George Mason
Law Review

Why We Need Federal Administrative Courts

Michael S. Greve
Volume 28
Issue 2

In 2004, Michael and Chantelle Sackett acquired a parcel of land near Priest Lake, Idaho, and obtained local permits to build a home in a built-out subdivision zoned for residential construction. In 2007, when the Sacketts began construction, EPA officials ordered them to stop work and then sent them a “compliance order” claiming that the property contained a federally protected wetland. The order demanded costly restoration work, a three-year monitoring program during which the property must be left untouched, and off-site mitigation and substantial fines. Failure to comply, the EPA warned the Sacketts, might entail civil penalties up to $75,000 per day, as well as criminal sanctions.

The Sacketts’ administrative complaints were met with no meaningful reply. Their subsequent federal lawsuit, contending that the EPA lacked jurisdiction over the property, was met with the agency’s objection that the compliance order was not a final agency action and therefore not subject to judicial review at all. In 2012, a unanimous U.S. Supreme Court held that the EPA’s order was indeed subject to judicial review.

Upon remand to the district court, further litigation ensued. In 2019, the district court determined that under the deferential standards of review that apply to the EPA’s interpretation of the Clean Water Act, as well as the agency’s own regulations and guidance documents, the EPA’s wetlands determination was supported by adequate record evidence and neither arbitrary nor capricious. The ruling is pending on appeal.

The Supreme Court’s decision is Sackett v. EPA.1566 U.S. 120 (2012). The district court’s opinion is unreported.2Sackett v. EPA, No. 08-CV-00185, 2015 WL 1825522 (D. Idaho Apr. 21, 2015).


We in the United States pride ourselves on our independent judiciary, and rightly so. In certain respects, though, it is a bit of a myth. Apart from criminal proceedings and a small enclave of constitutional rights, virtually all disputes between citizens and federal agencies are decided by agency-appointed tribunals, not judges. Typically, the tribunals’ decisions are reversible by the agency and then subject only to highly deferential judicial review. This “appellate review” regime originated over a century ago, in the battle over administrative determinations of railroad rates.3Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum. L. Rev. 940, 956–59 (2011). It found its canonical formulation in the Supreme Court’s foundational decision in Crowell v. Benson4285 U.S. 22 (1932). and, in 1946, was effectively codified in the Administrative Procedure Act (“APA”).5Administrative Procedure Act, Pub. L. No. 79-404, 60 Stat. 237 (1946) (codified as amended at 5 U.S.C. §§ 551–559). The model has since been overlaid with a blanket of administrative common law, and it has been subject to a great deal of improvisation outside the APA’s default provisions.6See, e.g., Gillian E. Metzger, Foreword, Embracing Administrative Common Law, 80 Geo. Wash. L. Rev. 1293, 1295 (2012) (arguing that judicially created “administrative law doctrines . . . venture too far afield from statutory text or discernible legislative purpose”). Despite the constant tinkering, however, and despite widely shared misgivings about its constitutional foundations and practical operation, the appellate review model has proven immune to serious challenge.

The APA’s seventy-fifth anniversary provides an occasion to rethink that model and to mount that challenge. This Article confronts the appellate review model with a stark but realistic and compelling alternative: a system of independent administrative courts, endowed with the institutional capacity and incentives to provide meaningful protection for citizens’ rights.

Many countries in the world feature such a judiciary.7France, Italy, and Germany are among those countries. Many other countries have adopted models of adjudication that lie somewhere between the extremes of independent administrative courts and our appellate review model. For a useful discussion, see Michael Asimow, Five Models of Administrative Adjudication, 63 Am. J. Compar. L. 3, 15–24 (2015). Prominent among them is Germany’s system of administrative courts. It rests on constitutional commands that categorically forbid administrative tribunals and instead require that disputes between the executive and private citizens must always be adjudicated by an independent court. While we cannot simply import that model, we can replicate its essential features, well within the confines of the Constitution and our legal traditions. In fact, as this Article will argue, institutional reforms along German lines might help us recover foundational domestic legal traditions that the U.S. system lost or abandoned by adopting the appellate review model.

Agency adjudication,8In a narrow sense, “agency adjudication” means administrative hearing procedures and dispute resolution; in a broader sense, it includes enforcement actions and any other individual agency action with external legal effect. Cf. 5 U.S.C. § 551(6)–(7). I will use the term in both senses; the intended meaning will appear from the context. the principal subject of this Article, has re-emerged as a prominent topic of scholarly debate9See Kent Barnett, Against Administrative Judges, 49 U.C. Davis L. Rev. 1643 (2016) [hereinafter Barnett, Against Administrative Judges]; Kent H. Barnett, Some Kind of Hearing Officer, 94 Wash. L. Rev. 515 (2019); Jonah B. Gelbach & David Marcus, Rethinking Judicial Review of High Volume Agency Adjudication, 96 Tex. L. Rev. 1097 (2018); Thomas W. Merrill, Fair and Impartial Adjudication, 26 Geo. Mason L. Rev. 897 (2019); Christopher J. Walker, Constitutional Tensions in Agency Adjudication, 104 Iowa L. Rev. 2679 (2019); Christopher J. Walker & Melissa F. Wasserman, The New World of Agency Adjudication, 107 Calif. L. Rev. 141 (2019). The recent literature on the subject features proposals, similar in many respects to the project of this Article, to replace agency adjudication with independent courts. See Steven G. Calabresi & Gary Lawson, The Depravity of the 1930s and the Modern Administrative State, 94 Notre Dame L. Rev. 821, 86465 (2018); Michael B. Rappaport, Replacing Agency Adjudication with Independent Administrative Courts, 26 Geo. Mason L. Rev. 811 (2019). For a brief discussion of some of those proposals, see infra notes 165174 and accompanying text. and hard-fought litigation.10See, e.g., Kisor v. Wilkie, 139 S. Ct. 2400 (2019) (54 decision); Lucia v. SEC, 138 S. Ct. 2044, 2049 (2018) (discussed infra notes 61–63 and accompanying text); Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, 138 S. Ct. 1365 (2018). In some measure, the renewed attention has been prompted by practices that are widely perceived as problematic, irregular, or even abusive. Examples include the imposition of civil fines by bureaucratic edict;11Aaron L. Nielson, D.C. Circuit Review – Reviewed: “I vote for Chenery I, not Chenery II, Yale J. on Regul.: Notice & Comment (Nov. 24, 2017), “non-final” enforcement actions that effectively thwart private citizens’ businesses or use of their land;12See U.S. Army Corps of Eng’rs v. Hawkes Co., 136 S. Ct. 1807, 1813 (2016); Sackett v. EPA, 566 U.S. 120, 130–31 (2012); Rhea Lana, Inc. v. Dep’t of Labor, 824 U.S. 1023 (D.C. Cir. 2016). the holdup of permits or licenses, and their conditioning on well-nigh extortionate demands;13Philip Hamburger, The Administrative Threat 40–42 (2017); Mark Fenster, Failed Exactions, 36 Vt. L. Rev. 623, 624–25 (2012). sudden changes of agency policy,14See Note, Judicial Review of Agency Change, 127 Harv. L. Rev. 2070, 2081–83 (2014). accomplished by means of adjudication and without fair warning to the parties;15Aaron L. Nielson, Visualizing Change in Administrative Law, 49 Ga. L. Rev. 757, 790 (2015). the opportunistic shifting of enforcement proceedings from Article III courts into agency tribunals;16See Jean Eaglesham, SEC Wins with In-House Judges, Wall St. J. (May 6, 2015, 10:30 PM), and the administrative “death squadding” of invention patents that cannot be canceled in any U.S. court.17See, e.g., Alden Abbott, Erika Lietzan, Adam Mossoff, Kristen Osenga, Brian O’Shaughnessy, Hon. Randall R. Rader & Robert Stien, Crippling the Innovation Economy: Regulatory Overreach at the Patent Office 12 (2017),; Tony Dutra, America Invents Act Post-Grant Oppositions After Two Years: Benefit or ‘Death Squad’?, Pat. Trademark & Copyright J. (Sept. 19, 2014),; Tony Dutra, Rader Regrets CLS Bank Impasse, Comments on Latest Patent Reform Bill, Pat. Trademark & Copyright J. (Nov. 1, 2013), (quoting Randall Rader, former chief judge of the Federal Circuit, describing PTAB adjudicators as “acting as death squads, killing property rights”). However, concerns of this sort (not universally shared) partake of a broader, more fundamental, and constitutionally grounded critique of administrative government.

Begin with a rock-bottom proposition: in matters of private right, disputes between citizens and the government must be adjudicated by an independent judge.18The distinction between private and public rights and the corollary question of when and to what extent Congress may commit adjudication to administrative rather than Article III adjudication have received notoriously inconsistent treatment in Supreme Court jurisprudence. With all due caution, though, the Court’s recent decision and opinions in Oil States Energy Services, LLC v. Greene’s Energy Group, LLC, 138 S. Ct. 1365 (2018), suggest that six (and now perhaps seven) justices hold the position that private rights can be adjudicated conclusively only in Article III courts. Justice Gorsuch’s dissent, joined by Chief Justice Roberts, stated the position explicitly and claimed that the Court’s majority “does not quarrel with this test.” Oil States Energy Servs., 138 S. Ct. at 1381 (Gorsuch, J., dissenting). But see id. at 1379 (Breyer, J., concurring) (“[T]he Court’s opinion should not be read to say that matters involving private rights may never be adjudicated other than by Article III courts, say, sometimes by agencies. Our precedent is to the contrary.” (first citing Stern v. Marshall, 564 U.S. 462, 494 (2011); and then citing Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 853–56 (1986))). However, it turns out that aside from criminal convictions and outright violations of constitutional right, virtually no government interference with citizens’ private sphere of conduct is a matter of private right.19Here and throughout the Article, any account of the public rights “exception” refers to matters within the scope of administrative law. Tort and contract disputes with the federal government involve different questions. Congress may superintend interstate commerce as a matter of public right and commit disputes arising under regulatory regimes to administrative agencies, at least so long as those agencies are governed by tolerably fair procedures and some form of judicial review, however deferential, remains available.20See, e.g., Yakus v. United States, 321 U.S. 414, 433 (1944). That, in a nutshell, is the appellate review rule, to which Article III adjudication is the exception.21See Merrill, supra note 9, at 905 (“Over time . . . the Court came to understand the public rights exception as covering virtually everything other than claims of private right grounded in state law, such as contract and tort claims.”); Gillian E. Metzger, Foreword, 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1, 22 (2017) (noting the Supreme Court has “acknowledged that public rights do not require Article III adjudication, and [has] appeared to sanction a broad definition of public rights”); Joshua I. Schwartz, Nonacquiescence, Crowell v. Benson, and Administrative Adjudication, 77 Geo. L.J. 1815, 1839 (1989) (“[M]ost administrative adjudication f[alls] within the public rights category.”).

The model runs up against the intuition that we cannot possibly entrust a single government actor (an administrative agency) to write the rules, to prosecute you under those rules, and to then adjudicate the dispute.22Actually, we can. And we can then commandeer federal and even state courts to make the rulings stick. So says the law. See Testa v. Katt, 330 U.S. 386, 390–91 (1947); Yakus, 321 U.S. at 433–34. And, the argument continues, neither the procedural protections of the APA nor deferential judicial review can cure that basic rule-of-law defect.23See, e.g., Calabresi & Lawson, supra note 9, at 864–65; Philip Hamburger, Chevron Bias, 84 Geo. Wash. L. Rev. 1187, 1196–97 (2016). That line of attack, surely, has force. It resonates with deeply held convictions that we cannot “sport away” the rights of individuals quite so easily.24Marbury v. Madison, 5 U.S. (1 Cranch) 137, 166 (1803).

The dominant, functionalist defense of the appellate review model is that agency adjudication is constitutionally acceptable in some version—with this proviso or that tweak.25See, e.g., Richard H. Fallon, Jr., Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915, 946–47 (1988). A far more robust defense has come from a scholar who is utterly beyond suspicions of harboring functionalist sympathies: John Harrison, Public Rights, Private Privileges, and Article III, 54 Ga. L. Rev. 143 (2019). And in any event, it is the best we can do under modern conditions of social complexity, rapidly changing circumstances, and mass adjudication.26See, e.g., Paul R. Verkuil, Valuing Bureaucracy: The Case for Professional Government (2017); Adrian Vermeule, Law’s Abnegation: From Law’s Empire to the Administrative State 25 (2016); Jody Freeman, Private Parties, Public Functions and the New Administrative Law, 52 Admin. L. Rev. 813, 816 (2000). Those arguments, too, have force. Who in their right mind would want a relative handful of Article III courts deciding, in the first instance or on de novo review and perhaps with a jury on the premises, millions of disability, veterans’, Social Security, workmen’s compensation, or asylum claims? Reimbursement rates for Medicaid providers? Patent claims? Please.

This Article confronts the opposing sides with the omitted third option: a system of independent administrative courts.

The functionalist “best-we-can-do-under-modern-conditions” set of arguments rests on two suppressed, interconnected premises—one conceptual, the other institutional. The conceptual premise is the encompassing sweep of the public rights “exception,” coupled with what Professor Mila Sohoni has called the APA’s “rights neutrality.”27Mila Sohoni, Agency Adjudication and Judicial Nondelegation: An Article III Canon, 107 Nw. U. L. Rev. 1569, 1581 & n.60 (2013). “Rights neutrality” is a bit of an oversimplification. See discussion infra Part I.A. and accompanying notes. However, the nuances do not affect the point in the text. According to that argument, as a matter of general administrative law doctrine, it makes no difference whether a government agency denies disability benefits or prohibits private citizens from building a house on their own land. It is against this conceptual backdrop that the functionalist view gains undeniable plausibility and, moreover, ties in with institutional considerations. The functionalist defense of the appellate review model looks well-nigh unanswerable (1) within a rights-neutral conceptual framework and (2) given a stark institutional alternative between prompt, expert administration and laborious Article III adjudication. Take a transatlantic step outside that framework, though: the appellate review model and its defense look deeply suspect.

In here-relevant respects Germany is a country much like ours: a complex, fast-paced society, with governmental programs as generous and ambitious as ours.28Too, and at the risk of cultivating stereotypes, the Germans appear to put a higher premium on administrative efficiency and expertise than do we. And yet, conceptually and institutionally, German administrative law is the polar opposite to our appellate review model.29For detailed explanation of the following two paragraphs, see infra Part II.

