Shared Flashcard Set


Administrative Law (Kalt 2011)
Admin Class Summaries

Additional Law Flashcards




Administrative Law Practice

Introduction to administrative law
Getting to know the APA
1. Administrative law deals with the laws and standards that constrain agency action, including especially the Administrative Procedure Act (APA).

2. Agencies perform tasks that closely resemble legislation (prospective, general applicability) and adjudication (retrospective, discrete application).

3. There are four recurring and overlapping themes in this class: accountability to the political process; rule by experts; legitimation through quasi-judicial process; and micro-political participation.

4. The majority of the lawyering in administrative law is done before getting to court. If you have to sue in administrative law, it usually means that you will probably lose, because courts are highly deferential to agencies.

5. Agencies provide notice of their actions in the Federal Register. Their regulations are codified in the CFR.

6. Because of the deference given to agencies on substantive matters, the only way for many people to attack agency action effectively is by attacking the procedures used.

7. Although we will talk about the APA a lot, in almost every case in real life, there will be a statute specific to the agency that applies. Very often the statute will simply advert to the APA. Other times, however, the statute may have something different to say, and when this happens, the statute controls.

"Where do rules come from?": sources of rules; lobbying; and petitions
1. In determining the constraints facing the actions of a particular agency, you should look first to the statute that governs the agency, and only second (if necessary) to the APA. This is as true for rulemaking and its initiation as elsewhere.

2. There are two practical models of the instigation of rulemaking inside an agency. The first is the bottom-up process--technical staff consults the literature, data, and experts, and makes recommendations. The second is the top-down process--political reasons for making a new rule come from the head of the agency, Congress, a cabinet secretary, or the President. These are not mutually exclusive processes.

3. Furthermore, the impetus for either of the processes in #2 can come from outside the government as well. In other words, the public (individuals, corporations, interest groups, etc.) can feed the bottom-up and/or top-down modes by exerting pressure, through making connections and suggestions to staff, or by playing pressure politics through the government or the media.

4. If all else fails, individuals can always petition for rulemaking.

5. Some practical points on effective lobbying: (1) The agency is not a courtroom; (2) Be on the level about your interest, but realize that if all you are saying is that you want the rule changed because it is hurting you, the agency is unlikely to be sympathetic--tell them why your proffered action is good for the public and for the agency and consistent with the agency's statutory mandate; (3) Be technical. If you have science on your side, the bottom-up approach can be a powerful one.

6. Pursuant to the APA (5 U.S.C.) § 553(e), any interested party can petition an agency to initiate rulemaking. The agency must promptly respond (§ 555(b)) and, if the petition is denied, give an explanation (§ 555(e)). As a practical matter, though, courts have been reluctant to give the "promptness" requirement any teeth. Delays are either accepted or given only weak stimulus from the court. When the court does act, it guarantees only that the petition is answered, not that it is granted. But even if the petitioner loses, it might still force the agency to put its cards on the table, or provide other helpful external benefits.

7. Congress may direct an agency to promulgate rules on a certain subject. It may provide a time limit for the rulemaking, though courts do not always enforce these strongly unless the statute provides explicit penalties for failure to meet the deadline.

Denial of petitions
Exceptions to the APA rulemaking process
1. The U.S. Court of Appeals for the D.C. Circuit is the most important court in administrative law, because so many agencies are located in D.C. Indeed, many statutes assign review of agency action to that court, in order to provide uniformity under the aegis of an expert court. As a result of these factors, D.C. Circuit precedents on administrative law, while not binding on other circuits, are often accorded extra weight by them.

2. A person who petitions for rulemaking and is denied by the agency can seek review of the decision in an Article III court. The court will review the agency decision under the deferential "arbitrary and capricious" standard. Rather than ask whether the petition should or should not have been granted, the court will just ask whether the agency (1) gave a satisfactory explanation for its action, and (2) that explanation reflected a rational connection between the agency's decision and the facts in the administrative record. Thus, the agency might prevail even if the court would have concluded that the agency lacked a preponderance of the evidence on its side. (On the other hand, if the explanation wasn't good enough, the opposite might occur too: an agency could have a preponderance on its side but lose the case because its explanation was inadequate.) In any case, the remedy is generally a remand for a proper explanation, if one can be constructed; the Article III court won't itself grant a petition.

3. APA (5 USC) § 553 on informal rulemaking contains several procedural requirements: (b) notice of proposed rulemaking, (c) public comment, (d) publication of the final rule [and as discussed last week, (e) openness to petitions].

4. Military and foreign affairs matters are exempt from all of these requirements. Matters relating to "agency management or personnel or to public property, loans, grants, benefits, or contracts" are similarly exempt, though as to the last half of that, Congress or the agencies themselves generally require these procedures anyway, through specific statutes superseding the APA, or via self-imposed procedural rules.

5. Interpretive rules and policy statements are exempt from § 553's notice and comment requirements; we will cover them in more depth later in the semester.

6. Procedural rules are exempt from notice and comment. The standard for distinguishing procedural rules (exempt from APA notice and comment) from substantive rules (not exempt) is muddled and frustrating. Most courts just try to determine the extent to which the rule affects the "primary conduct" of the regulated entity as opposed to merely affecting how the entity presents its conduct. The more that the former (primary) conduct is affected, the more likely the rule is to be considered substantive.

Typical APA rulemaking I (overview; notice; introduction to comment)
1. Instances where the agency for good cause finds notice, comment, and publication to be "impracticable, unnecessary, or contrary to the public interest" are potentially exempt from notice and comment as well. This "good cause" exception is not read as broadly as it sounds in (5 USC) § 553, because doing so would give the agency an incentive to overuse it, and the exception would swallow the rule.

2. If a statute directs the agency to make rules "on the record after opportunity for an agency hearing," then the agency must use the formal rulemaking process (5 U.S.C. §§ 556, 557). If the statute specifies other specific procedures, hybrid rulemaking will be used. Otherwise, informal rulemaking will be used (5 U.S.C. § 553). [The agency can always add procedures voluntarily if it sees fit, as long as its statute doesn't preclude it.]

3. In Vermont Yankee Nuclear Power Corp v. NRDC, 435 U.S. 519 (1978), the Supreme Court said that courts should not add more mandatory agency procedures "absent constitutional restraints or extremely compelling circumstances." Generally speaking, therefore, the procedures mandated by the agency's own statute will be the limit.

4. APA notice requirements for informal rulemaking are simple: notice must include a notice of proposed rulemaking (NPRM) in the Federal Register which contains the data and methodology underlying the proposal; notice of the statutory basis for the rulemaking; the time, place, nature, etc., of any public proceedings; and either the gist of the proposal or (typically, but not mandatorily unless the agency statute specifies it separately) the full text of the proposed rule. The data and methodology requirement comes from case law, not the APA, despite Vermont Yankee.

