independent view of the matter for that of the [NLRB]."¹⁷⁴ Emphasizing the NLRB's careful consideration of both labor and immigration law, Justice Breyer noted that the Attorney General, who is ultimately responsible for enforcement of the immigration laws, agreed with the NLRB's position. Citing Chevron in his parting shot, Justice Breyer contended that "the [NLRB's] position is, at the least, a reasonable one. Consequently, it is lawful."¹⁷⁵

The majority did not expressly articulate a principle that Chevron deference is not available where an agency interprets an issue that is related to another statute. This proposition could be avoided by considering Hoffman Plastic Compounds to involve on the exercise of remedial discretion, not the exercise of interpretive authority. It could also be avoided by considering the case to involve highly specialized questions arising under labor and immigration law, so that it is not broadly applicable to the generality of deference jurisprudence. Still, the fact that the dissent specifically argued for Chevron deference can be used to argue that the Court rejected Chevron deference in these circumstances.

B. Does It Still Matter Whether Chevron Applies?

Perhaps Chevron's greatest contribution is its recognition of the distinct roles of agencies and courts with respect to the implementation of statutes in the administrative state. Justice John Paul Stevens wrote that the lower court had erred by adopting "a static judicial definition" of the ambiguous statutory term.¹⁷⁶ He went on to describe the distinctive Chevron framework, under which statutory ambiguity constitutes an implicit delegation of policymaking authority to the agency.¹⁷⁷ Whatever a court might think of the statute, if it finds ambiguity it must accept the agency's decision as long as it is reasonable. This limited judicial role is dictated not only by the delegation itself, but also by the very nature of courts and agencies. By contrast to courts, agencies both possess substantive expertise in the relevant area and are part of a "political branch of the Government," accountable to the people through the elected President.¹⁷⁸

Scholars have lavished attention on Chevron's impact on the nature of deference where the statute is ambiguous.¹⁷⁹ As a matter of theory, agencies qualifying for Chevron deference now seem strong candidates for interpretive authority, more clearly recognized as legitimate policymakers under a congressional delegation. As a matter of fact, Chevron may have resulted generally in more widespread judicial acceptance of agency positions.¹⁸⁰ Certainly when agencies are able to convince courts to review their decisions under [32 ELR 11472] the reasonableness standard of Chevron Step Two, agencies have been overwhelmingly successful in defending their policy choices in implementing their statutes.¹⁸¹

Chevron's significance depends upon this distinctive framework. In applying Step One, a court determines whether the policy decision has been delegated to the agency.¹⁸² This assures that if the analysis reaches Step Two, the court is fully aware that the agency, not the court, is the congressionally authorized decisionmaker. Hence, the court should examine the agency's position with the respect—the deference, if you will—that is due to the body assigned to make the decision. Under Chevron's framework, this is not merely a question of comparative expertise in the substantive area or in the interpretation of statutes.¹⁸³ It is not, in Skidmore's terms, a question of whether the various aspects of the agency's analysis render its position sufficiently persuasive to convince a court that might otherwise not be convinced.¹⁸⁴ This is a question of power and political legitimacy.

Much less attention has been paid to the role of the court as a decisionmaker with respect to an ambiguous administrative statute where Chevron does not apply.¹⁸⁵ On one level, the role of the court is simple. If the court determines that the statute is not ambiguous, that is the end of the matter. To use Justice Stevens' language, the result is a "static judicial definition,"¹⁸⁶ presumably not to be changed in the absence of congressional action.

But what is the court's role if the statute is ambiguous? In addressing this question, it is important to emphasize that statutory ambiguity always creates a policy space. Someone, agency or court, must make the policy choice. If Chevron applies, the agency holds the decisionmaking authority. Its job is not to decide what the law is, but to make the delegated policy choice. The court's role is limited to a review for reasonableness and a remand to the agency if the agency's interpretation fails that test. This is appropriate in that the agency is a legitimate political actor ultimately subject to political and electoral review. By the same token, as Chevron itself recognized,¹⁸⁷ statutory ambiguity permits an agency to change its policy choice as long as its new position is reasonable.

If Chevron does not apply, however, the decisionmaking authority lies in the court. Once again, the practical reality is that statutory ambiguity has inherently created a policy space. But in the absence of Chevron deference there is no legitimate political actor available to make the choice. At this point, it is the court's function to determine, as well as it possibly can, what the law is under the relevant statutory language. In effect, the court is the decisionmaker of last resort, not a politically appropriate policymaker, but the only body available under the circumstances. Once the court has made that decision, it is presumably bound by principles of stare decisis.¹⁸⁸ Unlike the agency, it is no longer free to remake its policy choice. In theory, at least, the court did not make a policy choice, but a judgment about the ultimate question of what the statute dictates.

This circumstance makes it vitally important to maintain a clear distinction between the circumstances in which Chevron applies and those in which it does not. It is also important to maintain the distinction between the two types of deference. Under Chevron deference, the agency is the ultimate decisionmaker, with all that implies. Under Skidmore deference, the court is the decisionmaker, subject only to persuasion by the agency. If we conflate the two, we risk weakening agency authority as a political actor and increasing the judiciary's ability to intrude in the political arena.

Unfortunately, two decisions from the 2001 Term join earlier indications to suggest that the Court may blur the distinctions between Chevron and Skidmore deference. Language in at least four of the Court's opinions suggests that Chevron merely provided an additional justification for judicial deference to agency interpretations. If fully realized, this proposition threatens to undermine Chevron's sharp distinction between administrative and judicial authority in statutory interpretations.

1. Early Indications—Christensen and Mead

In Christensen, Justice Breyer in a dissenting opinion agreed with Justice Scalia that the agency's informal statement might be sufficiently authoritative to qualify for Chevron deference, but he did "not object to the majority's citing Skidmore, instead."¹⁸⁹ Disputing Justice Scalia's assertion that Skidmore deference was an anachronism, Justice Breyer briefly reviewed Skidmore's proposition that various factors may give an agency's interpretations "the 'power to persuade,' even where they lack the 'power to control.'"¹⁹⁰ He then said that "Chevron made no relevant change. It simply focused upon an additional, separate legal reason for deferring to certain agency determinations, namely, that Congress had delegated to the agency the legal authority to make those determinations."¹⁹¹

Taken out of context, this statement is jarring. It can be read as envisioning a regime in which the pre-Chevron multifactor approach to deference was simply supplemented by Chevron's addition of another factor, the recognition of a delegation of interpretive authority. This reading is particularly suggested by Justice Breyer's statement that he did not object to the majority's applying Skidmore where he thought Chevron might apply. He seems to be saying that it does not matter which analysis applies as long as one or the other triggers deference. If it does not matter which analysis triggers deference, presumably the choice makes no difference to the nature of the court's role in deferring to the agency decision. If Chevron dictated a stronger form of deference than Skidmore, presumably Justice Breyer would have objected to applying Skidmore since Justice Breyer believed the agency's interpretation should be upheld. Justice Breyer must have seen little difference between Chevron and Skidmore once it was decided that deference was due to a particular agency interpretation.