The irreducible, constitutionally grounded command of German administrative law is that every administrative act that interferes with citizens’ private sphere of action must be subject to de novo determination by an independent court.30Verwaltungsgerichtsordnung [VwGO] [Code of Administrative Court Procedure], as amended, art. 5, Anything resembling our form of administrative adjudication is categorically prohibited. The legal system is decidedly not rights neutral. It is organized, to the point of monomania, around private rights, not (as ours is) around the administrative “process.”31See infra notes 87–90 and accompanying text. Its constitutionally grounded purpose is to permit private citizens to go about their lives without undue government interference. The German term for that command is the allgemeine Handlungsfreiheit, meaning the general freedom to do as you wish until and unless the government tells you, with persuasive and constitutionally permissible reasons, to cut it out.32See Grundgesetz [GG] [Basic Law], art. 2, para. 1, translation at

The focus on rights in turn drives the institutional arrangements. If you want a judiciary that will guard against deprivations of right—not just sporadically, but on a systematic basis—you need enough judges to perform that task. Accordingly, Germany has built a robust system of independent administrative courts. Those courts adjudicate nothing but questions of private right—as distinct from, say, “public interest” complaints over an agency’s failure to follow proper procedures or to act with sufficient speed.33Id. art. 19, para. 4, art. 101, para. 1; Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] Apr. 17, 1991, 84 Entscheidungen des Bundesverfassungsgerichts [BVerfGE] 34, 58; Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] Nov. 12, 1958, 8 Entscheidungen des Bundesverfassungsgerichts [BVerfGE] 274, 332. But they adjudicate all such disputes de novo on all questions of law and fact.34See infra note 131 and accompanying text. And somehow, that system seems to work, better perhaps than ours, along all relevant dimensions—constitutionality, lawfulness, functionality, and public acceptance.35The emphasis is on “seems”: robust evidence is hard to come by. Anecdotally, though: one hears and reads quite a bit about excessive bureaucratic rigidity and “judicialization” in Germany. See, e.g., Friedrich Schoch, Gerichtliche Verwaltungskontrollen, in 3 Grundlagen des Verwaltungsrechts 743, 753 (Wolfgang Hoffman-Riem, Eberhard Schmidt-Assman & Andreas Vosskuhle, eds., 2d ed. 2012) [hereinafter Grundlagen] (noting scholarly criticism of the “Justizstaat”). Complaints about irregularity, unfairness, and inadequate legal protection, akin to those that run through the American debate, appear exceedingly rare. Thus, the defense of the appellate review model—against full-scale judicial adjudication—must rest on normative convictions about rights. It cannot rest on arguments about the ineluctable demands of the modern administrative state.

This Article hopes to show that it would be entirely possible, and highly desirable, to adapt the German model to our legal and institutional traditions and to substitute it for the appellate review model, over a wide range. To be sure, that enterprise would require a substantial legislative revision of the APA and amendments to significant number of agency organic statutes. It would require an extended debate, on the scale and of the duration of the debate that preceded the APA.36Cf. George B. Shepherd, Fierce Compromise: The Administrative Procedure Act Emerges from New Deal Politics, 90 Nw. U. L. Rev. 1557, 1560 (1996). But that debate is well underway.37See, e.g., Richard A. Epstein, The Dubious Morality of Modern Administrative Law (2020); Metzger, supra note 21; Jeffrey A. Pojanowski, Neoclassical Administrative Law, 133 Harv. L. Rev. 852 (2020); Adam White, Doing Justice to APA Modernization, Yale J. on Regul.: Notice & Comment (Dec. 16, 2019), (describing a summit convened by the U.S. Department of Justice to “modernize” the Administrative Procedure Act). This Article aims to reorient it in two ways: (1) away from questions of administrative procedure, and towards questions of private right; and (2) away from doctrinal questions—especially the “standard of judicial review”—towards institutional questions.

Therein lies a challenge not only to defenders of the appellate review regime but also its opponents. There are two doctrinal ways to contest that regime: one is to restrict its range, the other is to re-constitutionalize its operation. The former strategy calls for—at the end of the day—a re-assertion of “substantive due process” rights;38Harrison, supra note 25, at 207–09. the latter, for overturning judicial deference canons.39See, e.g., Hamburger, supra note 23, at 1188–89; see generally Christopher J. Walker, Attacking Auer and Chevron Deference: A Literature Review, 16 Geo. J.L. & Pub. Pol’y 103 (2018). While I am sympathetic to both lines of argument, both confront the objection that Article III courts as currently constituted lack the institutional capacity and the incentives to exercise meaningful control over agency adjudication. If you want to rethink and revise administrative law doctrines from the ground up, you’ll want to establish institutions that might be up to that task. You’ll want administrative courts.

Part I of this Article provides a brief description of the U.S. system of administrative adjudication and a conventional restatement of the central rule-of-law problem—that is, the conjunction of prosecutorial and judicial functions within a single government body. Part II describes the German system. While the hyper-conceptualism of that corpus juris will seem alien and perhaps obsessive to American lawyers, I hope to elucidate two central, already-mentioned themes. One is the real-world viability, and desirability, of an administrative law regime that pivots on the notion of private right. The other is the intimate connection between law and institutions: a rights-protective legal regime requires independent courts capable of providing that protection. Part III describes what a system of independent administrative courts, adapted to the U.S. constitutional system, might look like. The concluding Part IV returns to the what-can-we-learn theme of this Introduction. Again, we cannot airdrop Germany’s legal and institutional system into the United States. We can adapt it, however, and the comparative inquiry may help to reorient our domestic debate to the very the first question of our very first great administrative law case.40“1st: Has the applicant a right to the commission he demands?” Marbury v. Madison, 5 U.S. (1 Cranch) 137, 154 (1803). Cf. Thomas W. Merrill, Marbury v. Madison as the First Great Administrative Law Decision, 37 J. Marshall L. Rev. 481, 481 (2004).

I.     Administrative Justice and Its Critics

A.     The “Hidden Judiciary”

Under the appellate review regime, agencies internally conduct virtually all administrative adjudication in the federal system. There are fewer than 1,000 Article III judges, and administrative cases constitute only a small portion of their dockets.41See Annual Report of the Director: Judicial Business of the United States Courts, U.S. Courts tbl.1.1 (2018), (listing the number of federal judgeships). For statistics on judicial caseloads and types of cases, see Federal Judicial Caseload Statistics 2018, U.S. Courts (2018), In contrast, there are well over 10,000 agency adjudicators who decide millions of cases each year.42Barnett, Against Administrative Judges, supra note 9, app. A. Information concerning ALJ caseloads is available through the Office of Administrative Hearings. Off. of Admin. Hearings, The Office of Administrative Hearings: Biennial Report for Fiscal Years 2013–14, at 8 (2014), They adjudicate challenges to agency decisions, enforce orders and impose fines, issue or withhold permits or licenses, and in some cases, decide disputes among private parties. Some 2,000 are Administrative Law Judges (“ALJs”)—the great majority of them (85%) serve in the Social Security Administration.43Walker & Wasserman, supra note 9, at 154. Other administrative judges (“AJs”) bear various titles, such as “immigration judge.”44Barnett, Against Administrative Judges, supra note 9, at 1659. While their precise number is unknown, there are more than five times as many AJs as there are ALJs.45Walker & Wasserman, supra note 9, at 154.

Both ALJs and AJs are attached to and appointed by a particular agency.46For more detailed description and discussion, see Barnett, Against Administrative Judges, supra note 9, at 1654, 1659–60. ALJs were until recently screened and selected by the Office of Personnel Management (“OPM”); agencies then appointed ALJs from the OPM’s list. However, in the wake of the Supreme Court’s Lucia decision (briefly discussed infra notes 61–63 and accompanying text), the Trump administration issued an Executive Order exempting ALJs from the competitive service. See Exec. Order No. 13,843, 83 Fed. Reg. 32,755 (July 10, 2018). Most enjoy a measure of functional independence (such as removal and salary protections), and some agency adjudications—especially those conducted under the APA’s provisions for “formal” adjudication475 U.S.C. §§ 554, 556.—are heavily proceduralized. Formal adjudication generally requires an ALJ to preside over hearings and provides ALJ independence from ex parte communications, agency oversight, and the obligation to perform other agency duties.48Id. §§ 557, 3105. Parties enjoy substantial procedural protections, including notice of the proceedings, legal counsel, presentation of their case orally or in writing, submission of findings of fact and law, and a reasoned decision supported by substantial evidence in the record as a whole.49Id. §§ 556–557, 706; Barnett, Against Administrative Judges, supra note 9, at 1699 (“As other studies have demonstrated, the administrative state is balkanized with numerous varieties of hearings, although the trend is towards increased formality.” (footnote omitted)). However, AJ and ALJ decisions are almost always subject to review and reversal by agency heads.50Barnett, Against Administrative Judges, supra note 9, at 1655. This prospect of reversal distinguishes administrative judges from actual courts and judges.

Generally, parties must exhaust administrative procedures before contesting a final, adverse administrative decision in an Article III court.51See Sohoni, supra note 27, at 1572. Those challenges are “review” proceedings—“re-view,” that is—of an adjudicatory decision that has already been made at least once. Review proceedings are limited to the administrative record.52§ 706. Generally, the standard of judicial review depends on the form (or the degree of formality) of the agency action, not on how much or what is at stake for the private litigant.53See, e.g., FCC v. Fox Television Stations, Inc., 556 U.S. 502, 536–37 (2009) (Kennedy, J., concurring in part and concurring in the judgment). And it is highly deferential. Courts will overturn agency findings of fact only if they are “arbitrary and capricious” or bereft of “substantial evidence.”54Id. at 513 (majority opinion) (citing § 706(2)(A)); see also § 706(2)(E). If the agency acts with sufficient formality and with the force law, courts will also defer to the agency’s legal interpretation of ambiguous statutes and regulations.55See Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 844 (1984); see also Kisor v. Wilkie, 139 S. Ct. 2400, 2408 (2019); United States v. Mead Corp., 533 U.S. 218, 221 (2001); Auer v. Robbins, 519 U.S. 452, 461 (1997).

How well has this worked? By some measures, the “hidden judiciary”56Barnett, Against Administrative Judges, supra note 9, at 1645 (quoting Chris Guthrie, Jeffrey J. Rachlinski & Andrew J. Wistrich, The “Hidden Judiciary”: An Empirical Examination of Executive Branch Justice, 58 Duke L.J. 1477, 1478 (2009)). has operated adequately, or at least sufficiently well to remain out of sight and out of mind. On second thought, however, that may not be saying much: an agency’s adjudicative machinery has to do something truly dumb or suffer a near collapse to attract public attention, judicial suspicion, and critical scholarly discussion. Breakdowns have most commonly occurred in adjudication systems that handle a very large volume of cases, such as the immigration system57Under the Bush administration, newly hired immigration judges were former enforcement personnel; others were poorly qualified. Their bias against asylum seekers and prospective immigrants was sufficiently virulent to prompt federal appellate courts to reverse administrative asylum and deportation decisions at an astounding rate and in very harsh language. Barnett, Against Administrative Judges, supra note 9, at 1686. More recently, a massive backlog of asylum and deportation cases has again drawn public attention. See, e.g., Maria Sacchetti, Immigration Judges’ Union Calls for Immigration Court Independent from Justice Department, Wash. Post (Sept. 21, 2018, 6:46 PM), and the Social Security Administration.58Under the Obama administration, newspapers generated stories about ALJs who approved just about every disability benefit application to cross their desks. Damian Paletta, New Report Could Increase Scrutiny of Social Security Disability Judges, Wall St. J. (Nov. 14, 2014, 6:11 PM), In other instances, agencies with smaller portfolios have been suspected of abusing their adjudication system for political, agenda-driven purposes.59The Securities Exchange Commission (“SEC”) has cranked larger numbers of enforcement cases through its own adjudication system rather than the federal courts—arguably, for the purpose of obtaining easier convictions. Eaglesham, supra note 16. But see Urska Velikonja, Are the SEC’s Administrative Law Judges Biased? An Empirical Investigation, 92 Wash. L. Rev. 315, 321 (2017) (criticizing the findings of the Wall Street Journal study); David Zaring, Enforcement Discretion at the SEC, 94 Tex. L. Rev. 1155, 1199–200 (2016). The National Labor Relations Board (“NLRB”) has attempted to engineer far-reaching policy changes by means of adjudication under circumstances that suggest a pronounced anti-employer bias and partisan machinations. See Theodore J. St. Antoine, The NLRB, The Courts, The Administrative Procedure Act, and Chevron: Now and Then, 64 Emory L.J. 1529, 1550 (2015).

Irregularities of this sort may be episodic. Underneath the surface, however, lurk graver and more enduring doubts. Some arise from separation-of-powers concerns, while others have to do with elementary notions of due process and constitutional government.

B.     Administrative Adjudication and the Rule of Law

The appellate review system sits and fits somewhat uneasily within the constitutional system. It substitutes functionally independent agency adjudicators for independent Article III judges, agency procedures for due process, and a credible appearance of impartiality for institutionally cemented impartiality. Despite all high hopes and fond wishes, though, the contending forces of society have never come to rest entirely on that compromise. Lately, it has shown signs of increased strain.

One pressure point is the Supreme Court’s increasingly formalistic separation-of-powers jurisprudence.60See Kent Barnett, Resolving the ALJ Quandary, 66 Vand. L. Rev. 797, 863 (2013); Walker, supra note 9. In Lucia v. SEC,61138 S. Ct. 2044 (2018). the U.S. Supreme Court held that (certain) AJs are officers of the United States.62Id. at 2050–51. They must therefore be appointed in conformity with Article II, either by the President or the heads of “their” agencies. Put that holding together with the Court’s earlier pronouncements on the constitutional command to preserve presidential control over officer removal: the AJs and ALJs functional independence looks increasingly doubtful. In his Lucia concurrence, Justice Breyer warned (not for the first time) against a jurisprudence that risks “unraveling, step-by-step, the foundations of the Federal Government’s administrative adjudication system as it has existed for decades.”63Id. at 2064 (Breyer, J., concurring in the judgment in part and dissenting in part); see Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 519 (2010) (Breyer, J., dissenting); see also Kent H. Barnett, Regulating Impartiality in Agency Adjudication, 69 Duke L.J. 1695, 1711–14, 1719, 1728 (2020) (discussing the acute tension between Article II and due process requirements in the wake of Lucia); Linda D. Jellum, “You’re Fired!” Why the ALJ Multi-Track Dual Removal Provisions Violate the Constitution and Possible Fixes, 26 Geo. Mason L. Rev. 705, 733, 741–45 (2019); Walker, supra note 9, at 2683.

Meanwhile, scholars of a disposition very different from Justice Breyer’s have argued that “administrative justice” is an oxymoron and unconstitutional over a wide range of cases.64See, e.g., Philip Hamburger, Is Administrative Law Unlawful? 1–2 (2014); Calabresi & Lawson, supra note 9, at 865 (“When dealing with executive deprivations of rights, ‘procedural due process,’ not ‘substantive due process,’ is the oxymoronic phrase.”); Nathan S. Chapman & Michael W. McConnell, Due Process as Separation of Powers, 121 Yale L.J. 1672, 1720 (2012); Gary Lawson, Delegation and Original Meaning, 88 Va. L. Rev. 327, 334–35 (2002); Ilan Wurman, Constitutional Administration, 69 Stan. L. Rev. 359, 362 (2017). The principal rule-of-law difficulty is usually conceptualized as a due process concern: we cannot have the same set of officials sit as prosecutors and judges. In some settings, that intuition still dominates. For example, we would be horrified if the Department of Justice could adjudicate criminal offenses inside the agency, subject to direction and reversal by the Attorney General, and with near-conclusive effect in federal court.65Fallon, Jr., supra note 25, at 952 n.208 (“[C]riminal cases traditionally have been regarded as requiring judicial resolution . . . .”); Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559, 618 (2007); Sohoni, supra note 27, at 1578 & n.37, 1611–12 (noting the criminal law exception from the APA’s “transsubstantive” reach and ambition). Nor would we not change our minds on that score if the Department’s internal arrangements conformed to all the niceties of “process.” Ultimately, due process is not just about the “how” but also and mostly about the “who.” The agent who prosecutes you cannot convict you.66For extended discussion, see Chapman & McConnell, supra note 64 (reviewing the history and application of due process). That is why we have independent courts; that is how “the judicial Power” conferred in Article III hangs together with the Due Process Clause.

