Now More Than Ever: The Case for Specialized Nominees to the U.S. Court of Appeals for Veterans Claims, Ten Years On

CCK Law: Our Vital Role in Veterans Law

Bradley W. Hennings[1]

Introduction

A decade ago, David Boelzner, Jennifer Rickman White, and I published Now is the Time: Experts vs. the Uninitiated as Future Nominees to the U.S. Court of Appeals for Veterans Claims in this Journal.[2] The thesis was direct: two-thirds of judges appointed to the Court of Appeals for Veterans Claims (“CAVC” or “Court”) should be drawn from among lawyers experienced in the U.S. Department of Veterans Affairs (“VA”) benefits claims adjudication system.[3] The CAVC is a specialty court. Veterans deserve a bench steeped in the law that determines their benefits.[4]

The thesis has gained traction. Its central recommendation has not.

Since 2016, the appointment of Chief Judge Michael Allen—a veterans law professor who directed Stetson’s Veterans Law Institute and testified before both chambers’ Veterans’ Affairs Committees—and Judge Amanda Meredith—who spent seven years at the CAVC as a law clerk, executive attorney, and backlog-reduction director before serving as Senate Veterans’ Affairs Committee counsel—reflect a growing recognition that specialized knowledge serves this court well.[5] These were good appointments. They moved in the direction the 2016 article advocated.[6]

But not far enough. No former Veterans Law Judge has been appointed to the CAVC.[7] No private attorney from the veterans benefits bar has been appointed.[8] And the conditions that made the argument urgent in 2016 have intensified. The Court’s jurisdiction now includes class actions. Loper Bright has reshaped the interpretive tools available to CAVC judges. The Appeals Modernization Act has transformed the system the Court reviews. The veterans benefits bar has matured. And the Court faces multiple simultaneous vacancies.[9]

This Article updates the thesis. The need for specialized nominees has not diminished; by several measures, it has grown. An administration that has already demonstrated its commitment to veterans through record claims processing, a bipartisan VA Secretary confirmation, and an accelerated pace of VA nominations is well positioned to extend that commitment to the Court that reviews VA’s work.

I. What Has Changed Since 2016: The System

A. The PACT Act and the Claims Surge

The Sergeant First Class Heath Robinson Honoring Our Promise to Address Comprehensive Toxics Act of 2022 (“PACT Act”) was the largest expansion of VA benefits in a generation.[10] It established new presumptions of service connection for toxic exposure conditions, added twenty-three conditions to the presumptive list, and required VA to provide medical examinations before denying toxic exposure claims.[11] VA estimated that 3.5 million veterans might be newly eligible.[12]

The operational impact was immediate and sustained. VA processed more than three million claims in fiscal year 2025—its fastest pace ever—and the disability compensation and pension claims backlog dropped below 100,000 for the first time since May 2020.[13] Under Secretary Collins’s leadership, VA has maintained this pace while absorbing the largest expansion of benefits eligibility in the system’s modern history—a genuine operational achievement.[14] But backlog is a measure of processing speed, and processing speed is not the same as accuracy.

A significant share of the surge consisted of PACT Act presumptive claims—conditions with established presumptions of service connection that are, by design, faster to adjudicate.[15] When straightforward claims dominate the inventory, the backlog shrinks. That is partly a function of claim mix, not solely a reflection of systemic improvement.[16]

The question for the CAVC is downstream. Every incorrectly denied PACT Act claim becomes a Board appeal or a supplemental claim. Speed at the front end creates volume at the back end.[17] The Court will feel this within two to four years, as denied claims work through the appellate process. And PACT Act claims bring novel medical-legal questions—toxic exposure causation, the boundaries of presumptive eligibility, the interplay between new presumptive conditions and pre-existing ones—that the Court has not previously faced at scale.[18][19]

B. The AMA and Multi-Lane Appellate Architecture

The Veterans Appeals Improvement and Modernization Act of 2017 (“AMA”), implemented in 2019, replaced the single Legacy appeal path with three distinct review lanes: Higher-Level Review, Supplemental Claim, and Board of Veterans’ Appeals appeal.[20] The Board appeal lane was itself subdivided into direct review, evidence submission, and hearing dockets.[21] The AMA was intended to streamline the appeals process and reduce the chronic backlog that had defined the system for decades.[22]

The AMA gave veterans choices they did not previously have and created faster pathways for straightforward errors. It also introduced procedural complexity that the CAVC is only now beginning to confront.

1. The Transitional Overlay

The interaction between the AMA and the Legacy system created a transitional period of extraordinary procedural complexity—two appellate systems running simultaneously, with different rules for opt-in, withdrawal, effective dates, and lane selection.[23] As of this writing, Legacy appeals remain in the system. VA reported just under 36,000 unresolved Legacy cases as of July 2025.[24] The Board’s pending caseload reflects both AMA and Legacy cases with distinct procedural postures, and the CAVC must review Board decisions arising under either system—sometimes in the same case, where a veteran has claims in both tracks.

2. Lane Selection and Its Consequences

The AMA’s lane-selection architecture introduces strategic choices with significant downstream consequences. A veteran who selects the wrong lane—or whose representative selects it for them—may forfeit evidentiary opportunities, waive the right to a hearing, or trigger a different standard of review, all of which become issues on appeal.[25] The duty to assist, for example, operates differently in a Supplemental Claim lane than on the Board’s direct review docket.[26] Effective-date rules depend on whether the claim was filed under the old system or the new one, and whether the veteran opted in during the transition period.[27] They determine the scope of review, the standard of prejudice analysis, and the availability of remand as a remedy.