[32 ELR 11473]

In the context of Christensen, however, the statement did not seem remarkable. Justice Breyer was defending Skidmore deference against Justice Scalia's argument that it had been fully supplanted by Chevron. He was arguing that even if delegation theory had become the dominant basis for deference, there remained a role for some sort of deference based upon "the experience-based views of expert agencies."¹⁹² This did not necessarily mean that Chevron and Skidmore deference were the same, or that both involve some sort of multifactor balancing. It could have meant only that whatever sort of deference might be applied under Chevron, Skidmore's multifactor persuasiveness approach remained valid where Chevron did not apply.

In Mead,¹⁹³ Justice Souter's majority opinion presented what may be the Court's most extensive discussion of deference doctrine. After noting that express delegations of interpretive authority are particularly binding on the courts, Justice Souter said that, "agencies charged with applying a statute necessarily make all sorts of interpretive choices, and while not all of those choices bind judges to follow them, they certainly may influence courts facing questions the agencies have already answered."¹⁹⁴ He then proceeded to describe deference as follows:

The fair measure of deference to an agency administering its own statute has been understood to vary with circumstances, and courts have looked to the degree of the agency's care, its consistency, formality, and relative expertness, and to the persuasiveness of the agency's position . . . . The approach has produced a spectrum of judicial responses, from great respect at one end, . . . to near indifference at the other, . . . .¹⁹⁵

Justice Souter then quoted Justice Robert Jackson's famous articulation of Skidmore's multifactor test for determining the degree of deference due to an agency decision.

To this point, Justice Souter's opinion can be read as describing deference as essentially the multifactor type of analysis that existed prior to Chevron. Justice Souter then seems to echo Justice Breyer's Christensen dissent when he writes that "since 1984, we have identified a category of interpretive choices distinguished by an additional reason for judicial deference."¹⁹⁶ If we stopped at this point, Justice Souter's opinion might appear to treat Chevron as simply providing another basis for triggering deference, but not dictating a particularly strong form of deference.

Of course, we cannot stop at that point in Mead. Indeed, much of the rest of the majority opinion is devoted to recognizing the distinction between Chevron and Skidmore deference and justifying its refusal to apply Chevron deference. If Mead means anything, surely it stands for the proposition that Chevron and Skidmore both represent distinct and legitimate forms of deference, and that the choice can matter to the outcome of a dispute. Indeed, the opinion closes by emphasizing "that judicial responses to administrative action must continue to differentiate between Chevron and Skidmore."¹⁹⁷ Among other things, this presumably means that when Chevron applies, policymaking authority stays with the agency, while a Skidmore deference decision would be one in which the court determines the ultimate meaning of the statute with persuasive guidance from the agency.

2. Edelman Confuses the Chevron-Skidmore Distinction

If Mead confirmed the significance of the Chevron-Skidmore distinction, what are we to make of Justice Souter's majority opinion in Edelman? As noted above, Edelman upheld an interpretation embodied in a procedural rule issued by the EEOC. The nature or degree of deference due to the agency's interpretation was potentially at issue because the Court had not yet addressed Chevron's application to procedural rules.¹⁹⁸

In Edelman, however, the majority considered the nature or degree of deference to be "insignificant."¹⁹⁹ Although the Court agreed with the government's amicus assertion that express notice-and-comment rulemaking power is not necessary to support Chevron deference, it found "no need to resolve any question of deference here":

We find the EEOC rule not only a reasonable one, but the position we would adopt even if there were no formal rule and we were interpreting the statute from scratch. Because we so clearly agree with the EEOC, there is no occasion to defer and no point in asking what kind of deference, or how much.²⁰⁰

It may seem obvious that a court does not need to decide about the availability and nature of deference where the court clearly agrees with the agency's interpretation. Struggling with deference would seem to be a waste of time.

But this approach is unsatisfactory in the sense that it provides no guidance concerning the agency's authority with respect to the issue at hand. If the Court applies Chevron deference, it is clear that the agency retains the policymaking authority to change its interpretation at some point in the future. If the Court rejects Chevron deference, whether or not it applies Skidmore, decisionmaking authority rests with the Court. Until Edelman, at least, the Court has been clear in holding that its own statutory interpretations are binding on federal agencies.²⁰¹ The courts of appeals have been split on this issue, with some holding firm to stare decisis, and some allowing agencies to adopt interpretations different from those initially reached by the Court.²⁰²

Justice Souter addressed this concern in a footnote by saying that "we, of course, do not mean to say that the EEOC's position is the 'only one permissible.'"²⁰³ He explained that "not all deference is deference under Chevron, . . . and there is no need to resolve deference issues when there is no need for deference."²⁰⁴ Justice Souter appears to be saying that whenever any sort of deference is available, the agency has the policy space to change its interpretation if it has sufficient justification.

Justice O'Connor disagreed. She doubted

[32 ELR 11474]

that it is possible to reserve this question while simultaneously maintaining, as the Court does . . . that the agency is free to change its interpretation. To say that the matter is ambiguous enough to permit agency choice and to suggest that the Court would countenance a different choice is to say that the Court would (because it must) defer to a reasonable agency choice.²⁰⁵

With this position, Justice O'Connor would maintain the distinction between Chevron and Skidmore deference. The distinction is important not only because Chevron seems to require a greater degree of deference, but also because it recognizes the difference between the primacy of administrative authority under Chevron and the primacy of judicial authority when Chevron does not apply.

If Justice Souter is correct in asserting that there may be more than one "permissible" agency interpretation under any form of deference analysis, the result would be to undermine the certainty in non-Chevron judicial interpretations and alter the judicial role from one of presumed law finder to one of roving policymaker. As to the first point, if it is always true that an agency can change its position under some form of deference, agencies will be tempted to take that route, perhaps avoiding the political radar, rather than seeking congressional change of a judicial interpretation. This would be true even where the agency had not been granted a delegation sufficient to trigger Chevron deference.

As to the second point, if the court's role is always to determine the permissibility or reasonableness of an agency position, it will presumably no longer be bound by stare decisis, even where the court was the legally authorized decisionmaker. Presumably the court's previous decisions would have the same status as an agency's previous interpretations. An agency could overcome either one with a sufficient demonstration that its new position was permissible. More significant, a court would have the leeway to reject its own prior interpretation whenever it wanted to accept a new one offered by the agency. Judicial review will have lost its moorings.