Why do these basic intuitions give way so readily to an acceptance of administrative adjudication well-nigh across the rest of the board? Perhaps, no judicial determination in the first instance is required when the dispute is not about sending someone to jail but about something considered less serious, including private citizens’ money or land.67This proposition partly explains why the imposition of even very substantial civil fines does not require Article III adjudication in the first instance. See Atlas Roofing Co. v. Occupational Safety Health Rev. Comm’n, 430 U.S. 442, 449–51, 455 (1977). However, the “criminal exception” from the APA and Chevron canons is actually quite hard to square with the transsubstantive, “rights neutral” presumptions of general administrative law. For discussion, see Sohoni, supra note 27, at 1580–84, 1611–12. See also, e.g., Carter v. Welles-Bowen Realty, Inc., 736 F.3d 722, 730–32 (6th Cir. 2013) (Sutton, J., concurring). Moreover, short of criminal convictions or messing with constitutional rights, the APA model presumes that we want agencies to make policy by means of adjudication, to better and more flexibly to employ their expertise. If so, the adjudicators must be inside the agencies and reversible by agency heads; and reviewing courts must grant deference, lest they rather than the agencies wind up as policy-makers-in-chief.

Might it be unfair to subject regulated parties to adjudication by decision makers who are appointed by the regulating agency, imbued with its mission and institutional perspective, and reversible by that agency? Why, yes. That is why the APA contains formal adjudication procedures and why statutory judicial review has come to be overlaid with a “presumption of reviewability” of uncertain origins.68See Nicholas Bagley, The Puzzling Presumption of Reviewability, 127 Harv. L. Rev. 1285, 1294–95 (2014).

Nonetheless, many scholars have remained uncomfortable with the APA’s “fierce compromise.”69Shepherd, supra note 36, at 1557, 1559–60. Recent reform proposals include more formalized agency procedures,70Barnett, supra note 60, at 832–35. more functional independence for AJs comparable perhaps to that enjoyed until recently by ALJs,71See, e.g., Barnett, Against Administrative Judges, supra note 9, at 1647–49; Barnett, supra note 63, at 1732–34, 1736–40 (proposing increased transparency with respect to adjudicators’ independence); Robert J. McCarthy, Blowing in the Wind: Answers for Federal Whistleblowers, 3 Wm. & Mary Pol’y Rev. 184, 226 (2012). and “fair notice” requirements before an agency changes course by means of adjudication.72Aaron L. Nielson, Visualizing Change in Administrative Law, 49 Ga. L. Rev. 757, 790 (2015). Other scholars have urged more probing judicial review, at least on questions of law.73Kristin E. Hickman & Aaron L. Nielson, Narrowing Chevron’s Domain, 70 Duke L.J. 931, 938–39 (2021) (proposing to narrow Chevron deference to rulemaking proceedings, to the exclusion of adjudicatory proceedings); Jeffrey S. Lubbers, Closing Remarks, 59 Admin. L. Rev. 621, 627 (2007); Schwartz, supra note 21, at 1820–21 (arguing that judicial review must entail “effective supervision . . . of the administrative agencies’ determination of questions of law”); see also sources cited supra note 39.

None of these proposals, however, reach the root of the problem—the agency’s dual role as policy maker and adjudicator. Agencies may provide procedures approaching those of ordinary trial courts but, at the end of the day, the adjudicator is meant to bring a certain bias to the proceeding and to exercise discretion and expertise consistent with the agency’s mission. Thus, the most one can expect from administrative adjudication is an appearance of impartiality.74Cf. Barnett, Against Administrative Judges, supra note 9, at 1671 (arguing that “the mere appearance of impartiality is as salient as actual bias” because important ends are “protect[ing] the integrity of the adjudicating body and validat[ing] the process”). Similarly, within the framework of the appellate review model, the case for judicial deference is overwhelming. Ultimately, the judges are reviewing policy decisions, and in that domain they have no comparative advantage over agency administrators. The cost of running a more legalistic system, if nothing else, will invariably exert a gravitational pull (back) toward deference.75Vermeule, supra note 26, at 2–3. And the reform programs—more judicialized agency procedures, more intense judicial review—pull in opposite directions. The more “process” regulated parties have received within the agency, the less inclined an Article III court will be to second-guess its decision.

The only escape from that conundrum is a judicial system that subjects government action, so far as it interferes with a sphere of ordinary private conduct, to comprehensive, genuinely legal, and independent judicial control. Some modern countries operate just such a system, with a proud insistence that nothing less will satisfy the basic demands of the rule of law. The prototype is Germany’s system of administrative courts. Part II describes those courts and their law.

II.     Germany’s Administrative Judiciary

A.     Conceptual Foundations

The American notion of “judicial review” combines and conjoins two somewhat different orientations. One is to fend off unwarranted government interferences with private conduct, the other is to ensure the legality and regularity of government conduct. The distinction was reasonably clear so long as what we now call “administrative law” was governed by Marbury presumptions concerning the nature of judicial power and the separation of powers. The executive must not “sport away” the private, vested rights of individuals, and the courts’ obligation is to prevent any such occurrence.76Marbury v. Madison, 5 U.S. (1 Cranch) 137, 166 (1803). But that power comes with a built-in limitation: it affirmatively forbids a general judicial superintendence over the executive’s conduct. No private right, no judicial review.

This elementary distinction became blurred in the contentions over railroad rates and independent regulatory commissions in the early twentieth century. It is—or rather has become—thoroughly obscured under the APA, which affords judicial relief to any person who has suffered a “legal wrong” or to anyone “adversely affected or aggrieved by agency action within the meaning of a relevant statute.”775 U.S.C. § 702. The first part of the APA disjunction—“legal wrong”—was intended to capture then-existing forms of non-statutory review, which required some preexisting right; the second part was meant to capture statutory rights of review. However, the distinction became eviscerated after the Supreme Court’s decision in Ass’n of Data Processing Service Organizations, Inc. v. Camp, 397 U.S. 150 (1970). For a splendid account of this development, see Caleb Nelson, “Standing” and Remedial Rights in Administrative Law, 105 Va. L. Rev. 703, 706–09 (2019). It is yet more thoroughly eviscerated in the contemporary body of administrative common law. “Private attorneys general,”78William B. Rubenstein, On What a “Private Attorney General” Isand Why It Matters, 57 Vand. L. Rev. 2129, 2130–31 (2004); see Alyeska Pipeline Serv. Co. v. Wilderness Soc’y, 421 U.S. 240, 258–63 (1975). expansive standing doctrines,79See, e.g., Friends of the Earth Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U.S. 167 (2000); Massachusetts v. EPA, 549 U.S. 497 (2007). pre-enforcement review in rulemaking proceedings,80Daniel Boger, Note, Pre-Enforcement Review: An Evaluation from the Perspective of Ripeness, 36 Va. Env’t L.J. 77, 78–79, 78 n.3 (2018). elaborate agency procedures and “hard look review,”81See, e.g., Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins., 463 U.S. 29, 43–44 (1983); Citizens to Pres. Overton Park, Inc. v. Volpe, 401 U.S. 402, 416 (1971). a “presumption of reviewability”82Guerrero-Lasprilla v. Barr, 140 S. Ct. 1062, 1069–70 (2020) (discussing “the presumption favoring judicial review of administrative action” (quoting Kucana v. Holder, 558 U.S. 233, 251 (2010))).: these and related doctrines all serve to ensure what German lawyers call the “objective legality” (objektive Rechtmässigkeit) of executive conduct—irrespective of whether the alleged illegality interferes with anyone’s private sphere of action, or, in German parlance, “subjective public rights” (subjektive öffentliche Rechte). Such a “public law” model of judicial review practically compels judicial deference, lest the courts become the de facto executive. It also entails rights neutrality. With limited exceptions for constitutional rights and criminal convictions, virtually all rights are “public” and thus subject to agency adjudication and deferential review, subject to whatever Congress may ordain.

In diametrical contrast, the distinction between a public law model that seeks to ensure across-the-board legality and a private rights model that seeks to block government interference with citizens’ private sphere of conduct remains very sharp in German lawyers’ and judges’ minds. Just about everyone agrees that the models are not just different but fundamentally incompatible. That consensus is a product of the nineteenth-century struggle to subject the (royal) executive to legal constraints. It produced—in legal theory and in institutional practice—both models: the private rights solution (often called the “Southern German” model), and the public law solution (the “Prussian” or “Northern German” model).83See Michael S. Greve, The Non-Reformation of Administrative Law: Standing to Sue and Public Interest Litigation in West German Environmental Law, 22 Cornell Int’l L.J. 197, 236–38 (1989). Both models require a very large administrative judiciary, but entail very different doctrinal and institutional consequences.

The Prussian model will at the limit permit anyone to contest the legality of any administrative act for any reason. For instance, if we seek to ensure judicial oversight over the legality and the enforcement of every effluent permit issued to industrial facilities, any member of the public should be able to complain. It makes neither practical nor conceptual sense to require a plaintiff to purchase an airline ticket or to paddle past a permitted power plant to demonstrate an “injury in fact.”84But see, e.g., Summers v. Earth Island Inst., 555 U.S. 488, 494–95 (2009) (requiring particularized injury for standing that is not established by general harm to the environment); Lujan v. Defs. of Wildlife, 504 U.S. 555, 562–63 (1992) (noting that the “injury in fact” test of the standing doctrine requires that the plaintiff is an injured party or is otherwise “‘directly’ affected apart from their ‘special interest’ in th[e] subject” (quoting Sierra Club v. Morton, 405 U.S. 727, 734–35, 739 (1972))). Courts that entertain such “public interest” lawsuits, though, would have to be part of the executive, for elementary reasons of both practicality and the separation of powers. The power to demand the faithful execution and meticulous observance of the laws, and the power to determine whether those duties have been satisfied, is the power to run the system. Conversely, independent courts that operate on the private rights model must not adjudicate anything but disputes over individual rights, lest they lose—in a recurrent formulation—their “distance” (the German word on this rare occasion is shorter: Distanz) from the executive.

The conceptual apparatus of German administrative law, originally developed around the turn of the twentieth century, largely favored the private rights model. In its original formulation, the system had a decidedly positivist tinge, and institutional practice in the German states during the Weimar years remained muddled.85See Greve, supra note 83, at 236–38. However, the Federal Republic’s 1949 Basic Law made an unequivocal decision in favor of the private rights model, and over time, foundational doctrines of administrative law have become constitutionalized or heavily overlaid with constitutionally grounded doctrines.86For a splendid overview of German administrative law and its historical development, see Florian Becker, The Development of German Administrative Law, 24 Geo. Mason L. Rev. 453 (2017). Section B describes the institutional arrangements. Section C outlines some of the principal doctrines. Section D provides a brief account of the contemporary scholarly debate and legal landscape.

B.     The Constitutional and Institutional Framework

The German Constitution, originally enacted as the Grundgesetz (Basic Law) in 1949, provides that the Federal Republic is to be a Rechtsstaat. This means that the legislature is bound by the Constitution,87Grundgesetz [GG] [Basic Law], art. 20, para. 3 (“The legislature shall be bound by the constitutional order, the executive and the judiciary by law and justice.”). that the basic rights listed in the Constitution bind all branches of government as directly applicable law,88Id. art. 1, para. 3 (“The following basic rights shall bind the legislature, the executive and the judiciary as directly applicable law.”). and that all judges must be “independent and subject only to the law.”89Id. art. 97, para. 1; see also id. art. 92; id. art. 99. Most crucially for present purposes, it means that any citizen whose rights are violated by a public authority must have access to an independent court.90Id. art. 19, para. 4 (“Should any person’s rights be violated by public authority, he may have recourse to the courts. If no other jurisdiction has been established, recourse shall be to the ordinary courts.”). This command is reinforced by a guarantee of a “lawful judge” and an explicit prohibition against extra-ordinary courts.91Id. art. 101, para. 1 (“Extraordinary courts shall not be allowed. No one may be removed from the jurisdiction of his lawful judge.”). Article 10 provides a narrow exception for legal restrictions that “serve[] to protect the free democratic basic order or the existence or security of the Federation or of a Land.” Id. art. 10, para. 2.

The Federal Constitutional Court and the administrative judiciary have developed doctrines that give teeth to the principles of the Rechtsstaat. For example, statutes as well as executive rules and regulations with the force of law must meet standards of definiteness (Bestimmtheit) and proportionality, and the legislature may not delegate “essential” decisions to administrative bodies.92For a discussion of these separate requirements, see Franz Reimer, Das Parlamentsgesetz als Steuerungsmittel und Kontrollmassstab, in 1 Grundlagen, supra note 35, at 585, 622–39. Legal process, especially including access to independent courts, must be prompt, comprehensive, and effective.9384 BVerfGE 34, 58; 8 BVerfGE 274, 332.

The constitutional architecture just described categorically forbids anything resembling our ALJs or AJs—that is, administrative “judges” who may be removed by the executive or whose decisions may be directed or revised by the executive.94The constitutional command of an independent judiciary is reiterated in section 1 of the Verwaltungsgerichtsordnung (“VwGO”), the statute that governs the operation of the administrative judiciary: “Administrative jurisdiction shall be exercised by independent courts separated from the administrative authorities.” Verwaltungsgerichtsordnung [VwGO] [Code of Administrative Court Procedure], as amended, § 1, The Federal Administrative Court (Bundesverwaltungsgericht) is an independent court. The justices are appointed by the Minister of Justice and a federal-state committee; like regular judges of the lower courts, they may be removed or reassigned only for exceptional, constitutionally specified reasons.95Grundgesetz [GG] [Basic Law], art. 97, para. 2.

At variance with the U.S. model, there are no lower federal courts in the German system. The lower courts, trial as well as appellate, are state courts. Thus, the Federal Administrative Court sits atop a pyramid of administrative courts that are established by the states.96VwGO § 3. States are required to establish the inferior courts. Their operation, though, is principally governed by the Verwaltungsgerichtsordnung (“VwGO”), a federal statute enacted in 1960 pursuant to the federal legislature’s concurrent jurisdiction.97Grundgesetz [GG] [Basic Law], art. 74, para. 1.

Each of the sixteen states has one or more administrative courts,98North Rhine-Westphalia, for example, a state with a population of some 18 million, has seven administrative courts. See Aufgaben und Aufbau der Verwaltungssgerichtsbarkeit: Vielfältige Aufgaben bei öffentlich-rechtlichen Streitigkeiten, Die Justiz des Landes Nordrhein-Westfalen, as well as a single appellate administrative court. (One of them covers two states, Berlin and Brandenburg.) With a handful of exceptions, the administrative courts serve as courts of first instance in all challenges to administrative action or inaction. Typically, cases are heard by panels consisting of three full-time judges with lifetime appointments99VwGO § 15(1). and two honorary (lay) judges, who serve for a fixed period of years.100The honorary judges’ appointments and duties are governed by VwGO sections 19–34. They are actual judges, not jurors. They participate in all phases of the proceeding and vote on questions of law as well as fact. They must be German citizens, and they may not be lawyers, members of the civil service, legislators, or convicts. Specifics are governed by state law. Certain simple cases may be heard by a single full-time judge. Id. §§ 19–34. Appeals from the administrative courts are limited to cases that pose unusually difficult questions. The standard of review (of the administrative act, not the lower court’s decision) is the same as in the administrative trial courts,101Id. § 128 (“The [appellate court] shall review the dispute within the appeal on points of fact and law application to the same degree as the administrative court. It shall also consider newly-submitted facts and items of evidence.”). and here as there, cases are heard by panels of full-time and honorary judges.102In certain cases—foremost, challenges to large and environmentally sensitive public infrastructure projects (such as power plants and airports) and complex planning and rulemaking proceedings concerning such projects—the appellate courts act as trial courts. See id. § 48(1)–(2). The Bundesverwaltungsgericht will entertain further appeals only in cases that are of a precedential nature and, moreover, turn on questions of federal rather than state law.103See id. §§ 132, 137. In contrast to appellate court proceedings, review (Revision) in the Bundesverwaltungsgericht is limited to questions of law.104In a few cases, the Federal Administrative Court has original jurisdiction. See id. § 50.