5. The main conflict over notice occurs when a rule changes so much between the proposed version and the final version that an interested party effectively had no notice of the final version. How much change is too much? Many courts use the "logical outgrowth" test to determine whether notice is inadequate. This is a good test only to the extent that the "outgrowth" is from the proposed rule and any other materials (including, perhaps, other comments that have been docketed publicly) that interested parties have an adequate opportunity to comment upon. A better way to look at the notice question is to ask whether these materials put the issues or interests of a party "on the table." That is, should the party have known that its interests were implicated by the impending rule?
Rulemaking Typical APA Rulemaking II (comment; statement of basis and purpose) Tweaking the APA: hybrid rulemaking
1. Comment is discussed in (5 U.S.C.) § 553(c). Typically, comments are written. The APA does not provide any minimum or maximum length for comment periods, but some individual statutes do. Agencies often extend the comment period when doing so seems helpful. 2. Formal rulemaking does not allow for ex parte communications, but the APA places no explicit restrictions on ex parte communications in informal rulemaking (IRM). In IRM, ex parte communication issues involve three questions: (1) what contacts can take place?; (2) which contacts must be formally disclosed?; and (3) should others be given time to make comments in reaction to the disclosed communication? 3. On the first question, ex parte contacts themselves are generally legal. One exception is that the Constitution (via the Due Process Clause) restricts ex parte contacts in those rare IRM cases that involve parties competing adversarially for a limited privilege (e.g., a single TV license). Agencies might also face limits to ex parte communications from Congress (in the agency's own statute) or from themselves (via procedural rules). 4. As to the second and third questions, some courts have required disclosure of ex parte communications in IRM. While the more restrictive of such case law is suspect under Vermont Yankee, many agencies (and occasionally some courts) some follow these restrictive rules anyway. 5. In practice, ex parte communications are only an issue in IRM where they actually shaped the substantive decision reached by the agency. Courts want the record to include the grounds (the data and other technical bases) for the decision. That said, if technical information like this comes in through ex parte communications, the most that courts will require of the agency is to disclose those communications and, possibly give others an opportunity to comment. But if all of the relevant data is already reflected in the record, then political pressure to favor one set of technical arguments over another need not be docketed in this way, let alone set out for comment. 6. The APA requires a final rule to have a statement of basis and purpose in its preamble. Many statutes require it to also include responses to the comments, and because the preamble is an important source for courts reviewing the agency's decision, agencies tend to err on the side of comprehensiveness, making these preambles lengthy.

ADR in administrative law: negotiated rulemaking
Getting into a "real" court: judicial review and the Chevron principle
1. Various supplemental rulemaking requirement statutes require agencies as part of the rulemaking process to consider--formally--the impact of a rule on certain things. Some examples are NEPA (environmental impact), RFA (small organizations), PRA (paperwork), and E.O. 12866 (cost-benefit analysis). There are several other executive sources of a similar ilk, but unless they have strong judicial-review or institutional teeth, they don't amount to much.

2. Negotiated rulemaking is an ADR-like supplement to traditional notice-and-comment rulemaking that produces consensus-based NPRMs. It works best if the interest groups involved are able to compromise without giving up their fundamental beliefs; if there are adequate representatives of the important interests; and if there are not too many interests around the table. An inherent problem with NR, though, is that it treats the agency as a neutral facilitator rather than as an accountable advocate of its vision of the public interest.

3. Chevron is the most important case in modern administrative law. It establishes when courts reviewing agency rulemaking should defer to the way the agency interpreted the statute in its rule. Such agency interpretations are initially subject to de novo review, but Chevron potentially adds some deference to that.

4. The Chevron test has two steps. Step one is to ask whether the statute is ambiguous. If not, then the court gives no deference to the agency's interpretation--the statute can have only one meaning; if the rule being reviewed shares that meaning, it is approved, and if it doesn't share that meaning then the rule is struck down. If and only if the statute is ambiguous, the court moves to step two.

5. Some judges look only to the plain meaning of the statute to gauge its ambiguity; others use canons of interpretation, or look more broadly to the spirit of a statute, or look to legislative intent and legislative history.

6. Step two of Chevron asks whether the agency's interpretation of the statute is reasonable. Because the statute is ambiguous or has an intentional gap if we have gotten to this point, there are a range of possible reasonable interpretations. As long as the agency's interpretation is within that range--even if the court might prefer another interpretation--the agency's interpretation will be upheld. It is rare for an agency to lose at step two.

7. The question in Chevron cases is the classic legal-process question: who decides? If Congress has given an agency authority (either explicitly by expressly delegating interpretive authority, or implicitly by just being vague), then it wants the agency--not the court--to resolve any ambiguities. While a court should defend a statute against unreasonable interpretations, it should also leave it to the agency to choose from among a range of acceptable interpretations.

More "real" court: substantive judicial review
Practical considerations for judicial review
1. Chevron requires that the statute is this agency's to interpret, so it does not apply when an agency interprets another agency's statute, or a statute that applies broadly to multiple agencies, or a statute in which Congress has withheld that interpretive authority from the agency.

2. Courts are reluctant to accord Chevron deference when the part of the statute that the agency is interpreting concerns the extent of its own jurisdiction.

3. Section 706 specifies the standard of review for Article III review of agency action. For informal rulemaking, we use the "arbitrary and capricious" standard. For formal rulemaking or other instances in which there is a formal record, we use the "substantial evidence" standard. (For hybrid rulemaking, it is whatever the statute specifies.) Functionally, both standards are the same--looking for reasonableness. Despite the similarity of the standards as a logical matter, though, "substantial evidence" is seen as being a signal for slightly more searching review.

4. The informal rulemaking "record," such as it is, consists of the notice of proposed rulemaking (NPRM), the final rule (including its preamble), and any other materials used by the agency to reach its conclusion, such as comments and other docketed input.

5. In substantive review—NOT questions of law/Chevron, but rather questions of fact regarding the actual outcome of the rule—the court will look to see (among other things) if the agency gave an "adequate explanation" for its decision. If the agency loses, the remedy is a remand for an adequate explanation; the reviewing court will not itself rewrite the rule.

6. When a court reviews a rule passed via informal rulemaking, it will determine that the agency was arbitrary and capricious if the agency: (A) relied on factors that Congress did not want considered; (B) left out a crucial issue, either one that Congress wanted addressed or that logic suggests should be considered; (C) offered an inadequate explanation, meaning that it either does not adequately treat the data on both sides, or that the explanation just doesn't add up; or (D) made a decision that was just really stupid.

7. An agency can bolster its substantive choices by finding data to support them, undermining data that oppose it, or, in the absence of sufficient data, relying on its expertise to the extent that doing so is plausible.

8. When repealing a rule, an agency must be reasonable based not only on the original record, but also on the original rule too. In other words, the record for the second rule will include the first rule.