3. Utah Compounds the Confusion

In Utah, Justice Breyer undertook an extensive analysis of the statutory question of whether the Census Bureau's practice of "imputation" constituted a form of prohibited "sampling." In so doing, he addressed the meaning of the specific statutory term "sampling," relevant expert opinion, the history of the sampling statute, and its own prior opinion in another decision.²⁰⁶ Throughout his analysis, Justice Breyer gave no indication that he considered the statute to be ambiguous. By all appearances, this is a Chevron Step One analysis.

The picture seems to change, however, when Justice Breyer comes to what he termed "one further legal hurdle that Utah has failed to overcome—the [Census] Bureau's own interpretation of the statute."²⁰⁷ Justice Breyer made two points about the significance of the Census Bureau's interpretation. The first seems to support Justice Breyer's preceding Step One statutory analysis. The second suggests that there might have been a role for deference in the decision.

As to the first point, Justice Breyer noted that the Census Bureau (1) had recommended the statute to Congress, and (2) "had consistently and for many years interpreted the statute as permitting imputation."²⁰⁸ He then remarked that "Congress, aware of this interpretation, has enacted related legislation without changing the statute."²⁰⁹ This discussion appears to support the proposition that the Agency's position was the correct interpretation of the statute. Having recommended the statute, the Agency was presumably particularly knowledgeable about what was intended when the statute was enacted. Justice Breyer treated the longstanding agency interpretation as significant because it was presumably accepted by Congress, again an indication that this was the correct interpretation, not merely an acceptable policy choice.

Having thus seemingly emphasized the definitive nature of his interpretation, Justice Breyer observed that "although we do not rely on it here, under these circumstances we would grant legal deference to the [Census] Bureau's own legal conclusion were that deference to make the difference."²¹⁰ At best, this statement adds considerable ambiguity to what otherwise to be a definitive interpretation under Chevron Step One. At worst, it threatens to undermine the distinctive nature of Chevron deference.

Consider the agency's position in the wake of Utah. Would the agency be able to change its position and adopt an interpretation prohibiting reliance upon imputation? If Justice Breyer's opinion were a true Chevron Step One decision that Congress has spoken on the issue, that would be the end of the story. The Census Bureau would have no authority to adopt a different interpretation.

But Justice Breyer says that "under these circumstances" he would grant "legal deference" if granting deference would matter to the outcome.²¹¹ What are the "circumstances" to which Justice Breyer refers? Under Chevron, as most fully elaborated in Mead, three circumstances support granting Chevron deference: statutory ambiguity, a sufficient delegation of interpretive authority, and agency implementation of the delegated authority.²¹²

There seem to be two possible readings of Justice Breyer's statement. First, he may simply be emphasizing the strength of the majority's view that the statute permits imputation. He is saying, in effect, "we don't have to rely on deference because we think the statute is clear, but even if the statute were ambiguous the agency would prevail because Chevron deference applies and its interpretation is reasonable." If that is the case, the "circumstances" to which Justice Breyer refers are presumably the strength of the delegation of interpretive authority, the agency's long-standing implementation of the interpretation, and the reasonableness of the agency's position under the statute. The purpose of his statement would seem to be to make it clear that the sort of delegation at issue in Utah is sufficient to support Chevron deference.

Unfortunately, the language smacks of another possible meaning. It suggests that a court may pick and choose when to rely upon Chevron. Just as he gave no indication that the [32 ELR 11475] statute was ambiguous, Justice Breyer never stated clearly that the statute's meaning was plain or unambiguous. Having failed to make it clear whether he considered the statute to be ambiguous, Justice Breyer has left open the possibility that his statement means that the courts may resolve the meaning of an ambiguous administrative statute without regard to Chevron. Indeed, Justice Breyer seems to be making a reality of his own prediction, made in 1986, that the seemingly simple Chevron two-step "seems unlikely in the long run, to replace the complex" multifactor approach that preceded it.²¹³

When we combine this confusion with the implications of Edelman, we see courts picking and choosing when to rely upon deference and when to reach their own conclusions. This undermines the distinction between administrative authority and judicial authority. It threatens to inject the judiciary still further into the policy arena. It seems, indeed, to demote Chevron and agency authority. Once a distinctive delegation analysis that must be applied where the statute is ambiguous and the Mead tests are met, Chevron seems to become only another justification for a court, on its own authority, to accept an agency's interpretation.

Conclusion

The Court's 2001 Term signals two developments in the law governing deference to agency statutory interpretation. The first, continuing a trend begun in 2000, increased the complexity of determining whether Chevron deference applies to an agency interpretation. The second raises questions about the continuing viability of Chevron doctrine as a distinctively powerful form of deference.

As to the first trend, where once the Chevron threshold seemed to be a determination of statutory ambiguity, Barnhart and Utah strongly suggest that the Court will at times employ a multifactor analysis in grappling with deference. In particular, both place emphasis on the long-standing nature of an agency's interpretation. Unfortunately, the Court has not clearly articulated the role of a long-standing interpretation in the two-step Chevron analysis. While the long duration of an interpretation may be relevant to either congressional acquiescence (a Chevron Step One question) or to the reasonableness of an agency's interpretation, it should have no bearing on whether Chevron deference should be applied to a particular interpretation. The Court relies upon the point in its general discussion of whether to defer to the interpretation, but it fails to state clearly whether the discussion relates to the question of whether Chevron deference applies to the interpretation, or to the question of whether the interpretation is reasonable. At this point, however, it is clear that agencies should emphasize the long-standing nature of a challenged interpretation. We know the Court may consider the point significant. We're just not sure why.

Although the Court's apparent return to multifactor analysis has muddied the waters, some of its 2001 decisions may have added a little clarity. Utah may have helped clarify deference doctrine somewhat by making it clear that an agency such as the Census Bureau, which has a clear delegation of authority to implement a statute, may qualify for Chevron deference when implementing the statute despite the absence of any relatively formal procedures for reaching the interpretation. Edelman suggests that the Court would defer to a procedural rule that has been issued with notice and comment, but it ultimately leaves that issue open. Hoffman Plastic Compounds indicates that an agency is less likely to receive Chevron deference where it interprets a statutory provision whose meaning is affected by a different statute not delegated to that agency.

Despite appearing at times in dicta, the above trend toward increased complexity seems to be real. The Court clearly articulated several factors and employed those factors in its deference analysis. By contrast, the second development is in its early stages. In Utah and Edelman, the majority finds it unnecessary to resolve the question of whether Chevron deference applies to the interpretations at issue. If the Court is saying that the agency's interpretation is obviously correct, so there is no need to go further, perhaps no harm is done. It is possible for an agency's interpretation to be either obviously correct under Chevron Step One or obviously reasonable under Chevron Step Two. If so, the Court may feel no need to resolve the question of which step governs the outcome. If that is what the Court has done, the only problem is that the decision fails to tell the agency or a later court whether the agency has the discretion to change its position in the future.