3. The VLJ Hearing Assignment Problem

One emerging issue illustrates the kind of procedural complexity that demands system-level understanding from the reviewing court. Under the AMA, the VLJ who conducts a Board hearing is not required to be the VLJ who decides the case.[28] This is a departure from the ordinary principle that the factfinder should observe the witness. The CAVC has addressed this issue, and the question is currently on appeal before the Federal Circuit.[29] A judge reviewing this issue must understand not just the legal principle at stake but the operational realities—how Board hearings are scheduled, how cases are distributed to VLJs, and what it means for credibility determinations when the judge who observed the veteran’s demeanor is not the judge who weighs the evidence.

4. Evidentiary Window Ambiguities

The AMA’s evidence-submission docket permits veterans to submit additional evidence to the Board within a defined window. But the statutory language specifies that evidence must be “submitted by appellant and his or her representative.”[30] Does this exclude evidence submitted by VA itself during the evidentiary window? There is currently no court guidance or regulatory scheme that directly answers this question.[31] In a system where VA has a statutory duty to assist claimants in developing evidence, excluding agency-obtained evidence from the evidentiary window creates a tension that the CAVC will have to resolve.

5. Effective Date Preservation

The AMA’s effective-date-preservation mechanism is one of its most significant features—and one of its most technically demanding. Effective dates are “fixed in accordance with the date of receipt of the initial claim or date entitlement arose, whichever is later, if a claimant continuously pursues an issue by timely filing in succession any of the available review options.”[32] Unlike the Legacy system, a claimant who loses at the Board or even at the CAVC can file a Supplemental Claim and retain the original effective date—a fundamental change in the relationship between appellate finality and retroactive benefits.[33] The Federal Circuit in MVA v. McDonough addressed several AMA regulatory provisions, invalidating some as inconsistent with the statute and upholding others.[34] But many AMA-specific effective date questions remain unresolved, and the CAVC is the court that will confront them first.

6. The Implicit Denial Doctrine

In February 2026, the Federal Circuit held in Hamill v. Collins that the implicit denial doctrine—a judge-made rule from the Legacy system permitting VA to “deny” a claim without ever mentioning it in the decision—cannot coexist with the AMA’s explicit notice requirements.[35] The unanimous opinion concluded that the AMA “deliberately and clearly heightened the notice requirement for VA’s initial decisions beyond what was previously acceptable.”[36] The implicit denial doctrine was a creature of institutional practice inside VA. Whether it survived the AMA is a question that requires familiarity with both the doctrine’s operational origins and the AMA’s procedural architecture. The CAVC will confront Hamill’s downstream consequences for years.

The 2016 article observed that CAVC judges without system experience faced a steep learning curve.[37] The AMA made it steeper. A judge who has worked within both systems—who has adjudicated claims under Legacy and understands the practical consequences of lane selection under the AMA—is better positioned to evaluate whether the Board correctly applied the procedural rules that govern each case. A judge learning both systems simultaneously, from briefs, starts behind.

But there is a perspective that even Board-level experience does not fully capture. Private practitioners who represent veterans at the agency level experience the AMA from the ground. Lane selection is not an abstract procedural choice when the decision determines whether an effective date worth tens of thousands of dollars is preserved or lost. A Regional Office that has not scheduled a C&P exam despite a pending duty-to-assist obligation is not a hypothetical — it is a Tuesday. The distance between what the regulations require and what actually happens at the point of contact between the system and the veteran it serves is where the errors originate that eventually become CAVC appeals. A judge who has built the evidentiary records that the Board decides and the Court reviews brings a perspective that neither adjudicatory experience nor appellate practice alone can replicate.

C. Class Actions at the CAVC

In 2017, the Federal Circuit held in Monk v. Shulkin that the CAVC has authority to hear class action cases under the All Writs Act and the Court’s rulemaking power.[38] Monk was a watershed. Before it, the CAVC’s work was entirely individual—one appellant, one Board decision, one judicial review. After Monk, the Court gained the authority to address systemic failures in VA’s claims processing on a class-wide basis.

The Court has since certified several classes and amended its Rules of Practice and Procedure to establish class-action procedures.[39] Godsey v. Wilkie was the Court’s first certified class action, addressing unreasonable delay.[40] Freund v. Collins addresses legacy appeals processing.[41] Pending legislation would further expand the CAVC’s class action jurisdiction by providing supplemental jurisdiction over certain benefits claims.[42]

Class actions at the CAVC are qualitatively different from individual appeals. They require the Court to evaluate systemic VA practices, assess aggregate harm, and design remedies that operate across thousands of cases simultaneously. This is institutional litigation. It demands familiarity with how VA processes claims at scale—how Regional Offices handle large categories of claims, how the Board manages its docket, how policy directives translate into individual case outcomes. The class action docket, more than any other development since the Court’s founding, makes the case for specialized nominees.

D. AI, Unaccredited Assistance, and the Integrity of the Claims Process

Artificial intelligence has entered the veterans benefits claims process—not through VA, but through the private market. It has arrived outside the regulatory structure that governs the system.

Unaccredited claims consultants and AI platforms now offer veterans automated tools for generating what purport to be medical nexus opinions—the evidence that connects a current disability to military service.[43] Some platforms charge as little as twenty-five dollars per month for access to tools that produce medical opinion letters in minutes, without a physician’s independent clinical judgment.[44] Others charge flat fees of over a thousand dollars for AI-generated claims packages that include nexus letters, personal statements, and evidence strategies.[45] These entities operate outside the VA accreditation framework—without the fee restrictions, competency standards, or disciplinary oversight that apply to accredited attorneys, agents, and VSO representatives.[46]

1. The Evidentiary Problem

The veterans benefits system depends on the probative value of medical evidence. A C&P examination reflects a clinician’s independent assessment of an individual patient—history reviewed, examination conducted, opinion rendered on the basis of professional judgment. An AI-generated nexus letter that mimics this structure but is produced by software analyzing self-reported symptoms without clinical examination is a fundamentally different product.