Administrative adjudication: an overview
Adjudication procedures
1. In deciding whether or not to challenge a rule in court, one should consider the following factors: (1) Are there alternatives to litigation? Examples include lobbying the agency for reconsideration or a favorable interpretive rule, or lobbying Congress for a statutory fix. (2) What are the chances of winning? (3) What would you win if you did win? Recall that many times, the only victory likely is procedural; a remand may just forestall the inevitable. (4) What will the costs be, both in terms of litigation and good will between the agency and the client (who may be dealing with the agency more in the future, perhaps as a regulated entity)?

2. An appeal that causes a beneficial delay may or may not be unethical. A poor chance of success on the merits does not necessarily mean that a position is indefensible or that suing is automatically in bad faith. As long as there are good faith reasons to proceed, the added fact that delay is beneficial to the client will not be sufficient grounds for sanctions. But keep in mind the crucial fact that no court (let alone the agency itself) is likely to stay implementation of the rule unless there is a good chance of success on the merits.

3. Adjudication is defined as a process for producing an order, and an order is defined as a final agency disposition of an issue, other than rulemaking.

4. The APA provides formal procedures for adjudication in (5 U.S.C.) § 554, (5 U.S.C.) § 556, and (5 U.S.C.) § 557. Informal adjudication is possible too, and is sometimes called "non-APA" adjudication instead, because it is often still quite formal. Some minor parts of the APA still apply to informal adjudication, though, particularly provisions in (5 U.S.C.) § 555 that relate to due process.

5. Whether or not it follows the APA's "formal" procedures, an agency will generally promulgate (via procedural rulemaking) its own additional specific procedures for adjudication. The APA's requirements are specific, but they are not comprehensive.

6. Constitutional due process requirements always apply, but the formal APA requirements pretty much cover them; due process arguments tend to arise only in the context of informal adjudication.

7. As with formal rulemaking, Congress can indicate that it wants the agency to use formal adjudication by using the so-called magic words: that adjudication should be "on the record after opportunity for an agency hearing." If the agency neither requires nor forbids formal adjudication, the agency can choose informal or formal adjudication, and some do choose the latter.

8. The book lays out the APA requirements for formal adjudication pretty well. We discussed notice (time, place, and manner of the hearing, the legal authority for it, and the matters of fact and law asserted by the instigator of the proceeding), third-party participation (broad, but subject to restriction for practicality; the issue here is not just whether one can participate but the extent of that participation), and ALJs (administrative law judges have most of the procedural powers of an article III judge but cannot strike down statutes, the agency's regulations, or the agency's interpretations of either).

Applying adjudication procedures
Keeping your mouth shut: adjudication and ex parte communications
1. In formal adjudication, the APA guarantees that the burden of proof always rests on the "proponent" or instigator of the proceeding, unless the underlying statute provides otherwise--only Congress, not the agency, can move the burden of proof. This applies only to the burden of persuasion (i.e., the burden of establishing a certain level of evidence on your side), not the burden of production (i.e., the burden of coming forward with some evidence to avoid an automatic loss); the agency can shift the latter through its construction of more detailed procedures.

2. [This one was mainly covered in Class #8, but I wanted to present it all together.] Formal adjudication--the production of a final agency order via APA procedures--can be done either by the head(s) of an agency or by an ALJ or some combination--the statute will choose which, or the agency will choose if the statute leaves them the choice. ALJs are employed by the agency but are insulated from some hierarchical control. They are also kept insulated from the part of the agency (if there is one) that advocates the agency position before them. Finally, ALJs are also kept insulated from ex parte communications in adversarial cases in ways that the heads of the agency are not.

3. In formal adjudication, some agencies use an ALJ to compile the record and propose a disposition, which the head(s) of the agency then review de novo to produce a final order. In some agencies, displeased parties have the option of either appealing the ALJ's decision to the head(s), or just letting the ALJ's decision become final and proceeding directly to an Article III court. Still other agencies just make the ALJ's decision final, period. Others don't use ALJs at all.

4. In formal adjudication, the finder of fact and courts reviewing them use the "substantial evidence" (SE) standard in 5 U.S.C. § 556(d). (This is different from the deferential "substantial evidence" standard used by courts to review formal agency action. There, SE referred to the quantity of evidence the deferential court must find before affirming: an amount sufficient for a reasonable decision maker to reach the result the agency did.) The § 556(d) SE standard means that the ALJ must rely on evidence of sufficient quality. Under this qualitative standard, even hearsay may qualify as substantial evidence, but only to the extent that it is otherwise reliable or the opposing party had an opportunity to cross-examine it, etc.

5. One must demonstrate prejudice to successfully allege that one has not received the notice required by the APA for an adjudication.

6. There are strict limits in the APA (5 U.S.C. § 554(d) and 5 U.S.C. § 557(d)) on ex parte communications for formal adjudication. There are no explicit APA restrictions on them for informal adjudication, though many agencies have restrictions in their statutes or procedural rules.

7. In 5 U.S.C. § 557(d), it is specified that ex parte communications between interested persons outside the agency and the agency adjudicators, relevant to the merits of the adjudication, are forbidden. The most common remedy is to disclose the communications. Where appropriate, a more severe remedy is available: the agency may require a show cause hearing in which the improper communicator is given the burden to show cause why the tainted adjudication should not be altered to his detriment. This remedy is only available when the improper communicator is a party, though, as opposed to a mere "interested person."

8. There is a third remedy for ex parte communications in a formal adjudication, not mentioned in § 557. It is for an article III court reviewing the matter to vacate the result of the adjudication and remand for a new adjudication. For this third remedy, the question is whether the decision making process was so irrevocably tainted that the resulting judgment was unfair. One should consider: (1) the gravity of contacts; (2) whether they influenced the decision; (3) whether the improper communicator benefited; (4) whether adverse parties knew about it; and (5) whether remand would serve a useful purpose.

9. Section 554(d) limits contacts from interested parties inside the agency. An ALJ is insulated from contact with the agency's advocates. This does not apply to the heads of the agency, however, as they have to run the place and can't really be insulated from anyone.

Adjudication and constitutional due process I (individualization)
Adjudication and constitutional due process II (property, liberty, and procedure)
Adjudication and constitutional due process III (what process is due?)
1. In non-APA adjudication, the limit on ex parte communications is constitutional. The key aspect of due process here is that interested parties have a right to notice and an opportunity to be heard. The information communicated ex parte must be new and material to the deciding official, however. If it was merely duplicative, if the other party had a chance to respond, or if it had no effect on the decision maker, then there was no violation of due process. (Keep in mind that whether something is "new" is quite arguable, as what is "merely duplicative" to one side may be "powerful and effective reiteration" on the other.)