The Court's cavalier treatment of deference in these two opinions suggests, however, that a majority of the Court may not view Chevron as dictating a distinctive form of deference, one that recognizes the division of authority between agencies and courts. If Chevron is simply "an additional, separate legal reason for deferring to certain agency determinations,"²¹⁴ it may devolve into merely another justification for a judicial decision, rather than a requirement to respect agency authority.

It is far too early to conclude that Chevron has been demoted. The indications are indirect and indistinct. But they are there, and they may give heart to some lower courts eager to strike down agency positions. It behooves the Court to affirm its adherence to Chevron's recognition of a form of deference grounded in the authority delegated to the agency, rather than in mere respect for expertise or experience.

1. 467 U.S. 837, 14 ELR 20507 (1984).

2. Chevron's dominance is astonishing. As of July 1, 2002, a Westlaw search reveals that Chevron has been cited in 10,995 documents, including 3,959 law review articles. By contrast, its near contemporary, Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 13 ELR 20672 (1983), has been cited in 2,865 documents, including 890 law review articles. Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519, 8 ELR 20288 (1978), a blockbuster in its time, has been cited in 2,151 documents, including 858 law review articles. Even Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 1 ELR 20110 (1971), arguably the primary source of modern administrative law, has been cited in only 5,293 documents, including 1,084 law review articles.

3. Cooley R. Howarth Jr., United States v. Mead Corp.: More Pieces for the Chevron/Skidmore Deference Puzzle, 54 ADMIN. L. REV. 699 (2002). See also David Partlett, Administrative Law Discussion Forum—Introduction, 54 ADMIN. L. REV. 565 (2002).

4. Administrative Law Discussion Forum, 54 ADMIN. L. REV. 565 (2002).

5. 529 U.S. 576 (2000).

6. 533 U.S. 218 (2001).

7. Thomas W. Merrill & Kristin E. Hickman, Chevron's Domain, 89 GEO. L.J. 833 (2001). The Supreme Court cited Chevron's Domain in Mead in elaborating on the proposition that Chevron's applicability is a question of congessional intent. 533 U.S. at 230 n.11. Professor Merrill also submitted a brief amicus curiae in Mead, in which he argued, in essence, that the Court should recognized two distinct types of deference (as embodied in Chevron and in Skidmore v. Swift & Co., 323 U.S. 134 (1944)), and that the availability of Chevron deference is a function of a congressionally intended delegation of authority and the agency's exercise of that authority through the appropriate procedures. Brief for Amicus Thomas W. Merrill at 2-5 (Introduction and Summary of Argument), United States v. Mead, 533 U.S. 218 (2001) (No. 99-1434).

8. Mead could have provided a clear answer to the following question: "6. Does Chevron apply when a Chevron-qualified agency renders an interpretation in an adjudication?" Merrill & Hickman, supra note 7, at 850. The agency was "Chevron-qualified" because it had legislative rulemaking authority. United States v. Haggar Apparel Co., 526 U.S. 380 (1999). Mead arguably involved an adjudication. See William S. Jordan, III, U.S. v. Mead: Complicating the Delegation Dance, 31 ELR 11425, 11428 (Dec. 2001).

Mead might also have answered the following questions "1. Does Chevron apply to an agency that lacks legislative rulemaking authority? (An issue subject to conflict in the circuits.); and 2. Does Chevron apply to an agency that lacks both legislative rulemaking authority and the power to render adjudications having the force of law? (An unresolved issue.)." Merrill & Hickman, supra note 7, at 850. Although these questions were not directly at issue in Mead, one suspects that the Court could have provided clear answers in dicta in an effort to illuminate the Chevron landscape.

9. In chronological order, they are National Cable & Telecommunications Ass'n v. Gulf Power Co., 122 S. Ct. 782 (2002); Barnhart v. Sigmon Coal Co., 122 S. Ct. 941 (2002); New York v. Federal Energy Regulatory Comm'n, 122 S. Ct. 1012 (2002); Edelman v. Lynchburg College, 122 S. Ct. 1145 (2002); Ragsdale v. Wolverine Worldwide, Inc., 122 S. Ct. 1155 (2002); Department of Housing & Urban Dev. v. Rucker, 122 S. Ct. 1230 (2002); Barnhart v. Walton, 122 S. Ct. 1265 (2002); Hoffman Plastic Compounds, Inc. v. National Labor Relations Bd., 122 S. Ct. 1275 (2002); Verizon Communications, Inc. v. Federal Communications Comm'n, 122 S. Ct. 1646 (2002); Chevron, U.S.A., Inc. v. Echazabal, 122 S. Ct. 2045 (2002); National Railroad Passenger Corp. v. Morgan, 122 S. Ct. 2061 (2002); Utah v. Evans, 122 S. Ct. 2191 (2002).

10. These are Securities & Exchange Comm'n v. Zanford, 122 S. Ct. 1899 (2002), and Wisconsin Dep't of Health & Family Servs. v. Blumer, 122 S. Ct. 962 (2002).

11. See infra section II.A.

12. See infra section II.B.

13. It is often difficult to determine precisely whether the Court has upheld the agency under Step One or has held the agency's interpretation to be reasonable under Step Two. The following decisions did one or the other, sometimes both: Gulf Power Co., 122 S. Ct. at 786, 789-89; Sigmon Coal Co., 122 S. Ct. at 965; New York, 122 S. Ct. at 1022, 1028; Ragsdale, 122 S. Ct. at 1162, 1163; Rucker, 122 S. Ct. at 1233, 1236; Verizon Communications, 122 S. Ct. at 1678-79, 1687; Zanford, 122 S. Ct. 1899, 1903; Echazabal, 122 S. Ct. 2045, 2052.

14. 323 U.S. 134 (1944).

15. Blumer, 122 S. Ct. at 976, 980 (Stevens, J., dissenting); Morgan.

16. 122 S. Ct. 1646 (2002).

17. The court in Verizon at some times seemed to be answering an interpretive question, while at other times it seemed to be determining whether the regulation was arbitrary and capricious. Since some commentators argue that Step Two review is essentially the same as arbitrary and capricious review, this has served only to confuse the analysis still further. See William Funk. Court Further Explicates and Applies Chevron/Mead, 27 ADMIN. & REG. L. NEWS 8, 8-9 (2002).

18. See, e.g., William S. Jordan III, Deference Revisited: Politics as a Determinant of Deference Doctrine and the End of the Apparent Chevron Consensus, 68 NEB. L. REV. 454, 458-77 (1989), describing the pre-Chevron history of the deference debate.

19. There were also decisions in which the Court appeared to give no deference to agency interpretations. See, e.g., Packard Motor Car Co. v. National Labor Relations Bd., 330 U.S. 485 (1947), discussed in Jordan, supra note 18, at 465-67.

20. 322 U.S. 111 (1944).