VA has responded. The DBQ fraud-detection program flags suspicious medical evidence.[47] The rescission of 38 C.F.R. § 4.10 signals awareness that the evidentiary environment is changing.[48] These effects have not yet reached the CAVC in volume, but the trajectory is clear.

The Court’s framework for evaluating medical evidence—under Nieves-Rodriguez v. Peake[49] and Stefl v. Nicholson[50]—assesses probative value based on factual accuracy, reasoning, and methodology.[51] Whether an AI-generated opinion that produces a facially adequate rationale by assembling language patterns from training data—without the clinical judgment Nieves-Rodriguez and Stefl assumed—satisfies these requirements is a question of first impression.

2. The Upstream Claim Development Question

Evidence quality is only one dimension. AI platforms and unaccredited consultants also shape what claims are filed, on what theories, and with what evidence—before the claim reaches VA. A veteran who uses an AI platform receives a set of conditions to claim, a theory of service connection for each, and templated supporting evidence. Whether the platform verifies that claimed conditions are actually present, that nexus theories are medically supportable, or that evidence strategies reflect the specific rating criteria VA will apply varies by platform and is rarely transparent to the veteran.

By contrast, an accredited representative develops a claim through individualized assessment: interviewing the veteran, reviewing service records, identifying conditions plausibly connected to service, and developing evidence targeted to the specific rating criteria VA will apply. This is professional judgment applied to individual circumstances.

The CAVC will see appeals reflecting both approaches. A judge who knows what competent claim development looks like — who has built records and adjudicated them — can distinguish the two in ways that matter when the question is whether the Board’s reasoning adequately addressed the evidence before it.

3. The Accountability Question

These developments raise a structural question: what happens when the claims process is shaped by entities outside the regulatory structure designed to protect veterans? The entities generating AI evidence are not accredited. They file no Power of Attorney. They are not subject to VA’s competency standards, conduct rules, or disciplinary authority. If their evidence or strategy leads to an incorrect denial—or a grant later reversed—the veteran bears the consequence alone.

Some publish outcome statistics as marketing. Whether those statistics account for baseline grant rates, PACT Act presumptions, or concurrent attorney assistance is rarely transparent.

The policy question of how to regulate this sector is for Congress and the executive branch. The adjudicatory consequences will reach the CAVC regardless.

E. The Evolving Accreditation Environment

The 2016 article assumed a relatively stable regulatory environment in which veterans seeking help would work with accredited representatives.[52] That assumption no longer holds. A substantial and growing segment of the claims assistance market is now occupied by entities operating without VA accreditation, charging fees ranging from $25.00 to over $20,000 for services that accredited representatives are either prohibited from charging for or are subject to fee review.[53]

To see why this matters for the CAVC, one must see what accreditation provides. The VA accreditation framework is not a licensing formality. It is a consumer protection architecture built around accountability to the veteran.[54] Accredited representatives are subject to competency standards,[55] conduct rules prohibiting misleading veterans or delaying claims,[56] regulated fees—with a categorical prohibition on charging for initial claims assistance[57] and caps on appeal fees subject to VA review[58]—a Power of Attorney requirement establishing a formal relationship VA can monitor,[59] and disciplinary authority that can result in suspension or cancellation of accreditation.[60]

Entities that operate without accreditation are not subject to these requirements. The obligations apply only to those who submit to the framework. The result: veterans who receive assistance from unaccredited entities have no recourse through VA’s disciplinary process, and the entity has no continuing obligation VA can enforce.[61]

Some entities characterize their services as “coaching” rather than representation, arguing they fall outside VA’s accreditation requirements.[62] Whether this distinction is legally sustainable is being actively litigated. The functions performed—advising veterans on which disabilities to claim, what evidence to gather, how to describe symptoms—are the same functions accredited representatives perform.

The fee structure creates a notable asymmetry. Under 38 U.S.C. § 5904(c)(1), accredited attorneys and agents may not charge any fee for initial claims assistance.[63] Unaccredited entities face no such restriction. Whether this asymmetry serves or undermines congressional intent is a question pending legislation is attempting to resolve.

The litigation and enforcement record reflects the growing recognition of this problem. Ford v. Veterans Guardian resulted in class certification under state consumer protection statutes.[64] The Texas Attorney General settled with a major claims coaching entity.[65] California enacted a ban on unaccredited claims consultants effective February 2026.[66] VA’s Office of General Counsel has issued more than forty cease-and-desist letters.[67] Pending federal legislation—the CHOICE for Veterans Act and the GUARD VA Benefits Act—would expand VA’s enforcement authority and, in some formulations, create a pathway to accreditation for entities willing to submit to oversight.[68]

The emergence of new models of claims assistance does not diminish the relevance of system experience. It increases it.

II. The CAVC-Federal Circuit Relationship After Loper Bright

The 2016 article focused on the CAVC in isolation. But the CAVC exists within a vertical appellate structure in which the Federal Circuit—an Article III court with no specialization in veterans law—reviews the CAVC’s legal conclusions de novo.[69] That relationship has intensified since 2016.

A. After Loper Bright

In Loper Bright Enterprises v. Raimondo, the Supreme Court overturned the Chevron doctrine, returning interpretive authority to the judiciary where it belongs.[70] For the veterans benefits system, this is an opportunity: courts can now exercise independent judgment about what veterans law statutes and regulations mean, rather than defaulting to VA’s preferred reading. But exercising that authority responsibly in a system this complex requires judges who know the regulatory architecture they are interpreting. The case for specialized nominees is strengthened, not weakened, by the end of reflexive deference.