The administrative courts decide around 125,000 cases each year.105See Carsten Günther, Ass’n of the Councils of State & Supreme Admin. Jurisdictions of the Eur. Union, Administrative Justice in Europe: Report for Germany 21 (2016), The cases range from the mundane (such as questions of student discipline or promotion) to the momentous, including questions of constitutional right. (The administrative judiciary, not the Federal Constitutional Court, is the principal protector of constitutional rights in the ordinary course of events and government conduct.) The administrative courts’ jurisdiction extends principally to all public-law disputes between citizens and the government, with four general exceptions.106VwGO § 40(1) (“Recourse to the administrative courts shall be available in all public-law disputes of a non-constitutional nature insofar as the disputes are not explicitly allocated to another court by a federal statute.”). For “of a non-constitutional nature,” see infra note 108. The Bundestag has made more-than-occasional (and to many minds ill-advised) use of the “insofar” clause and committed to the ordinary (civil) courts matters that by all rights belong to “core” administrative law, such as antitrust and energy regulation. For discussion and critique, see Schoch, supra note 35, at 799–814 (criticizing the legislature’s and the ordinary courts’ “inflationary” use of the “insofar” exceptions). See also id. at 813–14 (criticizing the “splintering” of judicial controls and the attendant legal uncertainty). First, tax matters are committed to a separate system of tax courts.107Finanzgerichtsordnung [FGO] [Fiscal Code], § 1, Second, benefit determinations under social and health insurance regimes—as one might expect, a large volume of cases—are committed to a separate system of independent “social” courts (Sozialgerichte).108Sozialgerichtsgesetz [SGG] [Social Court Act], Sept. 23, 1975, Bundesgesetzblatt, Teil I [BGBl I], as amended, § 1. There is no theoretically compelling reason to separate the tax and, especially, the “social” courts from the general administrative judiciary; in fact, there are periodic calls to create a unitary administrative judiciary. Nothing has come of it—perhaps because the system works well enough; perhaps because the envisioned change would require an amendment to article 95 of the Grundgesetz. For discussion, see Friedhelm Hufen, Ist das Nebeneinander von Sozialgerichtsbarkeit und Verwaltungsgerichtsbarkeit funktional und materiell begründbar?, 42 Die Verwaltung 405 (2009). Third, certain constitutional questions must be adjudicated by constitutional rather than administrative courts.109Grundgesetz [GG] [Basic Law], art. 93. The text of VwGO section 40 seems to preclude all constitutional claims in administrative courts. However, according to the dominant legal opinion and the courts’ consistent decisions, the preclusion extends only to “immediate” constitutional claims by constitutional actors, which includes political parties and a few other non-governmental actors but not individual citizens or enterprises. For discussion, see Wolf-Rüdiger Schenke, Streitigkeiten verfassungsrechtlicher Art im Sinne des § 40 VwGO, 131 Archiv des öffentlichen Rechts 117 (2006). Fourth, the administrative courts have jurisdiction only when government acts in its sovereign capacity and with the force of law (hoheitlich). Private-law disputes between the government and its citizens—for instance, over government contracts or the executive’s activities as a market participant—are adjudicated by the ordinary civil courts.110Gerichtsverfassungsgesetz [GVG] [Courts Constitution Act], May 9, 1975, Bundesgesetzblatt, Teil I [BGBl I], as amended, § 71,. The courts’ respective jurisdictions depend on the characterization of the initial administrative act. The intricate set of doctrines that govern this field are beyond the scope of this summary.

C.     Verwaltungsrecht in a Nutshell

German administrative law rests on two elementary principles that sharply contrast with American law. First, administrative adjudication—“judicial review of agency action,” in our parlance—almost always involves the legality of an individual administrative act (“adjudication” or enforcement), as distinct from administrative rulemaking.111See limited exceptions briefly discussed infra notes 151–153. Second, judicial review focuses single-mindedly on the plaintiff’s substantive rights, as distinct from questions of procedure.

Basics. The Archimedean point of the German system is the Verwaltungsakt (administrative act), meaning a final agency decision that tells an individual citizen, with binding force, what is or is not lawful.112Verwaltungsverfahrengesetz [VwVfG] [Administrative Procedure Act], May 25, 1976, Bundesgesetzblatt, Teil I [BGBl I] as amended, § 35 (“An administrative act shall be any order, decision or other sovereign measure taken by an authority to regulate an individual case in the sphere of public law and intended to have a direct, external legal effect. A general order shall be an administrative act directed at a group of people defined or definable on the basis of general characteristics or relating to the public law aspect of a matter or its use by the public at large.”). Note that administrative rules and regulations are not administrative acts in this sense, on the theory that they still require an individualized agency decision—an administrative act, in other words—to gain force and effect against individual citizens. However, the universe of administrative acts that are subject to judicial review is broader than the technical definition in VwVfG section 35. Foremost, “simple” administrative acts (einfaches Hoheitshandeln or schlicht-hoheitliches Handeln) that are not addressed to anyone in particular are administrative acts and may be challenged under forms of action that parallel those that are available for formal administrative acts. An administrative act must be communicated to the addressee, and it must be accompanied by reasons stating the legal basis and (when applicable) discretionary grounds for the act.113VwVfG § 39(1). The act is effective until it is revoked or changed; it does not require further judicial sanction.114Id. § 43. The executive need not go to court to enforce its orders. In fact, it cannot do so in administrative matters because it has no rights, only powers. However, a timely objection or lawsuit will have a suspending effect,115VwGO § 80(1). and every single administrative act must be challengeable in an independent court by any individual citizen who can credibly claim that his or her rights were violated by that act.116Grundgesetz [GG] [Basic Law], art. 19, para. 4; VwGO § 42(2) (“in seinen Rechten verletzt”). With extremely limited exceptions (such as presidential pardons), the system leaves no room for administrative acts beyond judicial jurisdiction (gerichtsfreie Hoheitsakte). It likewise blocks attempts to render acts “committed to agency discretion by law” unreviewable.117Cf. 5 U.S.C. § 701(a)(2) (declaring that actions committed to agency discretion by law are not subject to judicial review under the APA). Judges decide whether executive officials have exercised their discretion within legal bounds and free from error, not whether courts have jurisdiction to review that discretion.118VwVfG § 40; VwGO § 114. Article 19, paragraph 4 of the GG has been read to limit the legislature’s power to shield the exercise of administrative discretion against judicial control. Bundesverfassungsgericht [BVerfG] [Federal Constitutional Court] Apr. 17, 1991, 84 Entscheidungen des Bundesverfassungsgerichts [BVerfGE] 34, 58. For discussion, see Schoch, supra note 35, at 926–54.

Administrative Procedure. German administrative law does provide for means of contesting adverse administrative decisions within the executive branch.119VwVfG § 28. The procedures are principally governed by the VwVfG. The statute governs federal agencies as well as state agencies insofar as they administer federal law (under Germany’s cooperative federalism, the usual case). Id. § 1. State laws largely replicate the VwVfG, often verbatim. The principal means of contesting an adverse decision is an “objection” (Widerspruch), usually addressed to the next-highest level of the bureaucracy.120VwGO § 69. For a concise account of the process, see Friedrich Schoch, Das Widerspruchsverfahren nach §§ 68ff. VwGO, 2003 Jura 752. The process resembles the APA’s adjudicatory provisions in some respects. The applicant has a right to be heard (in writing or in person) and to be represented.121VwGO § 71. Third parties whose rights could be affected may participate as of right, and the agency may invite the participation of additional parties in its discretion.122Id. § 65. The denial of an objection must again be communicated in writing and supported by reasons.123Id. § 73. As a rule, aggrieved citizens must exhaust this procedure prior to filing suit.124The requirement applies to administrative acts in the sense of section 35 of the VwVfG (but not otherwise). Cf. id. § 68(1)–(2). The “as a rule” qualification matters because administrative procedures are generally governed by state law, and some states (such as Lower Saxony) have done away with the objection procedure altogether—with the result that there is nothing to “exhaust..

For all that, one must not mistake the objection procedure for a scaled-down version of the APA’s adjudicatory provisions. The Widerspruch is directed to an executive body—not an administrative law “judge.” The right to invoke the procedure is circumscribed by what we would call judicial review provisions: it is limited to persons who could independently bring suit and to acts that are, or would be, subject to judicial review.125See id. §§ 70–72. And by near-universal consensus, the administrative procedure serves a “subservient function” (dienende Funktion). It may help individual citizens to obtain relief without litigation. It may help the executive to correct mistakes, enhance public confidence in the bureaucracy’s decisions, and reduce the courts’ caseload. In all events, however, the procedure remains instrumental to the protection of individual rights and to the executive’s and the courts’ unvarying obligation to determine the single substantively correct decision in each case. Tellingly, the VwGO characterizes the objection procedure as a “preliminary procedure” (Vorverfahren)—“preliminary,” that is, to the judicial determination, which is the actual “administrative process” (Verwaltungsprozess).126Part of the reason for this characterization is that the federal government has concurrent jurisdiction over the administrative courts, Grundgesetz [GG] [Basic Law], art. 74, para. 1, but not over the states’ internal administrative procedures. But the nomenclature remains telling, and it is easily lost in translation (e.g., the common translation of the Verwaltungsgerichtsordnung as “Administrative Procedure Code”). What we call “administrative procedure” (or “process”) is the Vorverfahren. Prozess corresponds closely to the APA’s judicial review provisions.

There are no freestanding procedural challenges to administrative acts under German law; such challenges may be brought only by parties with substantive rights claims, in connection with those claims and at the same time.127VwGO § 44a (“Appeals against procedural acts by authorities may only be asserted at the same time as appeals which are admissible against the factual decision.”). A procedural mistake (either in the original administrative act or in the objection procedure) will render a suit admissible—confer “standing to sue”—only if it may have materially affected the plaintiff’s substantive rights.128Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] June 7, 1991, 88 Entscheidungen des Bundesverwaltungsgerichts [BVerwGE] 286, 287; Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Dec. 22, 1980, 61 Entscheidungen des Bundesverwaltungsgerichts [BVerwGE] 256, 258; Michael Sachs, Verfahrensfehler im Verwaltungsverfahren, in 2 Grundlagen, supra note 35, at 799, 836. With some exceptions, procedural mistakes may be disregarded as negligible (unbeachtlich). Moreover, the executive may “cure” procedural mistakes—for example, a failure to supply adequate reasons for an administrative act—at any point of the ensuing litigation, up to oral argument in the case.129Verwaltungsverfahrengesetz [VwVfG] [Administrative Procedure Act], May 25, 1976, Bundesgesetzblatt, Teil I [BGBl I], as amended, §§ 45–46. Because the Federal Administrative Court has given the executive wide berth in these respects, challenges to administrative decisions solely on procedural grounds have become something of a suicide mission.130Sachs, supra note 128, at 844–45. The flipside is that meticulous observance of procedural niceties will never excuse the executive from de novo judicial review on all questions of law and fact.

Litigation. Administrative litigation in a German court is emphatically not a “review” proceeding in our sense of the term. The court is not bound by or limited to the parties’ submissions, let alone the administrative record (sparse as it is). The court determines the facts de novo and ex officio (von Amts wegen).131VwGO § 86(1) (“The court shall investigate the facts ex officio; those concerned shall be consulted in doing so. It shall not be bound to the submissions and to the motions for the taking of evidence of those concerned.”). If that requires independent experts, the court will appoint them; if it requires witnesses, the court will call them; and if it requires site visits, the court will conduct them.132Id. § 87(1). While the administrative courts’ inquiry will seem extravagant to American lawyers, it follows the same inquisitorial model that also governs civil adjudication in ordinary courts. Judicial proceedings are governed by the same rules of civil procedure (Zivilprozessordnung). The court’s legal determinations are likewise and always de novo; the idea of judicial “deference” to administrative interpretations of the law is unknown.

This demanding, intense form of judicial review applies in principle to every administrative act or omission, and some form of action and effective relief (injunctive, monetary, or declaratory) must be available to any citizens whose rights may have been violated. However, while judicial “review” is much deeper under German law than under ours, it is also much narrower. Access to the administrative courts is afforded every individual who can credibly claim to have been violated “in his rights”—but only such individuals, and (with rare exceptions discussed below) no one else.133See discussion infra Part II.D. It is incumbent upon the plaintiff to make that showing, and the court will independently ascertain whether it has been made. This inquiry—the Klagebefugnis that is the bane of every law student—is far more stringent than our “injury in fact” test, and it differs in kind: it is a legal determination.

The individual addressee of an adverse administrative act—a prohibition, or the denial of a legal entitlement—always has standing.134VwGO § 42. In contrast, difficult questions arise when a third party complains of an administrative act in the addressee’s favor. A prototypical case is a building permit that adversely affects a neighbor’s property. Such third-party lawsuits are quite common, but they are subject to very strict rules of admissibility. The third-party plaintiff must show that he was arguably affected in his rights, not merely his monetary (or aesthetic or ideological) interests.135Id. § 80. Without that showing, the case will be deemed inadmissible and even an obviously unlawful administrative act will remain valid and in force. These doctrines reflect a near-morbid fear of Popularklagen—that is, litigation on behalf of the public, in form or in substance.136A widely used introductory textbook conveys the flavor: “The principal reason for the requirement of standing remains the prevention of public interest lawsuits [Popularklagen] . . . . Least of all should a citizen turn himself, by means of an administrative lawsuit, into a custodian of the common good and in that way drag the administrative court into a conflict between different interpretations of the public interest.” Friedhelm Hufen, Verwaltungsprozessrecht 237, 238 (10th ed. 2016) (author’s translation). That fear, in turn, reflects the separation-of-powers understanding sketched earlier. Citizens have a sacrosanct right to protect their own “subjective” rights. They have absolutely no business acting as a “private attorney general” who ensure the “objective” legality of government conduct.