21. Id. at 131.

22. 323 U.S. at 138, 139.

23. Id. at 140.

24. Does this distinction actually make a difference to outcomes? Some research suggests that Chevron, which adopted a Hearst-type approach, may have significantly increased the agency's chances of winning. See, e.g., Peter H. Schuck & E. Donald Elliott, To the Chevron Station: An Empirical Study of Federal Administrative Law, 39 DUKE L.J. 984 (1990). See also William S. Jordan III, Judges, Ideology, and Policy in the Administrative State: Lessons From A Decade of Hard Look Remands of EPA Rules, 53 ADMIN. L. REV. 45, 70 (2001). Others believe that the distinction makes little difference, and that the courts will reach the interpretations they desire regardless of interpretive method. See, e.g., Russell L. Weaver, Some Realism About Chevron, 58 MO. L. REV. 129 (1993). Whether or not the different strands produce different results, the parties and the courts certainly argue about it enough for the matter to be important to litigants.

25. 467 U.S. at 842-43, 14 ELR at 20508-09 (footnotes omitted).

26. Richard J. Pierce Jr., Chevron and Its Aftermath: Judicial Review of Agency Interpretations of Statutory Provisions, 41 VAND. L. REV. 301 (1988); Ronald M. Levin, Judicial Review of Administrative Action in a Conservative Era (panel discussion), 39 ADMIN. L. REV. 353, 360 (1987).

27. See, e.g., Ronald M. Levin, Mead and the Prospective Exercise of Discretion, 54 ADMIN. L. REV. 771, 780-82 (2002), suggesting that factors such as those articulated in Skidmore should have a bearing on questions decided under Chevron Step One.

28. Prof. Russell Weaver uses the term "actual deference." See Russell L. Weaver, The Undervalued Non-Legislative Rule, 54 ADMIN. L. REV. 871, 871 & n.3 (2002). The point is to emphasize the distinction between a court's duty to give respectful consideration to an agency's interpretation under Skidmore, and the court's duty to recognize the agency as the primary authorized decisionmaker under Chevron Step Two.

29. Jordan, supra note 18, at 483-85; Weaver, supra note 24, at 131-38.

30. See, e.g., United States v. Mead Corp., 533 U.S. 218, 227-31 (2001).

31. See, e.g., Kenneth W. Starr, Judicial Review in the Post-Chevron Era, 3 YALE J. ON REG. 283 (1986); Pierce, supra note 26; Levin, supra note 26; Abner J. Mikva, How Should the Courts Treat Administrative Agencies?, 36 AM. U. L. REV. 1 (1986); Steven Breyer, Judicial Review of Questions of Law and Policy, 38 ADMIN. L. REV. 363, 373 (1986).

32. Chevron, 467 U.S. at 842, 14 ELR at 20508.

33. Merrill, Brief for Amicus, supra note 7, at 3, 15; Merrill & Hickman, supra note 7, at 835.

34. It is well established that an agency may generally choose to develop binding policy through formal adjudications, rather than through legislative rulemaking. See, e.g., Securities & Exchange Comm'n v. Chenery Corp., 332 U.S. 194, 202-203 (1947); National Labor Relations Bd. v. Bell Aerospace Co. Div. of Textron, Inc., 416 U.S. 267, 294 (1974).

35. Skidmore, 323 U.S. at 139-40.

36. The APA defines "rule," in relevant part, as "the whole or a part of an agency statement of general or particular applicability and future effect designed to implement, interpret, or prescribe law or policy. . . ." 5 U.S.C § 551(4), available in ELR STAT. Admin. Proc. This definition is not limited to statements that have been adopted through statutorily prescribed rulemaking procedures. It includes virtually any statement the agency might make interpreting a statute. If a statement has been issued through the rulemaking process and has the force of law, it is known as a "legislative rule." If a statement has not been issued through the rulemaking process and does not have the force of law, it is known as a "nonlegislative rule."

37. 529 U.S. at 576. See also, e.g., Equal Employment Opportunity Comm'n v. Arabian American Oil Co., 499 U.S. 244, 256-58 (1991). For several years, Prof. Robert Anthony had been the principal proponent of denying Chevron-style deference to informal agency statements. See, e.g., Robert A. Anthony, Which Agency Interpretations Should Bind Citizens and the Courts?, 7 YALE J. ON REG. 1 (1990).

38. Christensen, 529 U.S. at 587.

39. See William S. Jordan III, Judicial Review of Informal Statutory Interpretations: The Answer Is Chevron Step Two, Not Christensen or Mead, 54 ADMIN. L. REV. 719, 723, 725 (2002).

40. Sometimes such documents are written in language that appears to be binding. Since the documents have not gone through the notice-and-comment rulemaking process, they cannot be legally binding. See William Funk, When Is a "Rule" a Regulation? Marking a Clear Line Between Nonlegislative Rules and Legislative Rules, 54 ADMIN. L. REV. 659 (2002). Nonetheless, the fact that some such issuances appear to be binding has prompted a long, vociferous, and continuing debate over whether such statements are sometimes required to go through the notice and comment before being issued by the agency. See, e.g., Robert A. Anthony, Interpretive Rules. Policy Statements, Guidances, Manuals, and the Like—Should Federal Agencies Use Them to Bind the Public?, 41 DUKE L.J. 1311 (1992), Peter L. Strauss, The Rulemaking Continuum, 41 DUKE L.J. 1463 (1992); Peter L. Strauss, Publication Rules in the Rulemaking Spectrum: Assuring Proper Respect for an Essential Element, 53 ADMIN. L. REV. 803 (2001) (hereinafter Publication Rules). This debate is beyond the scope of this Article.

41. Christensen, 529 U.S. at 587.

42. Skidmore, 323 U.S. at 140 (cited in Christensen, 529 U.S. at 587). The Court had previously reached the same conclusion in Arabian American Oil Co., 499 U.S. at 257-58.

43. Mead, 121 S. Ct. at 2178 (Scalia, J., dissenting).

44. See Michael P. Healy, Spurious Interpretation Redux: Mead and the Shrinking Domain of Statutory Ambiguity, 54 ADMIN. L. REV. 673 (2002); Howarth, supra note 3, Jordan, supra note 39, Ronald J. Krotoszynski Jr., Why Deference?: Implied Delegations and the Misplaced Legacy of Skidmore, 54 ADMIN. L. REV. 735 (2002); Levin, supra note 27; Thomas W. Merrill, The Mead Doctrine: Rules and Standards, Meta-Rules and Meta-Standards, 54 ADMIN. L. REV. 807 (2002); Weaver, supra note 28.