2. Due process rights only attach in certain contexts. First, there must be individualized decision making. For instance, your right to notice and a hearing regarding taxes varies depending on the context. One that is based on an individualized assessment of personal, adjudicative facts about you (like the assessment of your home's value for property taxes) are sufficiently individualized. More general issues based on mass, legislative facts (like changing the millage rate) are not, and you have no due-process right to be heard. The resolution of mass issues is achieved politically, and participation is achieved through representation and the First Amendment, not the Fifth. Virtually any challenge to procedures in adjudication will involve individualized decision making, though, so this is usually not much of an issue for us here.

3. The second prerequisite for making a due-process argument is that there must have been a liberty or property interest at stake in the underlying case. A property interest can come from having an entitlement or a legitimate, not-unilateral expectation of a guaranteed benefit. A liberty interest can be (among other things) the interest in pursuing a particular profession. A liberty interest can be infringed by (among other things) a reputational harm that seriously impedes that pursuit, so long as that harm is accompanied by some other action by the agency (like formally disciplining you). More commonly and simply, it can also be harm to something you simply own (property interest) or bodily restraint (liberty interest).

4. Once you determine that due process rights apply, you must determine what process is due. For this, we turn to Mathews v. Eldridge, 424 U.S. 319 (1976). Under Mathews, we look at the general type of procedure (e.g., witnesses at an academic disciplinary hearing, social security disability benefit pre-deprivation hearing, etc.) that the plaintiff is claiming should have been used. We balance on one side the burden to the government that this additional procedure would represent. On the other side we look at the interest that is at stake for people in plaintiff's position, discounted according to the probability that adding the procedure in question will prevent the erroneous deprivation of that interest. (It is hard to plug in actual, precise numbers here, but it is somewhat easier to assess which side is weightier, which is all we are doing here.) Also note, though I didn’t make this point in class, that the assessment is done for the entire class of cases, not just for plaintiff's individual case, as that would get us into the merits.

Adjudication and constitutional due process III (neutrality)
Judicial review by the substantial evidence standard
A difficult line to draw: standards of review for law, fact, and mixed questions
1. Another due process right is the right to a neutral decision maker. A neutral decision maker is one who has no personal or financial interest in the case that would sway his or her consideration of the facts. We also look out for prejudgment of facts. Prejudging matters of law does not present a due-process problem, though. Combining an investigatory and decision-making role does not necessarily present a problem either.

2. When performing judicial review of a formal agency adjudication, using the "substantial evidence" standard means that the court looks to see that a reasonable person could have found enough evidence in the record that they would have reached the same conclusions as the agency did. However, it may be that considering other evidence necessarily leads a reasonable person to discount some of the evidence favoring the agency's result. In other words, the reviewing body must consider the evidence in light of the record as a whole. While the record may be viewed in the light most favorable to the agency or ALJ result, that light shines on all of the evidence in the record.

3. When an agency has an adjudication procedure involving both an ALJ and the head(s) of the agency, the ALJ's findings derived from the demeanor of witnesses ("testimonial inferences") receives "weight" from the agency head. Inferences that can just as well be derived from the cold record ("derivative inferences"), however, do not receive any such weight, though like all of the ALJ's conclusions they become part of the record. When an Article III court reviews such a case, it will review the head's decision under the appropriate deferential standard for reviewing facts, though it too will give weight to the ALJ's first-hand testimonial inferences. This means that the head should, if it wishes to differ from the ALJ yet survive Article III review, either rely on sufficient derivative inferences to constitute § 706 substantial evidence, or have a very good reason for rejecting the ALJ's testimonial inferences beyond just second-guessing the ALJ's first-hand assessment of demeanor.

4. In formal adjudication, fact questions ("was Lund on patrol that night?") are reviewed under the substantial evidence standard and questions of law ("what is the standard for determining eligibility for worker's compensation") are reviewed de novo--but perhaps with deference added on.

5. The Hearst line of cases suggests that deference may be appropriate on questions of law. As in Chevron, deference is obtained only when (1) the statute is one that the agency is in charge of applying; and (2) the statute is ambiguous and leaves it to the agency to resolve the ambiguity. At that point, the court should accept any agency interpretation that is reasonable. If Hearst deference does not apply, then review of questions of law remains de novo (though we will talk about other forms of deference later on).

6. A complicated issue arises over mixed questions of fact and law ("was Lund acting in the scope of his employment that night?"), which are questions that require of the agency both a factual assessment of what happened, and a legal assessment of the significance of what happened, to be applied as precedent in future analogous cases. This issue is less complicated when Hearst deference applies, because then both fact and law review is essentially deferential to the same degree. If Hearst does not apply, however, these mixed questions may not be reviewed with as much deference. The underlying factual findings will be, but not the application of the law to those facts. A helpful way to approach these issues is to ask first whether the agency used an acceptable legal standard (a question that may or may not be answered deferentially), and second, if so, whether there is substantial evidence to support the factual findings plugged into it.

Judicial review by the "arbitrary and capricious" standard
1. The de novo standard of review for questions of fact expressed in 5 U.S.C. § 706(2)(F) is used in the review of adjudications only when the agency's fact-finding ability is inherently inadequate. If, as to fact-finding, the agency simply failed to execute, the court will usually just remand the case back to the agency to do it right. There is also the possibility that the agency's specific statute will direct that actions begin in article III court in the first place. (Moreover, apart from 706(2)(F), the agency's statute might supersede the APA and provide directly for de novo review.)

2. One way for an agency to be arbitrary and capricious in an adjudication is to give an inadequate explanation of the basis of the decision. While courts are deferentially wary of second-guessing the thought processes of the agency, they do not hesitate to remand a case when insufficient information has been conveyed to even express what the agency was thinking.

3. By way of review, arbitrary and capricious review pervades administrative law, and wherever we find it, it includes an adequate explanation requirement. How much is "adequate" varies by context, though, and in some places very little is required.

4. Another version of arbitrary and capricious agency adjudication is inconsistency. Deferential review allows for a range of outcomes at the agency level, but at some point the agency is supposed to speak with one voice. The agency cannot be unreasonably inconsistent from one case to the next; other than in cases of first impression, the agency must justify an adjudicatory decision in terms of established agency precedent. This leaves room for the same sort of evolution that occurs in the common law (indeed, maybe even a bit more), but not for unreasonable or unexplained shifts of policy. In other words, the agency can have a major shift, but at the very least such shifts require confronting and satisfactorily explaining the departure.
Choice of Procedures and Non-Legislative Rules

Choice of procedures (adjudication versus rulemaking)
1. Sometimes, an agency's statute may leave room for it to choose to develop standards either through rulemaking or adjudication. The advantages of rulemaking are that: it provides clear, broad notice; in the enforcement context, it can tell everyone what to do, both negatively and positively, instead of creating standards around individual violations; and the comment process provides helpful input. The advantages of adjudication are that: it is faster and cheaper; it allows standards to develop case by case with hindsight, and in a more flexible, nuanced, and contextualized manner; and it allows handpicking the adverse party.