Generally, “subjective public rights” are derived not from private-law norms but rather from the Constitution. They include, importantly, the right to property137Grundgesetz [GG] [Basic Law], art. 14. and, yet more importantly, the “general freedom to do as you wish” (allgemeine Handlungsfreiheit)138Id. art. 2, para. 1.—not a textually specified right, mind you, but a judge-made doctrine that underpins the courts’ understanding of administrative law and, in particular, of “protective” statutory law. Statutory provisions create private rights when they “intend” or “aim at” the individual’s protection, in whole or at least in part.139The Federal Administrative Court has adhered to this formulation throughout. See, e.g., Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] June 7, 1993, 92 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 313; Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Mar. 16, 1989, 81 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 329; Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Nov. 3, 1988, 7 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 355; Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Feb. 25, 1954, 1 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 83. This “theory of [rights-]protective norms” (Schutznormtheorie) bears a certain resemblance to the U.S. Supreme Court’s “prudential” or statutory standing analysis, but it is much more stringent than the “not especially demanding” inquiry under U.S. law.140Cf., e.g., Lexmark Int’l, Inc. v. Static Control Components, Inc., 572 U.S. 118, 130 (2014); Match-E-Be-Nash-She-Wish Band of Pottawatomi Indians v. Patchak, 567 U.S 209, 225 (2012) (citing Clarke v. Sec. Indus. Ass’n, 479 U.S. 388, 399–400 (1987). Under German law, in contrast, the standing analysis often threatens to collapse into the merits, and cynics might suspect that the point of the distinction is to enable law school professors to flunk students regardless of whether they park their analysis under standing (Zulässigkeit) or merits (Begründetheit). However, the crystal-clear purpose of conducting a rigorous standing inquiry—even at the risk of front-loading it with merits questions—is to prevent public interest lawsuits. See, e.g., Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Oct. 29, 1963, 17 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 87. See also Hufen, supra note 136.. The theory presents no small degree of difficulty, especially with respect to environmental laws and other statutes that serve both collective, public-regarding and rights-protective functions. Sometimes, the courts classify such statutes on a wholesale basis. For example, animal protection statutes protect cats and dogs, but not their owners’ affections or monetary interests.141E.g., Verwaltungsgerichtshof [VGH] [Higher Administrative Court] Apr. 3, 1997, Neue Juristische Wochenschrift [NJW] 3429 (animal protection statute). Statutes for the protection of public forests protect the trees, not private perambulators.142Bundeswaldgesetz [BwaldG] [Federal Forest Act], May 2, 1975, Bundesgesetzblatt, Teil I [BGBl I], as amended, at 1037. Much more often, the courts parse the statutes provision by provision. Among the most common criteria to distinguish protective from non-protective norms is the question of whether the group(s) of individuals who supposedly enjoy protection can be limited and circumscribed, conceptually and as a practical matter. In other settings (for example, competitor lawsuits), the intensity or degree of interference with the third-party plaintiff’s expectations may determine the Klagebefugnis.143For example, mere monetary interests (such as lost profits) do not confer standing to challenge an administrative act (e.g., a business license) in a competitor’s favor. See Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Apr. 18, 1985, 71 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 183, 193. By contrast, a threat to the plaintiff’s “existential livelihood” may satisfy standing requirements in a competitor lawsuit. See, e.g., Bundesverwaltungsgericht [BVerwG] [Federal Administrative Court] Aug. 30, 1968, 30 Entscheidungen des Bundesverfassungsgerichts [BVerwGE] 191, 197. The distinctions are not always entirely clear or convincing. But the decisions leave no doubt about the judiciary’s resolve in maintaining the boundaries.

D.     Developments and Debate

Among the standard themes of American administrative law is the notion that a court-centered, rights-focused, rule-bound system of administrative law is incompatible with the demands of a modern, complex society, which requires far more expertise and flexibility than such a system could possibly permit. The German system belies that contention.

To be sure, over the past half-century, the system has shown strains. Acute controversies have arisen over large-scale public projects that affect the rights of potentially thousands of citizens—nuclear power plants at first, then airports, train stations, and similarly massive undertakings.144Greve, supra note 83, at 223. Misgivings over the judicial management and a perceived lack of effective legal protection in such “mass proceedings” were soon joined by more general complaints about an “enforcement deficit” especially in environmental law.145Id. at 203–05; Andreas Vosskuhle, Neue Verwaltungsrechtswissenschaft, in 1 Grundlagen, supra note 35, at 1, 8–12. Public agencies, critics argued, often failed to enforce ambitious statutes with sufficient rigor, and potential plaintiffs who might challenge such derelictions in court routinely foundered on the shoals of standing requirements that systematically favor regulated parties over broad public interests. At the same time, legal scholars noted a distressing legislative propensity to legislate in broad, general terms or by way of statutes that—unlike the conditional, “if-then” statutes paradigmatic in the traditional model—instruct public agencies to optimize multiple, often incompatible objectives.146Michael Stolleis, Entwicklungsstufen der Verwaltungsrechtswissenschaft, in 1 Grundlagen, supra note 35, at 65, 116–17. Predictably, such statutes have helped to produce a flood of administrative rules and regulations and strained the administrative judiciary’s capacity to exercise legal control.147Statutes conferring broad discretion and goal-oriented decision-making tax a judicial system that is designed to subject administrative action and discretion to legal controls—and never to a judicial second-guessing of the good sense or utility of administrative decisions. Reimer, supra note 92, at 652–58; Schoch, supra note 35, at 954–59. Finally, European integration has caused perplexities. While the EU’s growing body of administrative law has often been shaped in analogy to—or directly borrowed from—the German system, certain aspects of that system and foremost its relentless private rights orientation stand in considerable tension with EU law and its interpretation by the European courts.148Because the European Commission lacks sufficient resources to prosecute each and every treaty violation or failure to “transpose” and implement EU directives, every Member State is obligated in certain fields (environment, consumer protection, competition law, internal market law in general) to empower private individuals and interest groups to monitor compliance by giving them standing in national courts—meaning, in Germany, foremost the Verwaltungsgerichte. Prominently, the 1998 Aarhus Convention committed the signatory countries to “guarantee the rights of access to information, public participation in decision-making, and access to justice in environmental matters.” Convention on Access to Information, Public Participation in Decision-Making and Access to Justice in Environmental Matters (“Aarhus Convention”) art. 1, June 25, 1998, 2161 U.N.T.S. 447; see also Directive 2011/92, of the European Parliament and of the Council of 13 December 2011 on the Assessment of the Effects of Certain Public and Private Projects on the Environment O.J. (L 26) 1, 2 (codifying the Aarhus Convention mandates). Eight years later, the Bundestag enacted statutes to transpose this commitment into German law. The most important of these statutes, the Umwelt-Rechtsbehelfsgesetz, derogates from core principles of general administrative law by effectively granting participatory, procedural rights as well as standing to sue to environmental associations. Umwelt-Rechtsbehelfsgesetz [UmwRG] [Environmental Legal Remedies Act], Dec. 15, 2006, Bundesgesetzblatt, Teil I [BGBl I]; see Klaus Ferdinand Gärditz, Klagerechte der Umweltöffentlichkeit im Umweltrechtsbehelfsgesetz, 2010 EurUP § 210; Jan Ziekow, Das Umwelt-Rechtsbelfsgesetz im System des deutschen Rechtsschutzes, 2007 Neue Zeitschrift für Verwaltungsrecht 259. For the limits of this innovation, see infra note 156 and accompanying text.

Much of this will sound familiar to American scholars. Over there as here, environmental concerns (and more recently privacy law, or Informationsrecht) have powerfully influenced administrative law. There as here, a principal concern has been to mobilize the courts to remedy administrative inaction—an “enforcement deficit,” or industry “capture.”149Greve, supra note 83, at 204. In some respects and to some extent, German legislators and courts have responded by nudging the law in an “American” direction. As noted,150See supra notes 139–141 and accompanying text. the legislature has, in limited areas, granted organizations a right to participate in administrative proceedings and to sue, in derogation of the ordinary rules that govern agency procedures and standing to sue. Participation and other procedural requirements have expanded in domains where judicial controls have been deemed inadequate. And in some areas, the federal legislature as well as the courts have permitted proceedings that strongly resemble our forms of pre-enforcement review. The traditional administrative law model permitted substantially no such challenges, on the theory that administrative rules were not administrative acts at all but rather forms of infra-parliamentary legislation and therefore—short of constitutional review—neither susceptible to nor in need of immediate, independent judicial review (inasmuch as any rule or regulation would still require an administrative act to gain legal force vis-à-vis individual citizens). In response to the expansion of administration by rulemaking, the VwGO was amended to permit pre-enforcement rulemaking challenges in certain areas of federal law, such as land use planning.151Under VwGO section 47, pre-enforcement challenges to state rules and regulations may be authorized by state law. Verwaltungsgerichtsordnung [VwGO] [Code of Administrative Court Procedure], as amended, § 47, All states permit such challenges in one form or another. See, e.g., Schleswig Holstein in: Landesjustizgesetz § 67. A few organic statutes, most having to do with environmental matters, contain comparable provisions. Such actions are adjudicated (as most of them are in the United States) by appellate rather than trial courts, and there as here, the court’s determination will generally preclude any subsequent challenge in the course of enforcement proceedings.152In addition, the administrative judiciary has increasingly permitted peremptory rulemaking challenges under a somewhat ill-fitting provision of the VwGO that confers jurisdiction in certain (individual) actions for declaratory relief. See VwGO § 43(1).

Those pressures and temptations have generated a scholarship calling for a “new science of administrative law” (Neue Verwaltungsrechtswissenschaft). From the common-sense observation that many forms of contemporary administration and administrative law do not conform to the traditional model, adherents of the “new science” proceed to the general proposition that administrative law is not simply a means of protecting individuals but also, and perhaps first and foremost, a governmental instrument for the optimization of multiple public-regarding ends.153The most rigorous and impressive effort to develop this “new science” is a monumental, three-volume treatise (each volume clicking in at close to 1,500 pages) entitled—with no sign of intended irony—“Foundations of Administrative Law.” 1–3 Grundlagen, supra note 35. The protection of individual rights is among those ends—but only alongside equally important objectives, such as the democratic legitimation of administrative decisions, their public acceptance, and the efficacy and accuracy of public administration.154See, e.g., Andreas Vosskuhle, Neue Verwaltungsrechtswissenschaft, in 1 Grundlagen, supra note 35, at 1, 63. For critical perspectives, see Oliver Lepsius, Steuerungsdiskussion, Systemtheorie und Parlamentarismuskritik (1999); and—way harsh—Klaus Ferdinand Gärditz, Die “Neue Verwaltungsrechtswissenschaft”: Alter Wein in neuen Schläuchen?, in Die Verwaltung Beiheft 12, Zur Lage der Verwaltungsrechtswissenschaft 105 (Martin Burgi, ed., 2017).

While much of this sounds sensible, even committed advocates of the “new science” have conceded that German administrative law has shown few signs of yielding to their vision.155Gärditz, supra note 154, at 124 (noting that the contributors to the monumental Grundlagen volumes are actually quite skeptical about the project); see also id. at 126 & n.171 (collecting cases and literature references to the effect that the “new science of administrative law” has “left no traces on [legal] practice”). The legislature, executive agencies, and the courts have responded to real-world challenges principally by way of compartmentalization and improvisation—and more than occasionally, by retrenchment. For example, the Bundestag has made only sparing use of its authority to expand the intervention and litigation rights of environmental or other public interest organizations.156Uniformly, those reforms have come slowly, grudgingly, and in response to demands of international or EU law. In particular, the implementation of the Aarhus Convention, supra note 148, has been accompanied by a protracted battle between German institutions (both the legislature and the courts) and the European Court of Justice (“CJEU”), which has repeatedly declared Germany’s reforms inadequate and insisted on full compliance with the Convention. See Case C-115/09, Bund für Umwelt und Naturschutz Deutschland, Landesverband Nordrhein-Westfalen eV v. Bezirksregierung Arnsberg, ECLI:EU:C:2011:289, ¶¶ 1–2 (May 12, 2011); Case C-72/12, Gemeinde Altrip and Others v Land Rheinland-Pfalz, ECLI:EU:C:2013:422, ¶¶ 1–3 (Nov. 7, 2013). For a brief account of the ongoing saga, see Jerzy Jendrośka & Lorenzo Squintani, The Courts as Guardians of the Environment: New Developments in Access to Justice and Environmental Litigation, in International Comparative Legal Guides: Environment & Climate Change Law 6, 8 (Simon Tilling ed., 17th ed. 2020). The piecemeal proceduralization of administrative law has been accompanied by countervailing efforts, both at the federal and the state level, to streamline administrative procedures. The executive’s authorization to “cure” procedural mistakes even during the pendency of litigation, for example, was added to the VwGO in 1996.157See Gesetz zur Beschleunigung von Genehmigungsverfahren [GenBeschlG] [Approval Procedure Acceleration Act], Dec. 9, 1996, Bundesgesetzblatt, Teil I [BGBl I] at art. 6; Gesetz zur Änderung der VwGO und anderer Gesetze [Law Amending the Code of Administrative Court Procedure and Other Laws], Jan. 11, 1996, Bundesgesetzblatt, Teil I [BGBl I]; see also Hufen, supra note 136, at 35–36 (describing efforts, especially since the 1990s, to streamline administrative procedures and litigation; collecting statutory cites). In a similar vein, states have curbed the rudimentary administrative objection procedures, and, in what looks like a bit of overkill, a few states have abolished it almost entirely.158Hufen, supra note 136, at 68–73 (providing state-by-state overview and criticizing the trend as “faddish” and “rash”). The administrative courts for their part have dealt with novel legal problems and instruments principally by assimilating judicial doctrines to the traditional private rights model. Doctrines that govern discretionary agency decisions, for instance, have little in common with our forms of free-wheeling “arbitrary and capricious” review. The focus of the judicial inquiry is whether the alleged arbitrariness concerned or affected the plaintiff’s rights. For reasons explained earlier, that will usually be a question of whether the plaintiff has any business in court in the first place. Similarly, the courts have limited pre-enforcement review—by near-uniform consensus, the only form of action and relief that contemplates “objective legality” rather than private rights as the focal point of judicial examination—in accordance with Germany’s conventional standing norms, so far possible. The traditional private rights model has not broken by any measure. It has barely even bent to the point of allowing its critics an occasional field goal.

* * *

The German system illustrates that one can in fact have a regime of full-scale, independent judicial adjudication. Is it any better than our system of administrative justice? In large measure, the answer depends on one’s normative priors.

If one places a high premium on administrative flexibility, expertise, and democratic participation in the administrative process, German administrative law is an example of what not to do. German administration is far more rigid and legalized than ours. The adjudicators are independent judges. They are generalists, not experts. And the system is not for soi disant democratizers. It invites and in fact demands public “participation”—but not in the bureaucracy or in the form of interest group politics. Rather, members of the public participate either in the ordinary electoral process or else (or rather also) through the courts, where lay judges serve for the publicly advertised purpose of helping professional judges maintain their “distance” from the bureaucracy.159See, e.g., Hamburg Admin. Courts, Leitfaden für Ehrenamtliche Richterinnen und Richter 3–4 (2018), (explaining the point); see id. at 14–15 (explaining that separation of powers concerns bar civil servants and public officials from serving as honorary judges).

The picture takes on a very different coloration if one places a premium on citizens’ right to do as they wish until and unless the government tells them otherwise—the allgemeine Handlungsfreiheit that is rooted in the German Constitution and which suffuses the country’s administrative law.