Before Loper Bright, the CAVC had cited Chevron in 170 precedential decisions.[71] Between 2020 and the Loper Bright decision, the CAVC cited Chevron in 30 precedential decisions, though in the majority of those cases the Court resolved the question at Step 1 (finding the statute unambiguous) without reaching deference.[72] The practical effect of Chevron in veterans law was always modulated by the competing pro-veteran canon—the Gardner presumption that interpretive doubt is to be resolved in the veteran’s favor.[73]

The Gardner canon and the deference doctrines have coexisted uneasily for decades, and courts at both levels have struggled to reconcile them.[74] The Federal Circuit’s approach has been cautious—employing Gardner as one tool among many rather than as a dispositive tiebreaker.[75] A recent Harvard Law Review study found that over a ten-year period from 2013 to 2023, the Federal Circuit cited Gardner in the veterans law context thirty times but applied it in a veteran’s favor only twice.[76] A parallel study covering 2011 to 2016 found zero instances.[77] The canon appears most often in dissents or in passing.[78]

With Chevron gone, the opportunity is for the CAVC to develop a more coherent framework for applying the Gardner canon in statutory and regulatory interpretation. That work requires judges who can articulate why a particular provision is or is not ambiguous, who can engage with the Gardner canon as a substantive interpretive tool rather than a rhetorical flourish, and who understand the practical consequences of interpretive choices for the veterans whose claims turn on the regulations at issue. A judge who has watched this dynamic play out across hundreds of cases—from the practitioner’s side, from the adjudicator’s chair, or both—navigates it differently from one who encounters it for the first time on the bench.

B. Auer, Kisor, and Regulatory Interpretation

The deference question extends to VA’s interpretations of its own regulations. In Kisor v. Wilkie—itself a veterans benefits case—the Supreme Court narrowed Auer deference without overruling it.[79] The Federal Circuit’s handling on remand illustrated the tension: the court concluded the regulation was not genuinely ambiguous, reversing its prior holding that it was.[80] The result generated significant commentary.[81]

For the CAVC, regulatory interpretation is not an occasional exercise. It is the Court’s daily work. The VA rating schedule, the procedural regulations governing the AMA, the duty-to-assist provisions, the rules governing effective dates—these are regulatory instruments that CAVC judges must interpret in every case. A judge who has applied these regulations as a VLJ, who has briefed their meaning to the Court as a practitioner, and who has watched the Federal Circuit reinterpret them on further appeal, brings an interpretive fluency that cannot be acquired from reading a treatise.

C. The Remand Cycle

The CAVC-Federal Circuit relationship also plays out through the remand cycle. When the CAVC vacates a Board decision and remands, the case returns to the Board, then potentially back to the CAVC, then potentially to the Federal Circuit. Chief Judge Allen, writing before his appointment, described repeated remands as the “hamster wheel.”[82] The AMA was designed in part to break this cycle, and the evidence suggests improvement.[83] But whether the CAVC has the institutional capacity to make the AMA’s design work as intended is itself an argument for judges who know the system the AMA reformed.

In Byron v. Shinseki, the Federal Circuit held that even where the Board commits clear legal error, the CAVC ordinarily should remand rather than reverse unless the record permits only one possible outcome.[84] For practitioners, this means a winning appeal may still send the case back into the system. A judge who knows what a remand means in operational terms—how long the Board takes to act, how delay affects aging veterans—approaches the remand-versus-reversal question with a sense of practical consequence that enriches the legal analysis.

III. What Has Changed Since 2016: The Court Itself

A. The Outcome Variance Problem

The 2016 article cited concern about outcome variance in single-judge decisions.[85] That concern has since been rigorously quantified. In 2016, Ridgway, Stichman, and Riley published an empirical study reviewing more than 4,000 single-judge decisions over a two-year period.[86] Their findings were striking: significant outcome variation across judges, with clear examples of decisions that violated the Court’s own Frankel criteria—which require that single-judge decisions address only cases of “relative simplicity” whose outcomes are not “reasonably debatable.”[87] The study recommended “substantial changes” in the Court’s exercise of single-judge authority,[88] finding that the Frankel criteria[89] were not being faithfully applied.[90]

Outcome variance is troubling in any appellate tribunal. It is especially troubling at the CAVC, where the great majority of decisions are single-judge and non-precedential. The practical result: the outcome of a veteran’s appeal may depend not on the merits of the case but on which judge is assigned to decide it.[91] That is not the hallmark of a court that resolves individual appeals “fairly and efficiently.”[92]

The connection to judicial qualifications is direct. Outcome variance is amplified when judges have widely divergent levels of system knowledge. A judge who knows how VA rating criteria apply to specific conditions evaluates the same record differently from one who does not. Greater specialization would not eliminate variance—it is inherent in appellate adjudication—but it would reduce the portion attributable to unfamiliarity with the system under review.

With four potential appointments, the opportunity to affect the Court’s consistency is significant. If all four seats are filled by nominees without specialized experience, the learning-curve problem compounds. If even two are filled by practitioners steeped in the system, the effect on the Court’s institutional competence is immediate and durable.

B. The Current Vacancies

The Court faces multiple simultaneous vacancies. Two seats are currently vacant: one created by the death of Judge William S. Greenberg on March 16, 2026, while in active service,[93] and one created by legislation expanding the Court’s temporary authorization.[94] Two additional vacancies are approaching: the terms of Judges Coral Wong Pietsch and Margaret Bartley, both appointed in 2012, expire in 2027.[95]

The President could fill up to four seats on a nine-judge court—nearly half the bench.[96] This administration filled six CAVC seats during its first term, appointing Chief Judge Allen, Judge Meredith, and Judges Toth, Falvey, Laurer, and Jaquith.[97] Two of those six—Allen and Meredith—brought specialized veterans law experience. Four did not. The question is whether the next round extends the trajectory that the two began—or reverts to the pattern that produced the other four.