2. Adjudication is faster and cheaper both because no comment period is required and because adjudicatory notice is, as a practical matter, less effective. (As we saw before, while there may be participation in a formal adjudication that is just as wide as in rulemaking, 5 U.S.C. § 555 allows the agency to limit participation based on the need for the "orderly conduct" of business.)

3. The choice of procedures that an agency makes (to the extent that it has a choice) is reviewable, but only for arbitrariness and caprice.

4. Adjudication may present a problem regarding retroactivity, if it imposes a standard on a party that was too difficult for the party to anticipate. The party in the adjudication must be subjected to some consequences under the new rule; otherwise, the adjudication would be non-adversarial, purely prospective, and amount to rulemaking without all of rulemaking's constraints. If a rule is good enough to apply prospectively, it must be good enough to apply to the party in the adjudication. But third parties whose conduct occurred before the new rule was announced are more likely to be exempted from a retroactive effect when their cases come up before the agency. Moreover, the party in the original case might merit a weaker remedy, so long as it faces some sanction, and so long as the weaker remedy is applied consistently to similarly situated parties.

5. The standard for determining whether a standard announced in an adjudication should apply retroactively are: (1) the extent to which the rule is new/different and thus unexpected and upsetting to substantial reliance interests; and (2) the countervailing interest that the agency has in imposing the standard retroactively. This applies both to applying a particular sanction in the first case in which the new standard appears, and to applying that sanction or any sanction at all in subsequent cases. An agency's decision to apply a standard retroactively in adjudication is reviewed de novo by the Article III court. An agency's decision not to apply a standard retroactively in subsequent adjudications would probably only be reviewed under A&C, though.

6. Before an agency can choose to use rulemaking, it must be authorized by statute to do so.

7. A definitive rule can moot a statutory right to receive a formal adjudication. The rule may be so clear that there are no material facts left to adjudicate, in which case the regulated entity no longer has a right to formal adjudication. Some agencies have waiver provisions by which entities can argue that they should be exempt from a rule, but agencies generally are not required to do so, and they tend to be sparing in the use of such procedures when they do have them.

8. Agencies can make rules apply retroactively, but only if explicitly authorized by Congress in the agency's statute.

9. A court will defer to an agency's interpretation of its own vague rule, so long as that interpretation is reasonable. We will call this GE (or Seminole Rock or Auer) deference.

10. An entity may claim that even if the court accepts the agency's interpretation, it should not be applied in the entity's particular case because it had no notice of the agency's interpretation, but such a claim will only succeed if several factors are in place. First, there must have really not been any notice, and there are a few ways to give notice: agencies can provide direct pre-enforcement notice to the entity, or they can publish their interpretation as a nonlegislative rule, or a case might not be one of first impression. Second, it must be reasonable for the entity not to have anticipated that the agency might choose the interpretation that it did (under the "fairly obvious" standard). If the agency's position is one of a number of apparent possible interpretations, that is good enough for the agency to win under this standard. Third, the entity will have a hard time prevailing unless the penalty leveled against it is substantial and retrospective (e.g., a big fine, as opposed to a cease-and-desist order).
Choice of Procedures and Non-Legislative Rules

"When is a rule not a rule?" (non-legislative rules)
1. A third choice that an agency can use to implement standards from a statute (if the statute gives it a choice) is to use nonlegislative rulemaking. The principal advantage is that nonlegislative rules do not require notice and comment or any of the other complications of legislative rulemaking, so they can provide structure and relative certainty quickly and cheaply. The principal disadvantage is that they do not get the deference or have the binding force of law that regular rules or formal adjudications do. Also, notice and comment are not all bad.

2. FOIA (5 U.S.C. § 552) requires that nonlegislative rules be published in the Federal Register. This requirement is weak, though, because the consequence upon a violation is that the agency cannot use an unpublished nonlegislative rule to someone's detriment. Since nonlegislative rules are not formally binding, that rarely happens anyway.

3. One category of nonlegislative rules--policy statements--are distinguished by their failure to bind the agency (let alone the regulated entity). Policy statement set forth the agency's desires using hedged terms like "should" or "may," as opposed to setting forth requirements and using mandatory terms like "must" or "will."

4. Interpretive rules are binding in one sense--once an agency promulgates one, it is bound to follow it unless and until it comes out with a new interpretation. As a result, regulated entities are informally bound by it too, in the sense that they must follow the agency's interpretation or suffer the likely (if not quite official) consequences. In another sense, though, interpretive rules are not binding, because they do not create a new duty. In fact, because they focus and clarify the duty imposed by a vague statute or legislative rule, it can be said in the abstract that they actually reduce the potential duties of regulated entities.

5. An interpretive rule is determined to be such by looking at two factors. First, how does the agency characterize it? (This obviously cannot be dispositive.) Second, what is the source of the duty placed on the regulated entity? If the rule really is an interpretive rule, the actual duty at issue will come from the statute or legislative rule that the IR interprets. But creating a duty is different from adding detailed focus where there was none before.
Choice of Procedures and Non-Legislative Rules

How to get whipsawed by the government: estoppel
Deference by courts to agencies I: Mead
1. An agency generally can change an interpretive rule without going through notice and comment, just as it can do in passing one in the first place. If, however, (1) the interpretive rule interprets a legislative rule (as opposed to interpreting a statute); (2) the change in interpretation produces a substantial modification in the duties or rights of the regulated entities; (3) the old interpretation was formalized and official; and (4) reliance on the old interpretation was substantial, some courts might require notice and comment for the new interpretive rule.

2. It is exceedingly rare for a federal court (less rare for a state court) to find that the government is estopped from using an interpretation against a regulated entity on grounds that the entity had been told something different, unofficially, by an agency employee. In private law, such estoppel requires reasonable, detrimental reliance. When government action is at issue, courts find it difficult to conclude that relying on agency communications is reasonable. This is especially true if the communication was informal, oral, etc. Among other reasons for the government to be treated differently here are that the government acts pursuant to law, which anyone can find for themselves; and that estoppel amounts to a rogue employee and a court applying different law to the regulated entity than that which Congress passed. This is especially problematic--and unconstitutional--when the estoppel decision requires expenditure of government funds. Estoppel might succeed, though, if criminal sanctions are at stake. And even if estoppel doesn't work, the regulated entity can try to make a due-process, consistency, or notice argument.

3. The pre-APA case of Skidmore v. Swift provides for a form of weak deference that applies in cases where Chevron deference does not. It looks at factors such as the expertise of the agency, the care with which the issue was considered, etc. Skidmore deference is quite indeterminate, and its effect can range from no real deference at all to essentially giving the agency the full benefit of the doubt.
Choice of Procedures and Non-Legislative Rules

Deference by courts to agencies II: Walton


Introduction to reviewability
Introduction to standing
1. Chevron-like deference does not apply to every agency interpretation of a statute, though just when it does is not clear right now. Interpretations emerging from rulemaking or formal adjudication are not problematic, but others, like informal adjudication and nonlegislative rules, fall into a grayer area. Under Walton, the Court examines the centrality of the gap in the statute, the expertise of the agency, the amount of attention the agency gave the issue, and so on. In some cases--as in Walton itself--using this sort of analysis may mean giving Chevron-like deference to policy statements or interpretive rules interpreting statutes.