Consider, by way of illustration, how a Sackett-style case160See supra notes 1–2 and accompanying text. would shake out under German law. Up to a point, the answer depends on how one varies the facts. If the compliance order has binding force (a point not remotely arguable under German administrative law—it’s a prohibitory order addressed to an individual party, explicitly backed by a threat of sanctions) it is contestable in court immediately upon completion of the perfunctory objection procedure described above, and the court’s determination will be de novo. If there is no binding administrative act—say, if the agency sends an informal communication with menacing noises rather than a direct order—the landowner is entitled to submit a request for an appropriate permit to the agency. A timely denial of the permit must be accompanied by reasons; it is contestable in substantially the same manner as a prohibitory administrative act. If the executive fails to act on the request within a statutorily specified period (usually three months), the permit is automatically deemed granted.161See Verwaltungsverfahrengesetz [VwVfG] [Administrative Procedure Act], May 25, 1976, Bundesgesetzblatt, Teil I [BGBl I] § 42a (“Fictitious approval”). The relevant portion of the statute reads: (1) Upon expiry of a specified decision-making period, an approval that has been applied for shall be deemed granted (fictitious approval) if this is stipulated by law and if the application is sufficiently clearly defined in content . . . . (2) The decision-making period . . . shall be three months unless otherwise stipulated by law. The period starts upon reception of the complete application documents. It may be extended once by a reasonable period of time if this is warranted by the complexity of the matter. Any such extension of the decision-making period shall be justified and communicated in good time. Id. Comparable provisions do exist under U.S. law. See, e.g., 42 U.S.C. § 7475(c). However, they lack any bite. Typically, agencies can evade the deadline by sending a letter to the effect that the permit will be denied on schedule unless the applicant agrees to an extension—in which case the agency might perhaps approve at some point. What gives the German requirement its bite is that the initial denial will be subject to de novo review. And if for some exotic reason none of these forms of action are available, individuals may file a declaratory action to obtain certainty concerning the state of legal affairs.162Verwaltungsgerichtsordnung [VwGO] [Code of Administrative Court Procedure], as amended, § 43, Sackett simply cannot happen there.

There is something to be said, is there not, for a legal system that stabilizes legitimate private expectations, blocks government agencies from clouding titles for years on end, and protects a sphere of private conduct against unwarranted government interference? If so, Germany’s administrative law and judiciary may be a model. Part III sketches what such a model, adapted to our Constitution and legal traditions, might look like.

III.     Administrative Courts, American Style

In its current configuration American Administrative Law seems to present a stark alternative: either we accept the appellate review model, or we provide full-scale adjudication by Article III courts in the first instance. The key point that emerges from the German comparison is the omitted third: we could have independent administrative courts. That option was considered and rejected, in England as well as the United States, over a century ago.163Jin Banggui, A Preliminary Study of Comparative Administrative Jurisprudence, 1994 Asia Pac. L. Rev. (Special Issue) 132, 140. The time may have come to reconsider that fateful decision. Prominent scholars have begun to do so.164William Baude, Adjudication Outside Article III, 133 Harv. L. Rev. 1511, 1513–14 (2020); Harrison, supra note 25, at 176–77; Nelson, supra note 65, at 614.

Professors Steven G. Calabresi and Gary Lawson have proposed that “all the current ALJs assigned to agencies whose actions deprive a person of life, liberty, or property be defunded and that Congress should appropriate funds to create new Article III Administrative Law Courts, the judges of which should be nominated by the President and confirmed by the Senate.”165Calabresi & Lawson, supra note 9, at 862. The new Article III judges would be called “Federal Administrative Law Judges,” and they would enjoy tenure during good behavior and Article III salary guarantees.166Id. Vacancies would be filled through the ordinary appointments process of presidential nomination and Senate confirmation.167Id. The proposal “would not apply to the hundreds of statutory ALJs and hearing examiners who decide Social Security or disability cases or who rule on tax and immigration claims.”168Id. Rather, it would initially cover regulatory agencies such as the EPA, the NLRB, the FCC, the FTC, FERC, the SEC, and OSHA.

Professor Michael Rappaport has proposed a similar scheme.169Rappaport, supra note 9, at 825. His proposal aims to combine the rule-of-law advantages of independent administrative courts with the advantages of agency adjudication—expertise, and low-cost decision-making.170Id. Accordingly, Professor Rappaport would divide the newly created independent administrative courts171Like Professors Calabresi and Lawson, Professor Rappaport would establish the administrative courts as Article III courts. Id. at 826. However, he is open to the idea of establishing them as Article I courts instead. Id. at 828–29. into three sections (medical, scientific, and economic), each staffed with judges with professional expertise in those fields and each tasked with reviewing agency decisions that fall into the respective domain.172Id. at 826. The courts’ decisions would be reviewable only by federal appellate courts, not by agencies, and the agencies would receive either no or reduced deference for their findings.173Id. Like Professors Calabresi and Lawson, Professor Rappaport would limit the administrative courts’ jurisdiction to regulatory agencies and their decisions, to the exclusion of benefit or immigration decisions.174Compare Rappaport, supra note 9, at 834–35, with Calabresi & Lawson, supra note 9, at 862.

The institutional scheme presented in this Part aims in the same direction of re-constitutionalizing judicial control over executive adjudication by means of entrusting that task to independent courts.175It differs in some of the details. See discussion infra Part III.A. What this Article hopes to add is a more detailed and nuanced examination of the why, the how, and the for what of such a dramatic departure from a century-old system of administrative adjudication.176In a sympathetic review of the Calabresi–Lawson and Rappaport proposals, Professor Christopher Walker has sensibly urged attention to those questions. See Walker, supra note 9, at 2703. Here you go, Chris. Moreover, this Article will employ a strategy that Professors Calabresi and Lawson forcefully suggest177Calabresi & Lawson, supra note 9, at 862 (“Comparative constitutional law thus suggests that the U.S. administrative law system is primitive and underdeveloped . . . we think this reform [independent administrative courts] borrows something valuable from European administrative law.”).: the comparison to administrative law and courts in many European countries, here in its German instantiation.

Section A sketches the basic argument for independent administrative courts, stipulates, in bullet-point format, the essential, nonnegotiable features of such a judiciary, and explains the range over which those courts should replace the appellate review regime. The following Sections attend to questions of institutional design: the administrative courts’ establishment (Section B) and their proceedings and jurisdiction (Section C). Throughout, this Article will draw on the German comparison and, moreover, will stress the interplay between institutions and legal doctrine. The point of an independent administrative judiciary is to provide an institutional mechanism to counteract, and over time to cure, the doctrinal shortcomings of American administrative law. American administrative law needs something resembling the German conceptual category of “subjective public rights” and a judiciary that will affirm and protect those rights. Neither AJs nor Article III judges are up to that task. The former are meant to compromise rights claims in accordance with their agency’s mission, and the latter operate in Chevron’s rights-neutral, deferential universe. A suitably designed administrative judiciary just might be capable of developing, over time, a more rights-centered jurisprudence.

A.     Essentials

We need independent courts, for well-rehearsed reasons. But why separate independent administrative courts from ordinary (civil) courts? There are good and bad reasons. The principal bad reason is the notion that administrative acts are never simply a question of law and right but always implicate considerations of expediency and sound public policy. Pursue that train of thought: invariably, it arrives at the conclusion that the administrative tribunals must be made part of the administration, subject to independent judicial control only at the outer limits. That, of course, is our system.

The good reason for a separate system of administrative courts is the recognition that the state (the government, in our parlance) is an actor sui generis: unlike private actors, it can unilaterally declare what is to be lawful and what is not. Thus, unlike in civil litigation, the private litigant against the state never confronts an equal. At some level, legal doctrines will have to reflect that reality. For an obvious example, private citizens may do to or with one another as they wish, within the limits of the law; the executive needs an affirmative legal basis for any act.178On the flipside, of course, the government has defenses that are not available to private litigants, foremost including sovereign immunity. It may be best to commit those separate inquiries to separate judicial systems.

Pursue that thought, and it yields something like the German system. Its essential attributes are as follows:

* Administrative courts must be independent from executive control, supervision, and revision.

* The administrative courts hear only administrative cases, but substantially all administrative cases or at least, a very wide range of cases. Any further specialization—say, a separate system of environmental courts—would reduce the courts’ “distance” from the executive and thus replicate the inherent dangers of “expert” decision-making: tunnel vision, situational decisions, and excessive emphasis on short-term results over rules.

* The administrative courts must be open to every citizen who credibly claims that an administrative act has violated his private rights. However, the courts must remain closed to anyone else, especially including “private attorneys general.”179Under the German system, even actual attorneys general cannot sue in the administrative courts. That is because the government has no rights, only powers. See supra notes 134–136 and accompanying text.

* While executive branch agencies may use hearings and other adjudicatory devices, no agency decision merits judicial deference. Every judicial determination must be de novo, on all questions of law or fact.

* There must be enough administrative courts to adjudicate a large volume of cases.

As shown in the remainder of this Part, it would be perfectly possible to mimic this institutional design well within the bounds of the U.S. Constitution and our political and legal traditions, and so to replace the appellate review model with independent administrative courts, over a wide range.

What range, you ask? Good question. Some of the answers are sufficiently straightforward to be set forth in bullet-point format. Harder questions follow.

* Administrative courts should handle administrative cases (roughly, cases now falling under the APA). Tort and contract claims against the government fall outside that ambit, in the United States as in Germany; that should remain so.

* Tax disputes are committed to a separate judicial track, in Germany as here; there is no reason to upset that arrangement.180Unless, perhaps, one holds very strong views about and against “tax exceptionalism.” Cf., e.g., Kristin E. Hickman, Administering the Tax System We Have, 63 Duke L.J. 1717 (2014). That debate is beyond my purview here.

* Internal administrative matters—for a prominent example, government employment disputes—are best handled inside the government, subject to constitutional constraints (for example, due process181See, e.g., Cleveland Bd. of Educ. v. Loudermill, 470 U.S. 532 (1985). or First Amendment limitations182Garcetti v. Ceballos, 547 U.S. 410 (2006).) and whatever arrangements and access to Article III courts Congress may provide;183For the current arrangement, see 5 U.S.C. § 1204 (defining the powers and functions of the Merit Systems Protection Board), and id. §§ 7111–7120, 7701–7703 (defining the rights and duties of government employees with claims against the government, including a provision granting judicial review). that, too, should remain as it is.

* Criminal cases are and should be beyond administrative courts’ jurisdiction (there as here).

Other questions concerning the administrative courts’ jurisdiction are more involved. The answers depend upon the purposes one intends the administrative judiciary to serve.

This Article declared its purpose in the Introduction: the point is to re-limit the public rights regime and our rights-neutral APA, to rehabilitate meaningful distinctions between coercive interferences with private conduct and mere benefit programs, and to create an institutional apparatus with the capacity and the incentives to perform those tasks. Accordingly, the administrative courts’ jurisdiction should extend exclusively184The “exclusively” point matters. It is implied in and presupposed by, but to my mind not sufficiently emphasized in the Calabresi–Lawson and Rappaport proposals. For example, Calabresi and Lawson insist that all potential deprivations of right must be subject to independent adjudication, see supra note 9, at 861–62. I agree. They slight the corollary proposition that other administrative decisions need not be and must not be subject to judicial “review” at all, lest the judiciary be dragooned into some administrative scheme. This central tenet of German administrative law entails that pre-enforcement challenges and citizen suits are structurally incompatible with a rights-protective jurisprudence. For further discussion, see supra notes 112–113 and accompanying text, and infra notes 218–221 and accompanying text. to regulatory agencies and their programs, as opposed to benefit programs. It should extend only to adjudicatory agency decisions, to the exclusion of rulemaking proceedings and pre-enforcement challenges. And jurisdiction should be withheld for cases arising under “citizen suit” or comparably broad statutory review provisions. Proceedings of that nature can and should stay where they are, in Article III courts and Chevron’s domain.

B.     Establishing a Federal Administrative Judiciary

The simplest way of establishing independent administrative courts might be to proceed in conformity with Article III and to staff the courts with judges who are nominated by the President, confirmed by the Senate, and serve “during good behavior.” Professors Rappaport, Calabresi, and Lawson favor that approach, in slightly different variations.185See Calabresi & Lawson, supra note 9, at 862; Rappaport, supra note 9, at 826–27. This Article takes a different tack. It would be far preferable to establish a federal administrative judiciary pursuant to Article I.186The formulation is a bit misleading, inasmuch as inferior Article III courts, too, are established pursuant to Article I; but I will follow the conventional terminology. This proposed model for a Federal Administrative Court (“FAC”) closely resembles the U.S. Tax Court. Technically, the U.S. Tax Court is an executive body, as are the U.S. Court of Federal Claims and the U.S. Court of Veterans Appeals.187See About the Court, U.S. Ct. of Fed. Claims, (describing the court as an Article I court ); History, U.S. Tax Ct., (same); U.S. Ct. of Appeals for Veterans Claims, (same). (Not being Article III courts, they cannot be anything else.) However, the judges differ from administrative adjudicators in virtually all relevant respects. They are appointed for fixed terms by the President (with the advice and consent of the Senate), not by an agency. They cannot be removed or directed by any agency, and their decisions are not reversible by any agency (only by an appellate federal court). The FAC could and should be constituted in substantially the same form. It would sit atop a hierarchy of lower administrative courts, sufficient in number to replace agency adjudication and AJs over the range described below.188Lower administrative judges could be appointed by the Justices, within legislatively determined parameters and procedures. Again, the U.S. Tax Court provides a model. See 26 U.S.C. §§ 7441–7479. The precise number is a matter of conjecture and largely dependent on the scope of the administrative courts’ jurisdiction. One hundred administrative courts, with at least one administrative court in every state, may be a reasonable guess.

Admittedly, the Article I option introduces complications that would not arise if the administrative courts were constituted in conformity with Article III. The constitutional status of the U.S. Tax Court is notoriously uncertain,189See, e.g., Brant J. Hellwig, The Constitutional Nature of the United States Tax Court, 35 Va. Tax Rev. 269 (2016). and one can argue that the judges, as well as administrative judges, must—as executive officers—be removable by the President.190That, in any event, is Professor Baude’s position. Baude, supra note 164, at 1567–68. It is also the general direction—although perhaps not the inescapable conclusion—of the Supreme Court’s separation-of-powers jurisprudence. See Lucia v. SEC, 138 S. Ct. 2044 (2018); Free Enter. Fund v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477 (2010). Additionally, it may be the case that constitutional claims, and perhaps certain private disputes arising under federal regulatory schemes, must (under the Supreme Court’s jurisprudence) remain in Article III courts.191The answer depends on how one understands the Supreme Court’s meandering “private rights” jurisprudence and, moreover, on one’s views as to whether and under what circumstances private individuals may by consent waive their right to an independent adjudication of constitutional claims. For discussion of that latter question, see Baude, supra note 164, at 1555–57, and sources cited supra notes 18–21. For purposes at hand, I do not and need not express any firm view on those matters.

Still, several reasons militate in favor of the Article I option and against the Article III option. For one thing, any attempt to add some 200-plus federal district court judges would produce heated partisan conflict and insurmountable opposition from the existing Article III judiciary.192The Article III option would have been the cleanest solution to the Supreme Court’s decision in Northern Pipeline Construction Co. v. Marathon Pipe Line Co., 458 U.S. 50 (1982), which held that bankruptcy-related claims arising under (state) common law cannot be adjudicated by non-Article III tribunals. Id. at 52, 76, 87. The Judicial Conference and individual federal judges resolutely opposed proposals to appoint bankruptcy judges pursuant to Article III. Richard A. Posner, The Federal Courts: Challenge and Reform 130–39 (1999). Prior to the publication of the Depravity article, Professor Calabresi submitted the proposal to increase the number of Article III judges to the U.S. House of Representatives, as part of a broader package of legal reform proposals. The condemnation was instantaneous and bipartisan. Walker, supra note 9, at 2688. More important, Article III courts would invariably drift (back) towards the appellate review mode of adjudication: on-the-record review and deference to the experts. Reputational and professional incentives cut that way, and the prospect of appellate review will do the rest. Thus, if one wants to re-constitutionalize the domain where the executive threatens to interfere with legitimate private expectations, the administrative judiciary should be independent not only from the executive but also, so far as possible, from the Article III courts.