C. The Caseload Trajectory

The Congressional Research Service has noted that the Court “will likely need to maintain, if not expand, the number of judges in active service” to keep pace with its growing caseload.[98] The CBO estimated that pending legislation expanding CAVC jurisdiction would require the equivalent of ten additional attorneys.[99] An understaffed court cannot absorb this volume. A court staffed with judges who require years of on-the-job training cannot process it efficiently.

IV. What Has Changed Since 2016: The Bar

A. Maturation Metrics

In 2016, the article described a veterans benefits bar that was “maturing.”[100] In 2026, the bar is mature. The CAVC Bar Association has evolved into a significant institutional voice.[101] Law school veterans law clinics have proliferated.[102] The Texas State Bar Military and Veterans Law Section is considering submitting a formal proposal to the Texas Board of Legal Specialization to create a board-certified Veterans Disability Law specialty—the first such proposal in the country.[103]

B. The Depth of the Private Practice Bar

The bar now includes former Veterans Law Judges in private practice—three nationally, by practitioner-verified count.[104] It includes an estimated twenty-five to thirty-five former CAVC law clerks in private practice.[105] The range of institutional experience available in the private bar—adjudication, appellate review, quality review, and agency-level representation—is broader than at any point in the Court’s history. The pool from which specialized nominees could be drawn is correspondingly deeper.[106]

V. The Revised Proposal

A. Reaffirming the Two-Thirds Principle

The original proposal: at least two-thirds of future CAVC nominees should come from attorneys experienced in VA benefits adjudication.[107] The updated data reinforces the case. For every three seats filled, one nominee from the appellants’ bar, one from VA, one outside nominee.[108] With four seats open, that means at least two—and ideally three—nominees with deep system experience.

B. The Inside-VA Gap

No sitting or former CAVC judge has served inside VA as a claims adjudicator.[109] No one currently on the bench has decided a disability appeal as a Veterans Law Judge, sat in a Board hearing room and weighed a veteran’s testimony against the medical evidence, applied the rating schedule to a specific condition in a specific veteran’s case, or signed the decision. The Court reviews this work every day. But none of its judges have done it.

This is a gap without parallel among Article I specialty courts. Tax Court judges routinely come from tax practice. Patent judges on the Patent Trial and Appeal Board come from patent practice. VLJs adjudicate the very decisions the CAVC reviews, and their qualifications are analogous to those required of Administrative Law Judges—they are appointed by the Secretary with presidential approval and must demonstrate competence in veterans benefits law before deciding a single case.[110][111] Yet no former VLJ has been appointed to the CAVC.[112]

A judge who has worked inside VA brings something that no other professional background can supply: an understanding of how VA actually makes decisions. Not how decisions are supposed to be made according to policy manuals, but how they are made in practice—under the pressures of a docket that demands volume, in hearing rooms where veterans describe conditions they struggle to articulate, through the application of a rating schedule that requires clinical judgment the adjudicator does not possess. Institutional pressures shape decision quality in ways that are invisible from the appellate record: docket management, productivity metrics, remand compliance deadlines. The reasons-and-bases requirement is not just a legal standard. It is a drafting discipline—and a judge who has practiced it understands what it means when the Board’s analysis falls short. That fluency is not available from doctrine alone.

A distinction is warranted. Some attorneys have held VLJ commissions while serving primarily in policy or administrative capacities—deciding a limited number of cases incident to a role focused on institutional management. That experience is valuable, but it is different in kind from carrying a full adjudicatory docket as a primary assignment. The Court reviews the work of VLJs who decide cases full-time. A nominee who has done that work would bring an understanding that policy-level experience does not replicate.

With four seats available, appointing even one nominee with dedicated inside-VA adjudicatory experience would give the Court a perspective it has never had.

C. The Outside Perspective: Practitioners in the Field

Inside-VA experience is necessary. It is not sufficient. The ideal nominee would combine it with the perspective that comes from representing veterans outside VA—from the other side of the table.

The CAVC reviews Board decisions. But the record the Board decides—and the Court reviews—is built at the agency level, by practitioners working with veterans in communities across the country. Not in appellate offices. In Regional Offices and VA medical centers far from Washington, where an attorney sits across the table from a veteran trying to explain how a condition that began during a deployment twenty years ago has worsened to the point where he cannot work. The record is built through medical evidence development, strategic lane selection, hearing requests, and effective date preservation—decisions where the stakes are measured in years of retroactive benefits.

A judge who has done this work understands something that inside-VA experience alone cannot supply: what the system looks like from where veterans live. How long it takes to get an examination scheduled in a rural area. How a duty-to-assist failure at the Regional Office cascades into a Board remand that adds years to a case. How the distance between a regulation’s text and its local implementation creates the errors that become CAVC appeals.

The combination of adjudicatory experience within VA and practitioner experience representing veterans against VA would bring the Court something it has never had. Neither perspective alone is complete. Together, they close the gap between the record on appeal and the reality that produced it.

The 2016 article proposed that one of every three seats should go to the appellants’ bar. That category should be understood broadly enough to include practitioners whose primary experience is at the agency level—not just those who have briefed cases at the CAVC. The claims system is national. Its problems are local. A judge who has worked in the field, outside the Beltway, brings a perspective that complements both the adjudicator’s view and the appellate litigator’s view. All three belong on this court.