2. The agency's interpretation of vague rules (as opposed to vague statutes) is still accorded something much like Chevron deference under GE. Under Gonzales v. Oregon, though, the Court apparently will not apply GE deference in situations where the legislative rule just parrots a vague statute, and will instead treat this as though the interpretation is of the statute rather than the rule (though with Walton this means that the agency might end up with Chevron-like deference anyway).

3. In order to be able to review an agency action in an article III court, you must have jurisdiction (which is generally available from 42 U.S.C. § 1331) and a cause of action (generally available from (APA) 5 U.S.C § 702, subject to the limits of (APA) 5 U.S.C. § 701 and the requirements of finality, exhaustion, and ripeness). Individual statutes may expand, contract, or change jurisdiction or the cause of action.

4. The doctrine of primary jurisdiction may require that a case between two private parties be sent to an agency for determination of an issue that is within that agency's purview.

5. The constitutional requirements for standing are injury in fact, causation, and redressability.

1. Standing has two components: constitutional requirements that apply regardless of any congressional action; and prudential requirements that can be reduced or eliminated in particular situations by statute.

2. The constitutional requirements for standing are injury in fact, causation, and redressability. Injury in fact means that the plaintiff has suffered or is about to suffer a concrete and particularized injury. This can include economic or physical injuries, but also such things as aesthetic or recreational injuries. It does not include speculative future injuries, nor does it include generalized, abstract injuries like violating an interest in seeing the law properly followed or tax dollars well-spent.

3. Causation means that the injury is fairly traceable to the defendant. The pleadings must be specific in linking the defendant's actions to the plaintiff's injuries--locations must be precise and effects clear rather than speculative (e.g., the government wants to allow strip mining on this land; I go to that exact piece of land regularly and enjoy the view, which will now be spoiled as a result of the government's action; this point goes to evaluating injury-in-fact too). Finally, redressability requires that the relief that plaintiff has requested would likely, not just possibly, redress the injury.

4. Prudential requirements include a strengthened requirement of non-generalization, and a notion that the plaintiff is not litigating an issue that is really the interest of a third party. These requirements can be eliminated by statute, such as the APA's broad grant of standing to any interested party, or, if the APA does not apply, by specific broadening in a particular statute. They generally are. A third prudential requirement, that plaintiff be in the zone of interests of the statute, will be discussed later.

5. An association can have standing without itself suffering an injury in fact and despite the prudential limit on third party standing, if three factors are in place. One of the members of the group must have standing to bring the suit; the organization's purpose must be related to the subject of the litigation; and the suit must be for declaratory or injunctive relief rather than damages (the latter of which any § 702 case will be anyway, so we don't really discuss this factor).

6. Procedural injuries are a special case. If an agency has failed to fulfill a procedural requirement (e.g., not having notice and comment), this probably will not be a sufficient injury-in-fact without more (e.g., that the regulation in question will cause harm--economic or personal or aesthetic or whatever--to plaintiff). The Supreme Court has said, however, that procedural injuries represent an exception to the causation and redressability requirements (e.g., the plaintiff can get standing without showing that the regulation wouldn't have passed in its injurious form had notice and comment been executed properly).

7. Standing does not require a nexus between the injury (e.g., the inability to go hiking somewhere) and the issue to be litigated on the merits (e.g., the underlying legality of building a dam).

Cause of action
Agency action and failure to act
Unreviewability under the APA I (statutory preclusion)
1. The zone of interests (ZOI) test is a prudential standing requirement that applies in third-party sorts of cases. It requires that the plaintiff be within the ZOI of the statute that he or she claims has been violated. This requirement is derived from § 702. Plaintiffs who are regulated directly by the government in a way that they do not like don't need to conform to the ZOI requirement. The ZOI requirement can also be mooted if the statute provides a very broad cause of action, which would amount to a declaration by Congress that no prudential limits to standing should apply.

2. In determining the ZOI of a statute, one must examine the particular provision alleged to have been violated, but in the context of the entire statutory scheme. A ZOI argument does not entail showing that Congress specifically intended for plaintiff's group to be protected. Rather, it need only be shown that Congress intended to deal with the interests that plaintiff is dealing with. In other words, look for "what" and not "who" when gauging the zone of interests. Also, it need only be shown that Congress arguably intended to protect the interest in question.

3. Judicial review of agency action requires just that--agency action--so it can be hard to get review of a failure to act by the agency. The Supreme Court has required in such cases that the plaintiff show that the agency (1) has failed to take a discrete action like making a rule, issuing an order or license, granting relief, or leveling a sanction; and (2) that the agency was required by law to take that discrete action. The agency's mere failure to exercise its discretion in a certain way, therefore, is not otherwise actionable.

4. Although the APA provides for broad Article III reviewability of agency action, § 701(a) provides for two exceptions from reviewability. The first ((a)(1)) is for when the statute precludes reviewability.

5. Preclusion can be explicit or implicit. Express preclusion is rare, and generally not total. Indeed, total preclusion of judicial review of an agency's actions may be unconstitutional, especially when the agency's action is being challenged on constitutional grounds. Moreover, courts will read preclusive language narrowly, because there is a general presumption in favor of reviewability.

6. Implicit preclusion under § 701(a)(1) is a more complicated proposition. In the absence of express preclusion, a court will find that Congress has precluded judicial review of agency action only when congressional intent to do so is fairly discernable from the structure of the statute and/or the legislative history. Courts may speak of a "clear and convincing evidence" test, but they are not that strict in practice. Still, it’s not easy to show.

Unreviewability under the APA II (committed to agency discretion)
Fancy legal terminology for red tape I (finality)
1. The second exception in 5 U.S.C. § 701(a), for matters committed to agency discretion, resembles the political question doctrine from constitutional law. The question here is not whether the agency is given discretion--that would eliminate almost all reviewability. Rather, the issue is whether the matter is wholly within the agency's discretion, such that there is no basis for a court to act. This may be because of traditional limits on courts (such as when an agency is given "prosecutorial discretion") or simply because the statute gives no external standards to guide the agency (such as when an agency is required to take an action not when certain criteria are met, but instead just when "it deems" that certain criteria are met).

2. The "committed to agency discretion" test can be summed up by asking whether the statute gives the reviewing court any "law to apply." If there is no law to apply--i.e. there are no external standards that constrain the agency and that a court can apply in reviewing the agency's decision--then the matter is committed to agency discretion and is unreviewable. As a result, constitutional claims are usually reviewable, because constitutional claims generally represent external standards, and are rarely left up to someone besides the court to apply.