What, though, of private litigants’ right to gain access to an Article III court, somewhere in the course of the proceedings? There are two ways of skinning this cat: an adjunct model and a competitive model.

An “adjunct” model would permit appeals from the FAC to a U.S. Court of Appeals. The “competitive,” far preferable option is to make the FAC’s rulings final and preclusive—and to give private litigants a choice between resort to the administrative courts or else, to the APA’s adjudicatory procedures and judicial review provisions. Resort to the agency process would operate as a waiver of right to invoke the administrative courts’ jurisdiction, and vice versa.193Roughly analogous arrangements exist under patent and trademark law, where a disappointed applicant’s decision to seek de novo review in a U.S. District Court operates as a waiver of the right to appellate review and vice versa. See, e.g., Lanham Act, 15 U.S.C. §§ 1051, 1071; Patent Act, 35 U.S.C. §§ 141, 145–146. The tax code provides taxpayers with a choice between contesting an assessment in the U.S. Tax Court or else payment of the assessment under protest and a subsequent challenge in federal court. See 26 U.S.C. § 7422. This model has several key advantages.

First, it would ease transition problems that would invariably accompany the introduction of an arrangement that would require substantial amendments to the APA and to dozens of organic statutes with a bewildering array of agency adjudication systems. Under the competitive model, Congress could leave the existing machinery in place; it would merely have to create an add-on alternative, operating at private plaintiffs’ choice. Second, while the prospect of Article III review may seem reassuring, federal appellate courts would almost surely resist the administrative courts’ attempt to develop more rights-protective doctrines, under de novo modes of proceeding that are alien to American administrative law.194Their natural impulse will be to assimilate those proceedings to the existing corpus juris, and that enterprise would likely be accompanied by a vocal chorus of “anti-exceptionalist” scholars who insist on the normative force and transsubstantive reach of the APA and the appellate review model. A small selection of such “anti-exceptionalist” literature includes Stuart Minor Benjamin & Arti K. Rai, Who’s Afraid of the APA? What the Patent System Can Learn from Administrative Law, 95 Geo. L.J. 269 (2007) (patents); Hickman, supra note 180 (tax); Peter Lee, The Supreme Assimilation of Patent Law, 114 Mich. L. Rev. 1413 (2016) (patents); Melissa F. Wasserman, The Changing Guard of Patent Law: Chevron Deference for the PTO, 54 Wm. & Mary L. Rev. 1959 (2013) (patents); Melissa F. Wasserman, What Administrative Law Can Teach the Trademark System, 93 Wash. U. L. Rev. 1511 (2016) (trademarks). Third, and perhaps most important, a system of mutual claim preclusion—operating at the private claimant’s choice—would protect the administrative judiciary’s capacity to develop rights-protective administrative law in a common-law fashion and, moreover, give the courts an incentive to do so. You, dear private litigant (the system says), may have your biased agency proceeding and deferential—and therefore also biased—review in an Article III court. Or you can come to us. That proffer may well be credible because the administrative judiciary’s reputation and prestige would come to depend on its willingness to provide impartial adjudication and effective remedies. Over time, agency adjudication might atrophy. Or perhaps, some administrative agencies might step up their game and protect their own AJs from irrelevance and unemployment by some means—perhaps, a credible commitment to ensure their impartiality. That’s the thing with (institutional) competition: you never know in advance what exactly will happen. What you can confidently predict is that the dynamic will cut in the desired direction—the erosion of the appellate review model.

C.     Proceedings and Jurisdiction

Cases brought in the administrative courts would not be appellate actions for “review.” Rather, they would be original civil actions—a version of the bill in equity that was once a standard vehicle for judicial “review,” and which has survived to this day in a few pockets of public law.195See, e.g., Lanham Act § 1071; Patent Act §§ 145–146. The standard of review would be de novo. Courts would give zero deference to the administrator, on questions of law or of fact.

The proceedings in the administrative courts would be conducted in accordance with the Federal Rules of Evidence and the Federal Rules of Civil Procedure. They would not be limited to the agency record, so the parties could introduce extrinsic evidence. The cases would be litigated without a jury; judges would make all determinations of fact as well as law.196Obviously, Germany’s inquisitorial model (briefly described supra notes 131–132 and accompanying text) is incompatible with our legal traditions and institutional arrangements. However, even our party-driven model readily permits courts to obtain or elicit the requisite facts.

Unlike in an appellate review proceeding, the usual form of relief would not be a remand to the agency that would require further adjudicatory proceedings; it would be an affirmative, non-reviewable, final order to the agency to give the plaintiff the relief to which he is entitled (excepting claims for monetary relief). Successful claimants would be entitled to reasonable attorneys’ fees.197For this purpose, the administrative courts could be authorized and instructed to borrow the rules of 42 U.S.C. § 1988.

Precisely who should be able to invoke the administrative courts’ jurisdiction, in a challenge to exactly what? I have explained the German answer: the administrative courts are open to anyone who can credibly aver to have been violated “in his rights” by an adverse administrative act—paradigmatically, a coercive interference with the individual’s private sphere of action. At first impression, that precept translates quite readily into the U.S. context. Enforcement orders, compliance orders, permit or license denials, the mistaken inclusion of an individual on a governmental “no-fly” list198Ibrahim v. DHS, 62 F. Supp. 3d 909 (N.D. Cal. 2014); Alex Kozinski, Commentary, How a Pointless Lawsuit Took Off, Wall St. J. (Jan. 24, 2019, 7:26 PM),—anything that looks like an administrative act (in the German sense) or a final agency order (in ours)1995 U.S.C. § 551(6). in an individual case would be grounds to invoke the administrative courts’ jurisdiction. Upon inspection, however, complications arise from the Germans’ corollary proposition: while the courts must be open to individuals whose rights have been violated, they must remain closed to anyone else. Crucially, that includes individuals who may have an “interest” but no legally protected right, and “concerned citizens” who sue on behalf of the public. Those limitations do not translate quite so readily but they are essential. They require some rough distinction between rights and mere privileges or benefits (in our terminology) and, moreover, a bar against pre-enforcement rulemaking challenges and citizen suits. The remainder of this Section considers these matters in turn.

Of Rights and Privileges. At one level, the conceptual distinction between private rights—roughly, a constitutionally protected sphere of private liberty and property—and mere privileges—that is, private claims that arise under and are correlative to a regime of public right200See Harrison, supra note 25.—is quite intuitive. The government seizure of private land is plainly a coercive interference with private right, which requires adjudication by an independent court; a land grant (to use the classic nineteenth century example) belongs on the “public” side. But what is one to do with government licenses? Franchises? Patents? Asylum claims? Or, for that matter, with a vast array of government entitlements, from food stamps to Medicaid claims?

Strikingly, the seeming perplexity never caused any great consternation for German administrative lawyers and courts—far less consternation, certainly, than questions of third-party rights, citizen suits, or pre-enforcement review.201See Stolleis, supra note 146, at 111–17. Conceptually, it isn’t all that hard to analogize a license revocation or a denial of disability benefits to a coercive interference. Either way, the object of the challenge is an individual administrative act, and the dispute is between the government and the individual addressee of its action. Thus, as early as the 1950s, German administrative courts accommodated the Leistungsstaat—the entitlement state—to the inherited conceptual apparatus of administrative law. The accommodation never posed a danger to the administrative judiciary’s central role as the first line of defense against government overreach and imposition. The traditional model of administrative law was too deeply entrenched to permit, let alone invite, a displacement of “coercive interference with private conduct” as the lodestar of the entire system.

The American scenery presents a wholly different picture. The distinction between government-conferred benefits and private (though regulated) spheres of conduct is almost wholly elusive. The APA makes no such distinction, and the public rights “exception” sweeps the boards. Beginning in the 1960s, moreover, the Supreme Court insisted that some sort of constitutional process was due for the “new property” of government benefits and their administrative denials—not the full-scale due process of independent Article III adjudication, mind you, but some sort of fair administrative process202Mathews v. Eldridge, 424 U.S. 319 (1976); Goldberg v. Kelly, 397 U.S. 254 (1970). and an arbiter with a modicum of impartiality.203Schweiker v. McClure, 456 U.S. 188 (1982). Having thus expanded “rights” at one end, the Court contracted due process at the other and decided that the same kind of emaciated due process would also suffice for deprivations that we ordinarily associate with constitutionally protected “liberty” or “property.”204See, e.g., Hamdi v. Rumsfeld, 542 U.S. 507 (2004); Atlas Roofing Co. v. Occupational Safety & Health Rev. Comm’n, 430 U.S. 442 (1977).

Were we to revisit those doctrines, agency adjudication might prove unconstitutional in many of its present forms.205Baude, supra note 164, at 1557–58. However, such a revision is a long-term prospect at best. An administrative judiciary represents a kind of institutional shortcut.

Hard though it may be to draw firm distinctions between right and privilege in any particular case, it is not so hard to distinguish between regulatory and benefit programs, and (more doubtfully perhaps) between the administering agencies. Thus, it should be possible to resurrect the intuitive distinction by conferring jurisdiction over the respective programs, or rather claims arising under those programs, on one set of courts or another. Disability benefits, veterans’ benefits, food stamp entitlements, and federally supported student loans are benefit programs. While their administration may leave much to be desired, transferring their adjudication to administrative courts (as here envisioned) would threaten to transform those courts into something akin to small-claims tribunals and, more importantly, distract them from their central mission to serve as bulwarks against the regulatory state. It would be best, therefore, to limit the administrative judiciary’s jurisdiction to regulatory or enforcement agencies and their programs.206In this respect, the present proposal parallels the Calabresi–Lawson and Rappaport proposals. See Calabresi & Lawson, supra note 9, at 862; Rappaport, supra note 9, at 813. One might start with the EPA, OSHA, the SEC, the FTC, the NLRB, the FCC, the CFPB, the FERC, and the CFTC.

There is no need—certainly not at the outset—to cover all regulatory agencies. What does matter is to cover a meaningful number of agencies with a wide range of portfolios—health and safety regulation, financial regulation, energy, communications, consumer protection. The importance lies in the imperative need to resist the siren song of bureaucratic “expertise” and the concomitant pull toward judicial deference.207I resist Professor Rappaport’s suggestion to staff administrative courts with medical, scientific, and economic experts and to assign cases accordingly. See Rappaport, supra note 9, at 813–14; 826–27. Generalist judges have any number of ways to gather the requisite expertise—for example, from expert witnesses supplied and cross-examined by the parties. The relevant question is not really expertise but specialization. For reasons explained in the remainder of this paragraph, the administrative judges should remain generalists. If prompt, expert decision-making is the point of establishing administrative tribunals, then what can judicial review really contribute to the enterprise? Not a whole lot, is one ready answer.208According to Professor Vermeule’s insightful analysis, this is why Crowell’s protections of the judicial role and prerogatives (i.e., the insistence on de novo review of questions of law and of constitutional and jurisdictional fact) crumbled in very short order. Vermeule, supra note 26, at 16. Another ready answer, however, is that we want the initial decision to be examined from a different perspective by someone who is a generalist and precisely not a specialist, the better to counteract “expert” tendencies toward tunnel vision and agenda-driven decision-making.209See generally Martin Shapiro, Who Guards the Guardians?: Judicial Control of Administration (1988). The appellate review system slights the difference; the point of an independent administrative judiciary is to accentuate it. Functionally specialized courts, seeing the same cases, claimants, and defendant-agencies day in and day out, would soon surrender to cultural capture210Cf. James Kwak, Cultural Capture and the Financial Crisis, in Preventing Regulatory Capture: Special Interest Influence and How to Limit It 71, 94 (Daniel Carpenter & David A. Moss eds., 2014). and tend to develop a kind of specialized administrative “law of the horse.”211Cf. Frank H. Easterbrook, Cyberspace and the Law of the Horse, 1996 U. Chi. Legal F. 208. By way of example: the U.S. Court of Appeals for the Federal Circuit has constructed its very own system of patent-specific rules and procedures. Id. The Supreme Court, for one, has not deemed the experiment successful. See, e.g., Lee, supra note 194, at 1416–18. Let them see an FCC licensing case one day, a land use case the next, and some financial or labor dispute the day thereafter: What general legal rules will work over the entire range? The answers are hardly pellucid but at least, the question is right.

Citizen Suits and Pre-Enforcement Challenges. American administrative law invites a myriad of tangentially affected “stakeholders” and regulatory beneficiaries to invoke the federal courts’ jurisdiction, either by statute212See, e.g., 42 U.S.C. § 7604(a). or under exceedingly generous doctrines of constitutional and statutory standing,213See, e.g., Massachusetts v. EPA, 549 U.S. 497 (2007); Friends of the Earth, Inc. v. Laidlaw Env’t Servs. (TOC), Inc., 528 U.S. 167 (2000). and it routinely permits judicial pre-enforcement review of administrative rulemaking proceedings.214See, e.g., 18 U.S.C. § 1951; Abbott Lab’ys v. Gardner, 387 U.S. 136 (1967). These instruments presuppose a rights-neutral conceptual framework, and they make that apparatus operational. They are structurally incompatible with an administrative judiciary whose mission is to protect this private party from that unlawful executive imposition.215See supra Section II.A.

The incompatibility is obvious with respect to citizen suits, which explicitly put the interests of “concerned citizens” on a par with the holders of traditional rights to property or liberty. It is a bit less obvious, but nonetheless incontrovertible, with respect to pre-enforcement challenges. Pre-enforcement review often entails the statutory preclusion of subsequent challenges, which raises the question of what to do about the complaints of regulated parties who were legitimately unaware of them—or may not even have existed—at the time of the regulation and its pre-enforcement challenge.216In PDR Network, LLC v. Carlton & Harris Chiropractic, Inc., a case arising over preclusion under the Hobbs Act, the Court eventually sidestepped the questions suggested in the text. But those questions did occur to the justices, and they consumed quite a bit of oral argument time. See Transcript of Oral Argument at 1–70, PDR Network, LLC v. Carlton & Harris Chiropractic, Inc., 139 S. Ct. 2051 (2019) (No. 17-1705). But the difficulties are more profound. Pre-enforcement rulemaking review practically must be on-the-record review. By definition, there are no as-applied facts. The point of front loading the process is to provide legal certainty for affected parties and clear sailing for the administration. On that account, and because the appellate court’s determination will usually have preclusive effect, it is hard to resist demands for broad participation in the administrative process and in the judicial proceeding. The proceeding addresses the overall lawfulness and effects of a general rule and is policy laden, and so naturally pushes toward judicial deference. The game is adversarial litigation in name only; it is an interest group sport, and the reviewing court serves as the forum.217The most famous administrative law case of all is tellingly captioned “Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc.”—as if the EPA were a mere bystander.