D. Formalizing the Consultation Process

The 2016 article proposed that every administration should consult the CAVC Bar Association and NOVA prior to putting forth any nomination.[113] With four simultaneous appointment opportunities, the rationale for formalization is urgent. NOVA and the CAVC Bar Association should jointly establish a standing judicial nominations committee, prepared to evaluate candidates and provide assessments when vacancies arise.[114] The American Bar Association has performed this function for Article III courts for decades.[115] There is no reason the specialized veterans bar cannot do the same.

But institutional consultation should not be limited to external bar organizations. VA itself—through the Secretary, the General Counsel, and the Board’s leadership—has a direct stake in the quality of the Court that reviews its work. VA understands the Board’s operational needs, the remand cycle’s costs, and the kinds of guidance that translate into better adjudication. If any institution should have input on who reviews its work, it is the institution doing the work.

VI. Addressing the Counterarguments

A. “Veterans Should Serve on This Court”

Military service brings a perspective no credential can replicate, and the tradition of appointing veterans to this bench is worth preserving. The question is not whether veteran status matters—it does. The question is whether, among four available seats, at least one can go to someone whose career has been spent inside the claims system. The two-thirds framework preserves room for veterans in every appointment cycle.

B. “Outside Perspectives Prevent Capture”

The concern is legitimate, and the one-third allocation for outside candidates addresses it.[116] But capture risk must be weighed against competence risk. The greater danger to this Court is not that insiders will be too deferential to one side. It is that outsiders will lack the context to identify the errors the system routinely produces—and will remand cases that did not need remanding because they cannot distinguish error from reasonable adjudicatory judgment. The class action docket makes this concrete: evaluating systemic VA practices requires systemic understanding. A judge cannot design an effective class-wide remedy for a process the judge does not understand.

C. “The Selection Process Will Always Operate Through Political Channels”

The selection process for CAVC nominees, like all presidential judicial appointments, operates through political channels. That is by constitutional design, and this Article does not suggest otherwise. The question is whether those channels can be supplemented with structured input from the institutions that understand the Court’s work.[117] The CAVC resolves individual veterans’ appeals. No major political constituency has a stake in the ideological composition of this bench.[118] That makes it an ideal candidate for a process that brings expertise into the room alongside the political judgment that properly belongs to the President and the Senate.

D. “Loper Bright Simplifies the Interpretive Framework, Reducing the Need for Specialization”

One might argue that with Chevron overturned and interpretive authority restored to the judiciary, the CAVC’s work has simplified—that judges no longer need to navigate the complex interplay between agency deference and the pro-veteran canon. Loper Bright was the right decision. But its consequence for the CAVC is not simplification. It is the opposite. Every regulatory interpretation now becomes an exercise in independent statutory construction rather than a two-step deference inquiry. Independent judicial judgment in a system this complex requires deeper understanding of the regulatory structure being interpreted, not less. A judge who knows why a regulation was drafted the way it was—because she worked within the system that produced it—exercises that judgment with greater fidelity than one parsing the text cold. Loper Bright gave courts the authority to do their job. Specialized nominees are how this court does it well.

E. “Adjudicators Are Not Appellate Judges”

A VLJ is a factfinder. A CAVC judge reviews findings deferentially. One might argue that adjudicatory skill does not predict appellate skill. The argument misunderstands what VLJs do. A Veterans Law Judge applies legal standards, weighs competing medical evidence, assesses credibility, constructs the analytical framework required by the reasons-and-bases obligation, and produces a written decision under appellate scrutiny. These are appellate skills exercised from an adjudicatory posture. The transition is a shift in vantage point, not a change in discipline. And the adjudicator-turned-judge brings something no other appellate judge can: the ability to read a Board decision and know whether the analysis reflects genuine engagement with the evidence or a template that masks inadequate reasoning.

VII. Conclusion

The trajectory is right. Chief Judge Allen and Judge Meredith brought genuine expertise to this bench, and credit is due to the administrations that appointed them. The current administration has set a pace of VA engagement that outstrips its recent predecessors. The question is whether that commitment extends to the Court that reviews VA’s work.

It should. The Court faces categories of complexity that did not exist when the 2016 article was written. Class actions requiring systemic understanding. A multi-lane appellate architecture the Court is only beginning to interpret. A post-Loper Bright environment that demands deeper regulatory fluency. AI-generated evidence challenging assumptions embedded in the Court’s evidentiary framework. Empirically documented outcome variance that greater specialization could reduce. A claims process that was already, as one commenter noted, “too complex for the uninitiated lawyer.”[119]

The gap this Article identifies is specific. No sitting CAVC judge has ever carried a full adjudicatory docket inside VA. The Court reviews that work daily but has never included someone who has done it. Closing that gap does not require displacing military veterans, policy experts, or outside perspectives. The two-thirds framework preserves room for all three. It requires adding what the Court has lacked since its founding: a judge who understands the adjudicatory process not from the record on appeal but from the hearing room where the record was made.

The bar is mature. The pool is there. Four seats. Forty percent of the bench. Fifteen-year terms.

This is the moment. Our nation’s veterans would be well served if it is seized.[120]

Endnotes

[1] Bradley W. Hennings is a Partner at Chisholm Chisholm & Kilpatrick LTD (CCK Law). He is a former law clerk at the U.S. Court of Appeals for Veterans Claims (Judge Alan G. Lance, Sr.) and a former Veterans Law Judge at the Board of Veterans’ Appeals (appointed by the Secretary of Veterans Affairs with presidential approval). He is a Past President of the CAVC Bar Association, Chair-Elect of the Texas State Bar Military and Veterans Law Section, and a member of NOVA’s Congressional Committee. The views expressed in this article are solely those of the author and should not be attributed to CCK Law, the BVA, the CAVC, or any other organization or individual.