3. The three timing doctrines we will discuss are finality, exhaustion, and ripeness, which overlap and blend into each other a fair amount.

4. Finality asks whether the agency has completed an action, because only completed actions are supposed to be reviewable. It is specifically required by 5 U.S.C. § 704. There are two parts of the requirement (we discussed four parts, but they boil down to two). First, is the agency action the consummation of the agency's involvement? That is, is it definitive of the agency's position? Second, do legal consequences flow from the action? For instance, would violating the action result in a penalty? Does it affect the legal status of the plaintiff (as opposed to just affecting practical concerns)? Is immediate compliance expected? The agency's characterization of its own action is not dispositive.

Fancy legal terminology for red tape II (exhaustion)
1. Exhaustion, though it predates the APA, is required by § 704 too. The exhaustion doctrine prevents courts from hearing cases where the plaintiff has failed to follow a mandatory internal agency procedure required by statute or rule. The plaintiff must, in other words, go through all mandatory agency procedures before bringing a case to an article III court for review. Individual issues need to be exhausted too.

2. There are exceptions to the exhaustion requirement for cases in which exhaustion would cause unreasonable and prejudicial delay; the agency procedures do not provide for the type of relief sought (e.g., damages, or overturning the agency's procedures altogether); or in rare cases where the agency is so biased that it is futile to bother with exhaustion.

3. Ripeness asks whether a case is developed enough for a court to consider it. Ripeness requires that the case be ready for judicial resolution--that is concerns either a legal issue, or factual issues that are sufficiently developed. We must also balance the hardship that delay would cause the plaintiff against the interference that immediate review would present to the agency. The less the net hardship is on plaintiff, the more the case has to be ready for judicial resolution. Ripeness overlaps a fair amount with finality (which is a third element of ripeness) and exhaustion, and with standing too.

4. It is very important to consider not just whether a case is ripe for review, but whether the agency action will be stayed pending that review.
Agency Structure

Separation of powers, Shmeparation of powers
1. "Strong" views of separation of powers are formalist ones that require strict and technical separation. More moderate ones (which have prevailed) are functionalist. They look more forgivingly at whether the core power of a branch is being encroached upon unduly, and allow encroachments that the formalists would not.

2. Congress can delegate quasi-legislative power to executive agencies, so long as the delegation contains an "intelligible principle" that constrains the agency's discretion and provides a reviewing court with a way of knowing whether or not the agency is acting consistently with Congress's desires. [The more power is delegated, the more guidance must be given, up to "substantial guidance" for economy-wide legislation. The bar is fairly low even in the latter case, however, as we saw in Whitman.]

3. The Supreme Court has not used the non-delegation doctrine to strike down any legislation since the 1930s. The doctrine is still powerful, though, because it leads the Court to narrowly interpret statutes to avoid constitutional problems, even if that means rejecting a more intuitive interpretation of the statute. To find the "intelligible principle" that constrains agency discretion and renders a statute constitutional, a court may have to look to external sources such as statutory structure and legislative intent.

4. The delegation of final judicial authority to non-Article III decision makers is potentially a violation of the separation of powers. The issue is when such entities get the last word on matters of law and fact; Article III review, even if deferential, generally suffices to save the constitutionality of the structure. Earlier cases concentrated on a distinction between public and private rights, with the former not requiring Article III participation. For our purposes, though, the best approach now is that from the Schor case.

5. In Schor, the court distinguished between the personal right in having an Article III tribunal adjudicate one's rights, and the general structural interest in maintaining the separation of powers. The personal right can be waived; a person who voluntarily enters into a legal relationship with the knowledge that disputes may be definitively resolved by a non-Article III tribunal will not be able to challenge that tribunal's legitimacy. Even when not waived, though, the personal interest is functionalist: it is for an "impartial and independent federal adjudication" (which formal adjudication by an ALJ will generally provide) and not an Article III tribunal as such.

6. The structural interest cannot be waived, but it too is subject to the court's functionalist approach to the separation of powers--to be unconstitutional, the role of the non-Article III tribunal must encroach upon the core function of the Article III court. In other words, if the new tribunal is stripping away a significant part of the core power of the Article III court (like, say, deciding constitutional claims), that is a problem. If it is hearing minor disputes that are only incidentally within the Article III courts' purview (if that), then there is no problem. Again, keep in mind that the issue is only when the agency adjudicator gets the last word.
Agency Structure

The legislative veto

Inspections, Reports & Subpoenas

The power to demand information I (inspections)
1. The bicameralism and presentment requirement is viewed strictly (i.e., formalistically) by the Court. If an action is "legislative" in nature, it must be approved in identical form in both houses and signed by the President (or vetoed but then approved by 2/3 majorities in each house). As explained in INS v. Chadha, to be legislative in nature an action must have the "purpose and effect of altering . . . legal rights, duties and relations." Executive action that seemingly meets this definition, but which is pursuant to congressionally delegated power as discussed under the non-delegation doctrine, is still OK though.

2. Laws that give Congress the ability to efficiently review (and, if it wants, to overturn) rules are constitutional, so long as they require bicameralism and presentment to overturn the rule. The same is true of a law that would require Congress to affirm a rule before the rule could take effect. Allowing some mere subset of Congress to temporarily delay passage is probably OK, too, because mere delay probably does not count as a legislative act. Finally, giving the White House or OMB the ability to kill a rule before it can take effect is probably not a problem, because that can be considered just another part of the rulemaking process, which is OK there because it stays solely in the executive branch.

3. Agencies have two main limits on their ability to perform searches and inspections, etc. First, Congress must authorize the agency to do so, and Congress usually limits and specifically defines the bounds of an agency's search powers. Second, the Fourth Amendment requires that searches be reasonable and that warrants (where they are required) be specific and based on probable cause. But warrants are not required where there is no expectation of privacy, or where there is consent to a search or seizure, among other exceptions. They are also not required for searches of "pervasively regulated" entities if the search has an important governmental purpose and the agency's process offers other protections.

4. Even though a general warrant authorizing a general search would seem to violate the Fourth Amendment, it may nevertheless be enforced in certain situations. Even there, though, the search must be based on a general and prospective plan, neutrally applied--this limits the scope of such searches. For a more specific search (such as one arising out of a specific complaint), an agency must generally meet the Fourth Amendment's probable cause and specificity requirements, and the search will be limited to what is authorized by the warrant.

5. Remedies for an unconstitutional search or seizure by an agency are the same as in other contexts: Bivens actions or the application of the exclusionary rule. It is very rare, however, for the exclusionary rule to be used in the administrative, as opposed to criminal, context. Use of exclusion must be an effective deterrent, and the violation must have been quite egregious.
Inspections, Reports & Subpoenas

The power to demand information II (recordkeeping and reporting requirements)
The power to demand information III (constitutional limits)
1. Recordkeeping and reporting requirements are another tool of administrative agencies. Some statutes establish such requirements directly and leave it to the agency to provide more detail; others just authorize the agency to establish requirements. This authorization may or may not require that reporting requirements be established through the rulemaking process. Administrative subpoenas may be issued, but the authority to do so must be given to the agency expressly by statute.