45. Mead is described in detail in Jordan, supra note 8, at 11425-27.

46. Id. at 11427.

47. Id. at 11426 (discussing Mead, 533 U.S. at 226-27).

48. 533 U.S. at 227. Of course, a statute may explicitly delegate to an agency the authority to "elucidate a specific provision of the statute by regulation," Chevron, 467 U.S. at 843-44, 14 ELR at 20509. When that happens, an interpretation issued in the exercise of that authority "is binding in the courts unless procedurally defective, arbitrary or capricious in substance, or manifestly contrary to the statute." Id. at 844, 14 ELR at 20509. The difficult questions of Chevron's applicability arise where there is an argument that the U.S. Congress has implicitly delegated interpretive authority to the agency.

49. 533 U.S. at 230.

50. Id. at 231.

51. NationsBank of N.C., N.A. v. Variable Annuity Life Ins. Co., 513 U.S. 251, 256-57 (1995) (cited in Mead, 533 U.S. at 231 & n.13).

52. 533 U.S. at 241 (Scalia, J., dissenting).

53. See, e.g., Jordan, supra note 8; Jordan, supra note 39; Weaver, supra note 28.

54. Howarth, supra note 3.

55. Krotoszynski, supra note 44.

56. Professor Levin argues that Chevron deference should apply where a statutory interpretation is "rendered by an agency that possesses delegated power to implement the statute in question—regardless of whether the agency actually exercised such power in the case at hand." Levin, supra note 27, at 773. Thus, in his view, the additional Mead requirement that the agency has issued in the exercise of that authority (which Professor Levin calls the "format" doctrine) will shift too many cases from Chevron to Skidmore analysis. Professor Healy argues that if statutory ambiguity is viewed as the touchstone of Chevron doctrine, the result will be that the agency, the more politically accountable body, will have more interpretive power. By contrast, he argues Mead's new requirement of an affirmative showing of congressional intent will mean that interpretive power will shift to the courts, which are not politically accountable. Healy, supra note 44, at 679-81. See also John O. McGinnis, Presidential Review as Constitutional Restoration, 51 DUKE L.J. 901, 951 n.22 (2001), suggesting that Mead may be cutting back on Chevron's applicability to agency statutory interpretations.

57. Healy, supra note 44, at 8-9; Weaver, supra note 28, at 873-74; Russell L. Weaver, The Emperor Has No Clothes Christensen, Mead, and Dual Deference Standards, 54 ADMIN. L. REV. 173, 175-86 (2002).

58. See Jordan, supra note 39, at 725-726; Merrill, supra note 44. Interestingly, Professor Merrill favors reliance upon a theory that Congress has intentionally delegated authority to the agency. Id. at 809. In fact, he considers it important that Congress should have made a "considered judgment" to delegate the interpretive authority. Merrill, Brief for Amicus, supra note 7, at 19. His problem with the Mead formulation is that "the inquiry . . . collapsed into a formless consideration of "factors" of unspecified weight. Merrill, supra note 44, at 809.

59. 467 U.S. at 843-44, 14 ELR at 20508-09.

60. Id. at 866, 14 ELR at 20514.

61. Marbury v. Madison, 5 U.S. (I Cranch) 137, 177 (1803).

62. 122 S. Ct. 1265 (2002).

63. 122 S. Ct. 2191 (2002).

64. 122 S. Ct. 1145 (2002).

65. 122 S. Ct. 1275 (2002).

66. Three other decisions may also be said to involve the question of when Chevron deference applies to an agency's interpretation. The first two reflect straightforward application of principles that are clear in Christensen and Mead. In Wisconsin Dep't of Health & Family Servs., 122 S. Ct. 962, 976 (2002), the Court gave only "respectful consideration," (Skidmore deference) to an agency's informal opinion letter and policy memoranda. In Securities & Exchange Comm'n v. Zandford, 122 S. Ct. 1899, 1903 (2002), the Court granted Chevron deference to an interpretation reached in a formal adjudication, albeit an interpretation more than half a century old. Neither opinion suggests any complications in determining the availability of Chevron deference.

The third decision, National Railroad Passenger Corp., reiterates the proposition that the Equal Employment Opportunity Commission (EEOC) is not entitled to Chevron deference when it interprets Title VII, a point definitively established in Equal Employment Opportunity Comm'n v. Arabian American Oil Co., 499 U.S. 244, 257-58 (1991). Although the Court did not discuss the point, this would tend to confirm the apparent principle that Chevron deference is not available to agencies lacking the authority to make law through either legislative rulemaking or formal adjudication.

67. 122 S. Ct. at 1268.

68. Id. (quoting 42 U.S.C. § 423(d)(1)(A) (1994 ed.) and 42 U.S.C. § 1382c(a)(3)(A) (1994 ed., Supp. V) (emphasis supplied by the court).

69. Id.

70. Id. at 1269.

71. Id. (citing 20 C.F.R. § 404.1520(b) (2001)).

72. Id.

73. Id. (citing 20 C.F.R. §§ 404.1520(b), 1592(d)(2) (2001)).

74. Id. at 1268-69. The Fourth Circuit's ruling on this point may seem absurd because it seems to result in disability payments for a condition "expected to last" 12 months even if the condition actually lasted only a few weeks or days. This apparent absurdity is substantially mitigated by a statutory provision for a five-month waiting period before a claimant is entitled to receive benefits. 122 S. Ct. at 1269 (citing 42 U.S.C. §§ 423(a)(1)(D)(i), 423(a)(1)(D)(ii)).

75. Id. at 1269-70, 1273.

76. Id. at 1270, 1273.

77. Id. at 1274 (Scalia, J., concurring).

78. Id. at 1270.

79. Id. at 1274 (Scalia, J., concurring).

80. Mead, 533 U.S. at 229.

81. It is possible to envision an argument in which the long-standing nature of an agency interpretation contributes to the proposition that the agency intended a delegation of lawmaking authority on the point at issue. Immediately after discussing the long-standing interpretation, the Court noted that Congress had "frequently amended or reenacted the relevant [statutory] provisions without change." 122 S. Ct. at 1270. One could construct an argument to the effect that this constituted a congressional recognition of agency interpretive authority on the matter at issue. The Court does not take this approach, however. Instead, the Court views the various congressional actions as indicating that the agency's interpretation is "statutorily permissible" under Chevron Step Two. Id. at 1270-71.

82. Id. at 1274 (Scalia, J., concurring).

83. For a possible exception to that statement, see the argument at supra note 81.

84. This is consistent with the Court's previous treatment of long-standing interpretations under the Chevron framework. See, e.g., Smiley v. Citibank (South Dakota), N.A., 517 U.S. 735, 741 (1996) ("To be sure, agency interpretations that are of long standing come before us with a certain credential of reasonableness, since it is rare that error would long persist.").

85. 122 S. Ct at 1270.