[2] Bradley W. Hennings, David E. Boelzner & Jennifer Rickman White, Now is the Time: Experts vs. the Uninitiated as Future Nominees to the U.S. Court of Appeals for Veterans Claims, 25 Fed. Cir. Bar J. No. 2 (2016) [hereinafter Now is the Time].

[3] Id. at 371.

[4] Id.

[5] See Judges, U.S. Court of Appeals for Veterans Claims, https://www.uscourts.cavc.gov/judges.php (last visited Mar. 25, 2026). Chief Judge Allen was nominated in June 2017 after sixteen years as a tenured professor at Stetson University College of Law, where he directed the Veterans Law Institute. Judge Meredith served from 1997 to 2005 at the CAVC as a law clerk to Judge Kramer, executive attorney to Chief Judge Kramer, and director of the Court’s Task Force for Backlog Reduction before joining the Senate Veterans’ Affairs Committee staff.

[6] See Now is the Time, supra note 2, at 394–95 (proposing that nominees should come from the specialized veterans benefits bar). The other judges appointed since 2016—Judge Joseph L. Toth (former federal public defender and Army JAG officer), Judge Joseph L. Falvey, Jr. (former military judge and Army JAG officer), Judge Scott J. Laurer (former Deputy Legal Counsel to the National Security Council and Navy JAG officer), and Judge Grant C. Jaquith (former military officer and government attorney)—brought distinguished military and government service but not veterans benefits practice experience.

[7] See Judges, supra note 5 (listing current and former judges; no former VLJ identified among appointees).

[8] See id.

[9] See infra Parts III.B, IV.

[10] Pub. L. No. 117-168, 136 Stat. 1759 (2022).

[11] Id. §§ 103, 201–203, 301–303.

[12] U.S. Dep’t of Veterans Affairs, PACT Act: What You Need to Know, https://www.va.gov/resources/the-pact-act-and-your-va-benefits/ (last visited Mar. 25, 2026).

[13] U.S. Dep’t of Veterans Affairs, Press Release, VA Disability Compensation and Pension Claims Backlog Below 100,000 (2025).

[14] See Bradley W. Hennings, LinkedIn (2025) (observing that “every incorrectly denied claim that moves quickly through the Regional Office eventually becomes a Board appeal or a supplemental claim”).

[15] See id.

[16] Id.

[17] Id.

[18] See, e.g., 38 U.S.C. § 1116A; 38 C.F.R. § 3.320.

[19] Other novel questions of first impression are already working through the appellate process. See, e.g., Adams v. Collins, Nos. 26-1304, 26-1383 (Fed. Cir.) (pending appeal on obesity as a basis for secondary service connection); De Hart v. Collins, No. 24-2238 (Fed. Cir.) (pending appeal on notice-of-disagreement and radiculopathy rating issues).

[20] Pub. L. No. 115-55, 131 Stat. 1105 (2017) (codified as amended at 38 U.S.C. §§ 5104A–5104C, 5108, 7105).

[21] 38 U.S.C. § 7113; 38 C.F.R. § 20.700.

[22] See H.R. Rep. No. 115-135, at 2–4 (2017).

[23] See 38 C.F.R. § 19.2.

[24] U.S. Dep’t of Veterans Affairs, AMA Congressionally Mandated Report (Aug. 2024), at 11 (reporting remaining Legacy inventory).

[25] See 38 U.S.C. §§ 5104A–5104C (describing lane-specific procedures and evidentiary rules).

[26] Compare 38 U.S.C. § 5103A (duty to assist in supplemental claims) with 38 U.S.C. § 7113(a) (limiting new evidence on direct review docket).

[27] See 38 C.F.R. § 3.2400; compare 38 C.F.R. § 3.400.

[28] See Berry Law Firm, Modernizing Your Approach to the Appeals Modernization Act (Jan. 2025) (noting that “the VLJ who conducts and attends a Board hearing is not required to decide the outcome of the case” and that this issue “is currently on appeal before the Federal Circuit”).

[29] Id.

[30] 38 U.S.C. § 7113(b).

[31] See Berry Law Firm, supra note 27 (noting that “[t]here is currently no court guidance or regulatory scheme that directly answers this question”).

[32] 38 U.S.C. § 5110(a)(2).

[33] Cf. Legacy system, under which the only option after a CAVC affirmance was to reopen the claim, forfeiting the original effective date.

[34] Military-Veterans Advocacy v. Secretary of Veterans Affairs (MVA v. McDonough), 7 F.4th 1110 (Fed. Cir. 2021).

[35] Hamill v. Collins, No. 24-1543 (Fed. Cir. Feb. 4, 2026).

[36] Id. (Moore, C.J.).

[37] Now is the Time, supra note 2, at 392.

[38] Monk v. Shulkin, 855 F.3d 1312, 1318–21 (Fed. Cir. 2017).

[39] U.S. Ct. of Appeals for Veterans Claims, R. Prac. & Proc. 23.

[40] Godsey v. Wilkie, 31 Vet. App. 207 (2019).

[41] Freund v. Collins, No. 20-6475 (Vet. App.).

[42] Veterans Appeals Efficiency Act of 2025, H.R. 3835, 119th Cong. (2025) (bipartisan; introduced in both chambers); Review Every Veterans Claim Act of 2025, H.R. 2137, 119th Cong. § 3 (2025) (expanding CAVC jurisdiction over certain benefits claims). See Cong. Budget Office, Cost Estimate: H.R. 2137, at 3 (2025).