As explained in Part II, German administrative courts have responded to those pressures by maintaining sharp, rights-based standing requirements and by assimilating pre-enforcement challenges to the paradigmatic forms of administrative action, so far as possible.218See Klaus Ferdinand Gärditz, Die Resubjektivierung der Umweltverbandsklage zwischen prozessualem Vorbehalt des Gesetzes und unionsrechtlicher Rechtsschutzeffektuierung, 2014 EurUP 39, 39–44. The problems have proven manageable because the legislature has made only sparing use of either instrument, many of the most serious inroads into the private rights model have come from the EU rather than German law, and exceptions and compromises can be tolerated so long as it remains clear to all concerned that the traditional doctrines have nothing to do with docket control or pragmatic concerns and everything to do with the logic of the system. Precisely because the courts will protect rights, they must not entertain any other kind of complaint.

None of this is true in the United States. Administrative law makes virtually no distinction between a citizen’s claim to use his land free from official interference and another citizen’s claim to demand that interference, and Congress has been promiscuous in legislating broad standing and pre-enforcement provisions.219Even adamant critics of the administrative state and its law often ignore or overlook the differences between a K Street brawl and a citizen’s fight for his livelihood or possessions. They would not otherwise presume that the same (more aggressive) standard of judicial review should apply across the board, regardless of who sues over what. Here again, the sensible approach is to engineer an institutional solution: withhold administrative court jurisdiction over cases brought under statutory citizen suit provisions, do likewise with pre-enforcement provisions, and leave both to the Article III courts. Under that arrangement the administrative courts would find it easier to develop, over time, a jurisprudence that reasserts meaningful distinctions between regulated parties and mere “stakeholders.”

IV.     Conclusion: Once More of Rights and Institutions

The appellate model is very deeply entrenched, and a reform proposal of the scale described here quite probably exceeds the capacity of our political institutions. Still, the confluence of the Supreme Court’s “dismantling” of ALJ independence and the renewed judicial and scholarly attention to the defects of administrative adjudication provides a rare opportunity to confront the appellate review regime with the promising alternative of independent administrative courts. And while a proposal to learn from the administrative courts of Germany (of all places) may raise eyebrows, the notion of separate administrative courts is not entirely outlandish. Congress has considered something of the sort at various times in our history,220Notably, in the 1930s, a proposal of this kind surfaced in the early phases of the debates that eventually produced the APA. Shepherd, supra note 36, at 1559. Nothing came of it, probably (in my understanding of the somewhat confused debate) on account of a lack of imagination: the New Dealers loathed anything that looked like a court, while the old guard, led by the American Bar Association and Roscoe Pound, associated anything short of Article III courts with “administrative absolutism.” Id. at 1598–99. Interest group concerns also played a role. The patent bar and the tax bar feared that the Court of Customs and Patent Appeals, an Article I court created only in 1929, might be merged into a more general administrative judiciary and so lose its special concern for “its” more narrowly focused constituencies. Sapna Kumar, The Accidental Agency?, 65 Fla. L. Rev. 229, 259 (2013). and this Article has suggested that the U.S. Tax Court might provide a model. For additional illumination, Congress might look to the states, some of which have established independent administrative courts quite like the ones herein depicted.221Approximately half of the states have adopted a “central panel model,” which differs from the federal appellate review model in two and, in some cases, three key respects. See Allen Hoberg, Administrative Hearings: State Central Panels in the 1990s, 46 J. Nat’l Ass’n Admin. L. Judges 107, 109 (1994). First, central panel adjudicators are neither appointed nor removable by any individual line agency. Instead, the panel heads are typically appointed in conformity with the respective state constitutions’ provisions for the appointment of what, for purposes of the U.S. Constitution, we call “principal” or “inferior” officers. Id. at 117. Second, the central panels’ jurisdiction is not limited to disputes arising from a single agency but spans a wide range of agencies and issues. Id. at 131. Third, a few states bar agency reversals of panel decisions. (Louisiana provides a singular arrangement: it prohibits the administrative reversal of panel decisions, and it permits private litigants—but not the agencies—to appeal adverse panel decisions to the regular courts. La. Stat. Ann. § 49:992(B)(3)(b).) Where panel decisions are nonreversible, the panels begin to resemble ordinary (“Article III”) courts with administrative jurisdiction or—depending on the mode of appointment and the mechanics of appeals to the regular courts—institutional “adjuncts” to those courts. For a description of the states’ central panels models and their development, see Hoberg, supra. See also Allen C. Hoberg, Ten Years Later: The Progress of State Central Panels, 21 J. Nat’l Ass’n Admin. L. Judges 235 (2001).

The baseline proposition is this: any system of public law that is worth having needs a reasonably coherent, robust set of rights that run against the government. Those rights correspond to, and serve to fend off, coercive government interference with ordinary private conduct—the right to hire and fire; to sell your product at a price that willing customers will pay; to earn your marginal product in the labor market; to make such use of your land as you see fit, without harming others; to go about your business without fear of extralegal impositions. The allgemeine Handlungsfreiheit, to borrow a phrase. And to ensure it, you need independent courts.

Those precepts resound deeply in Anglo-Saxon jurisprudence. And yet, they are deeply enshrined in German law and explicitly repudiated under our law. Germany’s system of “subjective public rights”222It truly is a system, not a collection of entitlements. Cf. Ottmar Bühler, Die Subjektiven öffentlichen Rechte und ihr Schutz in der Deutschen Verfassungsrechtsprechung (1914); Georg Jellinek, Das System der öffentlichen subjektiven Rechte (2d ed. 1905). These century-plus old works have remained well-nigh canonical. See the exhaustive and illuminating treatment by Hartmut Bauer, Geschichtliche Grundlagen der Lehre vom subjektiven öffentlichen Recht (1986). was conceived a century-plus ago in a Kantian and positivist frame of mind. After the horrifying Nazi experience, jurists more fully recognized the logic of a rights-protective regime, and after 1949, that logic became constitutionalized and institutionalized. In sharp contrast, the adoption of the appellate review model in the United States—hardly inevitable, but a fait accompli by the 1930s—in due course produced the rights-neutral, process-centered regime of the APA.223When Senator McCarran famously described the APA as a “bill of rights,” he did not mean substantive rights; he meant the APA’s procedural protections. Verbatim: “[The APA is] a bill of rights for the hundreds of thousands of Americans whose affairs are controlled or regulated in one way or another by agencies of the Federal Government. [The APA] is designed to provide guarantees of due process in administrative procedure.” 92 Cong. Rec. 2149 (1946) (statement of Sen. McCarran) (emphasis added). Those path dependencies matter. Germany’s rights-focused system is a product of (sometimes very bitter) experience, not a conceptual apparatus that one could parachute into the permanent construction site of American administrative law.224A single example shall suffice. In the 1970s, when a handful of German lawyers proposed to introduce U.S.-style citizen suits (Popularklagen) into German environmental law, Professor Felix Weyreuther—then-President of Germany’s Federal Administrative Court—wrote an entire book condemning the practice and noted that it had been favorably contemplated exactly once: by the Nazi regime and its so-called jurists. Felix Weyreuther, Verwaltungskontrolle durch verbände? 82–84 (1975). Justice Weyreuther had a point. See Greve, supra note 83, at 239 n.215 (citing and discussing Nazi jurists’ advocacy of public interest lawsuits). He expressed a widely shared, long-held view. See, e.g., Oberverwaltungsgericht für das Land Nordrhein-Westfalen [OVG] [Higher Administrative Court for North Rhine-Westphalia] 1953, Monatsschrift des Deutschen Rechts [MDR] 572 (association lawsuit dismissed for lack of standing and criticized as an impermissible delegation of public power that smacks of fascism). Where we sheer on selfless defenders of owls and wetlands, the German legal establishment smells a Blockwart. You cannot make that up, and you cannot transport the underlying sensibility into American law. Hopefully this Article has shown, however, that one can conceive of an administrative judiciary that fits our own legal traditions and that might, over time, provide robust protection for “subjective public rights,” although quite probably not under that heading.

However illusory that prospect may appear, the prospect of leaving administrative law reform to ordinary Article III courts seems equally doubtful.225For a more cheerful assessment, see Merrill, supra note 9, at 915–916 (urging judicial reconsideration of the public rights “exception”). For many contemporary critics of the administrative state, judicial deference—Crowell, Chevron, and associated canons—is the great white whale. Harpoon it, which is to say have Article III courts decide questions of law de novo, and behold the rule of law.226See Hamburger, supra note 64, at 309–21. That will not work, because it cannot work. As Professor Adrian Vermeule has powerfully argued, Crowell’s appellate review framework systematically pushes toward judicial deference.227Vermeule, supra note 26, at 13.

An obvious response to this dilemma is to rehabilitate the notion of private right and to rethink the public right “exception.” Here and there one can find judicial opinions that point in that direction, and a handful of intrepid scholars have seriously engaged the enterprise.228See Nelson, supra note 65, at 613; see also Baude, supra note 164, at 1579 (if one takes “a more radical view of these doctrines” [i.e., the right-privilege distinction and unconstitutional conditions], one would have to confront “the extensive twentieth-century practices and precedents of agency adjudication. When such cases should be overruled is a question about the scope of stare decisis and related doctrines”); Harrison, supra note 25. That intellectual endeavor, like this Article’s, zeroes in on the question of right, and it is closely tied to institutional concerns (albeit of a slightly different order).229My own proposal, as noted throughout, focuses on the question of institutional (judicial) capacity. In contrast, the Federal Courts scholarship focuses on the constitutional question of which institutions may or must adjudicate what sorts of claims. My proposal may appear at odds with the position that private rights must always be adjudicated by Article III courts; but that is not necessarily so. For example, if I understand Professor Baude correctly, he would deem either the “adjunct model” or the “competitive model” described in Part III.B. constitutional, at least in some configuration. See Baude, supra note 164, at 1575. However, the Supreme Court is exceedingly unlikely to revise its public-rights jumble of doctrines except at the outer margins. Hence, this Article’s proposal for an institutional shortcut: commit the adjudication of certain “public” rights, within the scope of the jurisdiction described earlier, to administrative tribunals—just not agency tribunals but actual courts.

This Article does not hold out administrative courts as a solution to all that ails the administrative state and its law. It has said virtually nothing about the adjudication of benefit programs—the vast bulk of administrative adjudication, and perhaps much in need of pragmatic reform. Nor has it said anything about, say, the nondelegation doctrine, or “unorthodox” agency maneuvers to evade judicial review;230Cf. Abbe R. Gluck, Anne Joseph O’Connell & Rosa Po, Unorthodox Lawmaking, Unorthodox Rulemaking, 115 Colum. L. Rev. 1789 (2015); Michael S. Greve & Ashley C. Parrish, Administrative Law Without Congress, 22 Geo. Mason L. Rev. 501, 522 (2015). and besides, what about Chevron deference? The short answer is that sometimes, intellectual and institutional compartmentalization helps. Administrative courts might make some seemingly intractable administrative law questions more manageable and susceptible of more sensible answers.

Start with the Chevron question: perhaps, the problem is not so much the doctrine itself but its “domain.”231For a powerful argument along these lines, see Hickman & Nielson, supra note 73. Judicial deference in pre-enforcement challenges is one thing. The “facts” that matter in those proceedings are not the kind of facts on which courts have any comparative advantage. Law matters, but often one marvels at how little law there is to be had in those cases. (Chevron itself is a fine example.) The questions that courts are good at handling—who exactly did what to whom? How does that rule apply to these facts?—are of little or no relevance because no one has done anything to anyone just yet.

It is a very different thing to bring that same mindset to bear on adjudicatory cases: where the question is precisely who did what to whom; where the facts are readily ascertainable by courts; where the dispute is not among contending interest groups but between a private citizen and the government; and where, typically, a lot more law will be available. It makes a great deal of sense to have those types of cases decided by judicial bodies that operate outside Chevron’s shadow.232Barring the actual adoption of that institutional solution, the mere proposal may help to re-orient the debate. The logic explained in the text also applies to the so-called “Brand X” problem—that is, the question of whether an agency may depart from an interpretation of an ambiguous statutory provision that has been deemed “reasonable” by an Article III court to another reasonable interpretation of that same provision. The general answer is that it may. See Nat’l Cable & Telecomms. Ass’n v. Brand X Internet Servs., 545 U.S. 967, 1000 (2005). However, it is one thing to execute that maneuver by means of notice-and-comment rulemaking. It is a very different and far more problematic thing to effect the change by means of formal adjudication and to spring the new interpretation on an unsuspecting private party who reasonably relied on the old one. Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1148 (10th Cir. 2016); De Niz Robles v. Lynch, 803 F.3d 1165 (10th Cir. 2015). See also Richard W. Murphy, Judicial Deference, Agency Commitment, and Force of Law, 66 Ohio St. L.J. 1013, 1040–41 (2005).

The same reasoning applies to other controverted questions of American administrative law. Under the Chenery II233SEC v. Chenery Corp. (Chenery II), 332 U.S. 194 (1947). doctrine, for example, the choice between rulemaking or adjudication is left to the agencies’ well-nigh unreviewable discretion, as their “expertise” or convenience might commend. Of late, the doctrine has come under fire, as an open-ended invitation to bureaucratic abuse.234See Nielson, supra note 11. However, Article III courts will be loath to challenge the doctrine, not least because it is hard to think of a plausible theory that would tell us when and why an agency must proceed in one way rather than the other. An administrative court system, in contrast, would render the Chenery doctrine a virtual nullity almost by default. Adjudication would be conducted outside of the agencies and de novo, and the verdicts would be unreviewable by the executive. If agencies wanted to buy themselves deference, they would have to write rules (provided they have the statutory authority), reviewable by Article III courts in the ordinary fashion. No judicial theory is needed: the choice is up to the agencies.

For a final example, it may be a mistake to have the same “standing to sue” rules apply to rulemaking challenges as well as truly adversarial adjudication. The point of pre-enforcement proceedings is to have every interest represented, so the standing inquiry perhaps ought to focus on the parties’ internal governance and legitimation,235In the limited areas where German law permits “association lawsuits,” it often requires the plaintiff-organizations to satisfy such requirements. See, e.g., Umwelt-Rechtsbehelfsgesetz [UmwRG] [Environmental Legal Remedies Act], Dec. 15, 2006, Bundesgesetzblatt, Teil I [BGBl I] § 3. as opposed to their ability to identify an “aggrieved” member with contrived-for-litigation “injuries in fact.” In contrast, the point of adversarial proceedings is precisely to focus on rights, and that requires standing doctrines that keep any other “concerned citizen” or “aggrieved” party out of the proceedings. The U.S. Tax Court will not tolerate interventions by parties who claim that someone else has failed to pay taxes due, so why should administrative courts entertain comparable claims?

* * *

The APA’s “fierce compromise” was forged against the backdrop of the Great Depression and World War II, in the thrall of heroic assumptions about the capacity of expert administrative government and for an administrative machinery that would set railroad rates and dispense radio licenses.236Shepherd, supra note 36. None of that has much to do with the modern administrative state and its discontents. There may be reasons to treat an antediluvian political compromise like an ersatz Constitution. Since I cannot readily think of such a reason, this Article proposes a more modern institutional arrangement that more closely approximates the actual Constitution—even if it is borrowed from someplace else.

Share this article

Share on twitter
Share on facebook
Share on linkedin

Share it

Share on twitter
Share on facebook
Share on linkedin