2. The Paperwork Reduction Act, 44 U.S.C. § 3501 et seq., which applies to any reporting requirement that affects 10 or more people (besides criminal investigations and civil discovery), requires agencies to subject a new reporting requirement to notice and comment, and to critically analyze the requirements to make sure that it is necessary, compatible with existing requirements, makes allowances for small businesses, not confusing, etc. Also, OMB must sign off on it and assign a control number to the reporting/record requirement.

3. Subpoenas (the power to issue coming from statute) may implicate the Fourth Amendment. Subpoenas must be relevant to the agency's purpose authorized by the statutory grant of subpoena power; and they must be reasonable. Moreover, they must follow the statutorily mandated procedures and limitations.
Public Access

From the Ironically-Named Statute Department . . . : Freedom of Information Act
1. The Freedom of Information Act (FOIA, 5 U.S.C. § 552) represents a strong statement on the side of openness in the balance between the needs for accountability and confidentiality. It gives anyone the ability to obtain public records from an agency. On the other hand, it is often slow in its workings, and it is subject to numerous exemptions.

2. FOIA is highly bureaucratized, and often those seeking information do better going through informal channels. Formally, though, agencies must have their own rules (promulgated pursuant to FOIA) telling would-be FOIA requestors the who, what, and where of the agency’s records.

3. FOIA requires that the agency reply to a request within 20 days, a time limit that is widely violated. Courts will allow judicial review after 20 days, and the agency's ability to collect fees is affected if they violate the limit, but most requestors decide just to wait rather than alienate the government and spend money on litigation.

4. Agencies expend resources in reviewing requests, finding documents, and copying and shipping documents. Requestors must pay the government for copies after a certain amount. Non-educational/non-scholarly/non-news requestors must pay some search costs too. Commercial requestors--who represent the bulk of FOIA requests--must pay for all of this plus some review costs.

5. FOIA has its own judicial review provisions that, unlike the APA default provisions, put the burden of proof on the agency rather than the proponent. Review is de novo, and not limited to the record, and the agency pays fees and costs if it loses.

6. A request must be reasonable in its specificity. This includes not just clarity in what is being sought, but also with reference to the index: agencies have indexes that spell out what records they have and where, and the request must track the index to make it reasonable for the agency to be able to find the documents.

7. Agencies only must turn over "agency records." Whether something is an agency record depends on balancing whether the agency created, stores, controls, and/or uses the record.

8. There are numerous exemptions from FOIA, including national security, internal personnel rules or agency practices, explicit statutory exemption, trade secrets and confidential business information, privileged material, privacy-sensitive material, and law enforcement materials.

9. The most complicated FOIA exemption is for confidential business information. Something is exempt under this provision if it was either (1) voluntarily given to the government but is not of a sort ordinarily disclosed to the public; or (2) required to be given to the government but its release would cause substantial competitive harm to the submitter.
Public Access

Freedom from information: reverse FOIA
Sunshine, openness, and how agencies avoid them I (FACA)
1. FOIA exemptions mean that the agency is not required to release the document to the requester. But they do not forbid the agency from releasing the document anyway, if it so chooses. In making the decision of whether or not to release a document, E.O. 12600 requires agencies to set up a process to (in some cases) notify and get the views of the entity that originally submitted the document.

2. So-called "reverse FOIA" suits allow the submitter of information to sue the government to prevent disclosure. FOIA only provides for when disclosure is required, not when it is forbidden, so agencies may disclose information that falls within the FOIA exceptions, so it is never the basis of a reverse FOIA suit. But if the submitter has some other basis for requiring non-disclosure (e.g., the Trade Secrets Act, 18 U.S.C. § 1905), he can use APA § 702 as a cause of action to prevent disclosure.

3. The Federal Advisory Committee Act (FACA), 5 U.S.C. App. 2, § 1 et seq., requires that advisory committees including non-governmental employees be created only after a showing that the public interest requires it, with a charter setting forth its organizing principles, and a sunset date. Membership must be balanced and inclusive, there must be agency oversight and control of the committee, and the meetings must be noticed and open. Advice from individuals, or from bona fide government employees, or communications regarding facts instead of opinion are exempt.
Public Access

Sunshine, openness, and how agencies avoid them II (Sunshine Act)

Attorneys Fees

Getting paid to sue the government I (EAJA; prevailing parties)
1. The Government in the Sunshine Act, 5 U.S.C. § 552b, requires that agency meetings be open. "Agency" for these purposes is a collegial decision making body, usually an independent agency, appointed by the President and confirmed by the Senate. Thus the FCC or NLRB are subject to the Act, but the Attorney General meeting with subordinates would not be. A "meeting" under the act is one where official decision making occurs, and where there is a quorum. This leaves a lot out. There are also specific exceptions made for instances where secrecy is more warranted.

2. Several statutes allow a party who prevails in a case (or in part of one) against the government to get the government to pay the party's legal costs. Besides specific statutes where Congress wants to help motivate lawyers to take cases, there is EAJA, which provides a general cause of action. In EAJA, however, one can only win fees upon prevailing and the government being unable to establish that its position was substantially justified.

3. In all of these statutes, being a prevailing party requires some level of judicial imprimatur be placed on the victory--a judgment, a consent decree, or possibly even a settlement in which the court retains jurisdiction all suffice; a plain settlement or a voluntary cessation of the complained-of conduct do not suffice (though Congress recently broadened FOIA’s fee-shifting provision to allow victories here).
Attorneys Fees

Getting paid to sue the government II (substantial justification; calculating awards)

Admin. Law Practice

Professional responsibility issues for agency lawyers
1. For EAJA purposes, the government will have an easier time avoiding paying fees if it can show that its position was supported by a lower court or by one side of a body of conflicting case law. This will not necessarily amount to "substantial justification," but it helps.

2. EAJA allows attorney's fees in the amount of $125 per hour, adjusted for inflation since 1996 (around $175-200 now). Additional amounts are available for cases in which the supply of qualified lawyers is severely limited or when the lawyer has a special non-legal qualification. Some courts allow higher amounts for special legal qualifications, but this requires something arcane--both in terms of the lawyer's skill and the requirements of the case--not just generally deep experience.

3. Government lawyers owe their client the same duty that private lawyers owe their clients. The client is the agency, though, and so the "client's interests" are infused with distinctive considerations. Part of this is because "the client" for an agency lawyer is more concerned with the public interest than the typical private client, and part is because agencies appear in court a lot and have an interest in taking more credible, less "out there" positions.
Supporting users have an ad free experience!