86. The Court cites North Haven Bd. of Educ. v. Bell, 456 U.S. 512, 522 n.12 (1982), in which the Court said that "in construing a statute, this Court normally accords great deference to the interpretation, particularly when it is longstanding, of the agency charged with the statute's administration." Since the Court made this statement before Chevron's clear distinction between applicability (Step One) and reasonableness analysis (Step Two), it is difficult to tell just what this statement means in the Chevron era. It is at least plausible, however, that the Court meant to import this principle into the reasonableness analysis of Step Two, rather than into the applicability analysis of Step One.

87. 122 S. Ct. at 1274 (Scalia, J., concurring). Justice Scalia appears, at this point, to be addressing the question of whether antiquity relates to whether Chevron deference applies to the interpretations in question (a Step One question), not the question of whether antiquity bears on whether the interpretations are reasonable (a Step Two question).

88. Id. at 1271.

89. Id.

90. Chevron, 467 U.S. at 863-64, 14 ELR at 20514.

91. 122 S. Ct. at 1271.

92. Id. (citing Chevron, 467 U.S. at 843, 14 ELR at 20509).

93. Id.

94. Id. (emphasis supplied).

95. Indeed, presumably all interpretations that have been through notice and comment were once informal statements, in the proposed rule if nowhere else.

96. Referring to its statement that prior informal status does not deprive an interpretation of Chevron deference, the Court said, "if this Court's opinion in Christensen v. Harris County suggested an absolute rule to the contrary, our later opinion in United States v. Mead Corp. denied the suggestion . . . . The want of [notice-and-comment] does not decide the case." 122 S. Ct. at 1271-72 (citations omitted).

97. Mead, 533 U.S. at 230.

98. 122 S. Ct. at 1272.

99. Id.

100. At least one court of appeals reached this conclusion with respect to certain rules issued by the U.S. Environmental Protection Agency. Hall v. EPA, 263 F.3d 926, 32 ELR 20083 (9th Cir. 2001).

101. 122 S. Ct. at 1272 (emphasis supplied). It is important to emphasize that the Court seems to employ these factors to determine the applicability of Chevron deference at Step One. In this passage, it is not using these factors to evaluate the reasonableness of the agency's interpretation under Step Two.

102. Id.

103. 122 S. Ct. at 1273 (emphasis supplied). Again, the highlighted language demonstrates that the Court considered these factors in making the Step One decisions about whether Chevron deference could be applied at all, not in making the Step Two decision about whether the interpretation was reasonable.

104. Id.

105. But the very same factor seemed to cut against a delegation of lawmaking authority in Mead. There, the Court found such a delegation unlikely where agency issued 10,000-15,000 letter rulings a year at various levels and offices within the agency. Mead, 533 U.S. at 233-34.

106. 122 S. Ct. at 1272.

107. 529 U.S. 120 (2000).

108. The question in Brown & Williamson was whether the agency had the authority to regulate nicotine in cigarettes. With respect to the significance of the importance of the issue, the Court said, "in addition, we must be guided to a degree by common sense as to the manner in which Congress is likely to delegate a policy decision of such economic and political magnitude to an administrative agency." 529 U.S. at 133. The Court held that Congress had not delegated the relevant authority to the agency. This would seem to cut against the proposition that Congress intended to delegate an important issue to an agency where, for example, Congress had not granted the agency explicit rulemaking or adjudicatory authority with respect to the particular issue.

109. 122 S. Ct. at 1272.

110. For a comprehensive list of pre-Chevron factors, see Colin S. Diver, Statutory Interpretation in the Administrative State, 133 U. PA. L. REV. 549, 562 n.95 (1985).

111. 323 U.S. at 140.

112. 122 S. Ct. at 1272, 1271.

113. 533 U.S. at 241 (Scalia, J., dissenting).

114. Utah, 122 S. Ct. at 2196.

115. Id.

116. Id. at 2199 (quoting 13 U.S.C. § 195). Utah also challenged the practice as a violation of the "actual Enumeration" language of the Census Clause, Art. I, § 2, cl. 3.

117. Justice Sandra Day O'Connor dissented with respect to the majority's conclusion that the statute authorized the use of imputation. 122 S. Ct. at 2207-12. Justices Clarence Thomas and Anthony Kennedy agreed with the majority's statutory interpretation, but they dissented as to the constitutionality of the practice. Id. at 2212-23. Justice Scalia dissented on the ground that Utah lacked standing to challenge the census. Id. at 2223-26.

118. The statutory prohibition refers to "the statistical method known as 'sampling.'" According to Justice Breyer the phrase "known as" and the quotation marks around "sampling" suggested a technical meaning. Id. at 2201.

119. Id. at 2201-02.

120. 525 U.S. 316 (1999).

121. Chevron, 467 U.S. at 842-43, 14 ELR at 20508.

122. Although the particular Chevron language quoted here suggests that agency involvement may be irrelevant at Chevron Step One, Chevron's note 9 at least opens the door to considering the agency's interpretation in deciding whether Congress has spoken clearly to the issue at hand: "If a court, employing traditional tools of statutory construction, ascertains that Congress had an intention on the precise question at issue, that intention is the law and must be given effect." Chevron, 467 U.S. at 843 n.9, 14 ELR at 20508-09 n.9. Prof. Peter Strauss has argued, for example, that Skidmore-type factors are among the "traditional tools of statutory construction" that must be considered at Chevron Step One. Strauss, supra note 40, at 835-36.

123. Utah, 122 S. Ct. at 2203.

124. Id.

125. Id.

126. Id. at 2203 (citing Chevron).

127. See infra section II.B.3.

128. 122 S. Ct. at 2211 (O'Connor, J., dissenting).

129. 529 U.S. at 587.

130. 533 U.S. at 226-27. See also Jordan, supra note 8, at 11427, suggesting that Mead transformed Chevron from a two-step to a four-step.

131. 533 U.S. at 231.

132. Id. at 227.

133. 13 U.S.C. § 4 (1990).

134. The Census Bureau's regulations appear at 15 C.F.R. pts. 30-101. They do not appear to include the interpretation at issue in Utah.

135. 13 U.S.C. § 141(a) (cited in Utah, 122 S. Ct. at 2203). The Secretary's rulemaking authority could also be a contributing factor, but there is no indication in the opinion that the Court was aware of that authority.

136. See supra note 124 and accompanying text.

137. See, e.g., Jordan, supra note 18, at 468-69.

138. This analysis can be reconciled with Christensen because Utah involved review of the agency's implementation of its mandate, while Christensen involved review of a mere agency opinion letter. 529 U.S. at 588. It can be reconciled with Mead because Utah involved a long-standing position taken by the agency in the exercise of its central function, while Mead involved one of many thousands of interpretations reached by the agency each year, one that could be reached by many agency offices throughout the country. 533 U.S. at 233-34.

139. 513 U.S. 251 (1995).

140. 533 U.S. at 231.

141. Id. at 231 n.13 (internal quotation marks omitted by the Court).

142. Id. at 231.

14313. U.S.C. § 141(a) (1990).

144. Merrill & Hickman, supra note 7, at 850.

145. 122 S. Ct. at 1145.

146. The discussion that follows addresses the significance of Edelman to the question of when Chevron deference may be applied to an agency interpretation. See infra section II.B.2. for discussion of the impact of Edelman on the continuing vitality of Chevron deference as we have known it to date.

147. Edelman, 122 S. Ct. at 1147 (citing 42 U.S.C. § 2000e-5(e)(1), 2000e-5(b)).

148. The statute requires that the charge be filed within 180 days of the illegal act unless the complainant has initiated proceedings in an appropriate state or local agency, in which case the filing deadline is 300 days from the alleged illegal act. 42 U.S.C. § 2000e-5(e)(1). The 300-day period applied in Edelman. 122 S. Ct. at 1147.

149. 29 C.F.R. § 1601.12(b) (1997).

150. 122 S. Ct. at 1149.

151. Although the EEOC is authorized to adopt procedural regulations, it has no power to adopt substantive regulations under Title VII. Id. at 1149-50.

152. Id.

153. Id. at 1150-52.

154. Id. at 1150 n.8.

155. Id. at 1153 (O'Connor, J., concurring).

156. See infra section II.B.2.

157. 122 S. Ct. at 1250.

158. Id. at 1150 n.7.

159. Id. at 1154 (O'Connor, J., concurring). Although he did not mention Chevron by name, Justice Thomas also found the EEOC regulation entitled to Chevron deference. In so doing, he stated only that the EEOC had promulgated the regulation "pursuant to its clear statutory authority to issue procedural regulations." Id. at 1153 (Thomas, J., concurring).

160. Id. at 1155 (O'Connor, J., concurring).

161. Id.

162. Mead, 533 U.S. at 230.

163. Id. at 226-27.

164. Id.

165. Id. at 1155 (O'Connor, J., concurring) (quoting Mead, 533 U.S. at 230-31).

166. One could make a similar argument about "interpretative" rules. Since Congress knows that "interpretative" rules are exempt from notice and comment, 5 U.S.C. § 553(b)(3)(A), available in ELR STAT. Admin. Proc. Congress must have intended any delegation of interpretive authority to include the authority to issue binding interpretative rules without notice and comment. However, despite the exception for "interpretative rules," substantive rules, including interpretations of substantive statutory provisions, generally do not become binding unless they are promulgated through notice and comment. See, e.g., Anthony, supra note 40, at 1312-15. Thus, when Congress grants substantive interpretive authority, it expects that the interpretations will not become binding unless they are embodied in regulations issued through the rulemaking process, or through implementation of the statute, rather than through a mere statement.

By contrast, procedural rules are binding on the public even if they are not issued through notice and comment. Thus, Congress would expect a delegation of interpretive authority as to procedural matters to be implemented through rules that do not go through notice and comment.

167. 122 S. Ct. at 1275.

168. Id. at 1278-79.

169. The ALJ also relied on a previous decision, Sure-Tan, Inc. v. National Labor Relations Bd., 467 U.S. 883, 903 (1984), in which the Court had held that backpay awards must be limited to aliens "lawfully entitled to be present and employed in the United States." Hoffman Plastics, 122 S. Ct. at 1279, 1282. The Court found no need to resolve arguments about the significance of Sure-Tan because the "legal landscape" had since changed significantly with the enactment of the IRCA. Id. at 1282.

170. 122 S. Ct. at 1279.

171. Id. at 1280-85.

172. Id. at 1282.

173. Id. at 1285-89.

174. Id. at 1289.

175. Id. at 1290.

176. 467 U.S. at 842, 14 ELR at 20508.

177. Id. at 843-44, 14 ELR at 20508-09.

178. Id. at 865-66, 14 ELR at 20514.

179. See, e.g., Ernest Gellhorn & Paul Verkuil, Controlling Chevron-Based Delegations, 20 CARDOZO L. REV. 989 (1999); Thomas W. Merrill, Capture Theory and the Courts: 1967-1983, 72 CHI.-KENT L. REV. 1039 (1997); Cynthia Farina, The "Chief Executive" and the Quiet Constitutional Revolution, 49 ADMIN. L. REV. 179 (1997); Yvette M. Barksdale, New Wine Bottles: Rethinking Political and Judicial Controls on Administration, 72 CHI.-KENT L. REV. 1337 (1997); A Pragmatic Approach to Chevron, 112 HARV. L. REV. 1723 (1999); E. Perry Hicks, Designation Without Conservation: The Conflict Between the Endangered Species Act and Its Implementing Regulations, 14 VA. ENVTL. L.J. 491 (2000); Thomas W. Merrill, Judicial Deference to Executive Precedent, 101 YALE L.J. 969 (1992); Jonathan T. Molot, The Judicial Perspective in the Administrative State: Reconciling Modern Doctrines of Deference With the Judiciary's Structural Role, 53 STAN. L. REV. 1 (2000); Ronald M. Levin, The Anatomy of Chevron: Step Two Reconsidered, 72 CHI.-KENT L. REV. 1253 (1997).

180. See Schuck & Elliott, supra note 24.

181. Id. See also Jordan, supra note 24.

182. This is the issue of whether Chevron applies, discussed supra in section II.A.

183. But see Krotoszynski, supra note 44.

184. See supra notes 21-24 and accompanying text.

185. See Publication Rules, supra note 40, at 838; Anthony, supra note 37, at 3 n.4, 41-42; Jordan, supra note 39, at 722-25.

186. Chevron, 467 U.S. at 842, 14 ELR at 20508.

187. Id. at 863-66, 14 ELR at 20513.

188. Publication Rules, supra note 40, at 837-38.

189. 529 U.S. at 596 (Breyer, J., dissenting) (complete citation omitted).

190. Id.

191. Id.

192. Id. at 597.

193. 533 U.S. at 281.

194. 533 U.S. at 227.

195. Id. (citing, inter alia, Skidmore, footnotes and citations omitted).

196. Id. at 229.

197. Id. at 238.

198. See supra notes 144-45 and accompanying text.

199. 122 S. Ct. at 1150.

200. Id.

201. Publication Rules, supra note 40, at 837-38.

202. Merrill & Hickman, supra note 7, at 852 n.96.

203. 122 S. Ct. at 1150 n.8.

204. Id.

205. Id. at 1154 (O'Connor, J., concurring).

206. 122 S. Ct. at 2293-99.

207. Id. at 2203.

208. Id.

209. Id.

210. Id. at 2203-04 (citing Chevron).

211. Id.

212. Mead, 533 U.S. at 226-27.

213. Breyer, supra note 31, at 373.

214. Christensen, 529 U.S. at 596-97 (Breyer, J., concurring).