[43] See, e.g., Turbovets Inc. v. VetClaims.ai, No. 2:25-CV-04733-JJT (D. Ariz.) (amended complaint filed Jan. 9, 2026) (alleging fraud, misrepresentation, and unfair competition between AI-based claims platforms).

[44] Id.

[45] See id. (documenting fee structures of AI claims platforms ranging from subscription models to flat fees exceeding $1,250 per veteran).

[46] See 38 U.S.C. § 5904 (accreditation requirements and fee restrictions for attorneys and agents); 38 C.F.R. § 14.629 (VA accreditation regulations). Unaccredited entities are not subject to these provisions because they do not seek accreditation.

[47] U.S. Dep’t of Veterans Affairs, DBQ Fraud Detection Program (2026).

[48] Rescission of 38 C.F.R. § 4.10 (2026).

[49] Nieves-Rodriguez v. Peake, 22 Vet. App. 295, 301 (2008).

[50] Stefl v. Nicholson, 21 Vet. App. 120, 124 (2007).

[51] Stefl, 21 Vet. App. at 124 (requiring that an adequate medical opinion “provide[] supporting rationale”); Nieves-Rodriguez, 22 Vet. App. at 304 (assessing probative value based on “factually accurate, fully articulated, sound reasoning”).

[52] See Now is the Time, supra note 2, at 374–76.

[53] See Ford v. Veterans Guardian VA Claim Consulting LLC, No. 1:23-cv-00756 (M.D.N.C.) (class certified 2025) (documenting fee structures ranging from $1,250 to over $20,000); see also Turbovets Inc. v. VetClaims.ai, No. 2:25-CV-04733-JJT (D. Ariz.) (documenting subscription-based AI platform fees).

[54] See 38 U.S.C. § 5904; 38 C.F.R. §§ 14.629–14.636 (establishing the accreditation framework, fee restrictions, competency standards, and disciplinary authority).

[55] 38 C.F.R. § 14.632(b)(1) (defining “competent representation” as “the knowledge, skill, thoroughness, and preparation necessary for the representation, including understanding of the applicable issues of fact and applicable provisions of title 38, United States Code, and title 38, Code of Federal Regulations”).

[56] 38 C.F.R. § 14.632(a)–(b) (enumerating conduct standards including duties of truthfulness, diligence, and prohibition on misleading veterans).

[57] 38 U.S.C. § 5904(c)(1) (prohibiting accredited attorneys and agents from charging any fee for services before an agency of original jurisdiction has issued a decision on the claim).

[58] 38 C.F.R. § 14.636(e)–(f) (providing that fees of 20% or less of past-due benefits are presumed reasonable; fees exceeding 33.33% are presumed unreasonable).

[59] See 38 C.F.R. § 14.631 (requiring filing of VA Form 21-22a to establish representative relationship); see also 38 C.F.R. § 14.629 (conditioning accreditation on compliance with regulatory framework).

[60] 38 C.F.R. § 14.633 (providing for suspension or cancellation of accreditation for violations of conduct standards, incompetence, or other misconduct).

[61] See Ford, No. 1:23-cv-00756 (alleging that unaccredited entities “hijack” the claims process without accountability); cf. 38 C.F.R. § 14.633 (disciplinary authority limited to accredited representatives).

[62] See, e.g., representations documented in Ford, No. 1:23-cv-00756, and in the Texas Attorney General’s enforcement action, in which entities characterized their services as “coaching” or “education” rather than representation before VA.

[63] 38 U.S.C. § 5904(c)(1).

[64] Ford v. Veterans Guardian VA Claim Consulting LLC, No. 1:23-cv-00756 (M.D.N.C.) (class certified 2025).

[65] Office of the Texas Attorney General, Settlement with VA Claims Insider (2025).

[66] Cal. Bus. & Prof. Code § 6190 (effective Feb. 2026).

[67] See VA Office of General Counsel, cease-and-desist correspondence to unaccredited claims consultants (40+ letters issued as of 2026). See also 38 U.S.C. § 5905 (criminal penalties for unauthorized fees, effectively unenforced since Congress modified the provision in 2006).

[68] CHOICE for Veterans Act; GUARD VA Benefits Act (both pending, 119th Cong.).

[69] 38 U.S.C. § 7292.

[70] Loper Bright Enterprises v. Raimondo, 144 S. Ct. 2244 (2024).

[71] Jonathan M. Gaffney, Cong. Research Serv., TE10108, Judicial Review of VA Regulations: An Overview (Dec. 2024).

[72] Id.

[73] Brown v. Gardner, 513 U.S. 115, 118 (1994).

[74] See Procopio v. Wilkie, 913 F.3d 1371, 1387 (Fed. Cir. 2019) (O’Malley, J., concurring) (noting “the court’s failure—yet again—to address and resolve the tension between the pro-veteran canon and agency deference”); Chadwick J. Harper, Give Veterans the Benefit of the Doubt: Chevron, Auer, and the Veteran’s Canon, 42 Harv. J.L. & Pub. Pol’y 931, 958–59 (2019) (arguing that courts should apply the veteran’s canon before turning to deference doctrines).

[75] Note, Codify Gardner, 138 Harv. L. Rev. 2093 (2025) (finding that over the period 2013–2023, the Federal Circuit cited Gardner thirty times in veterans law cases but applied it in the veteran’s favor only twice).

[76] Id. (replicating the methodology of a prior study covering 2011–2016, which found zero instances of the Federal Circuit using Gardner to rule in a veteran’s favor).

[77] Id.

[78] Id. (“Rather, the canon was most commonly employed in dissent when courts found against veteran claimants, or in passing, but not as determinative.”).

[79] Kisor v. Wilkie, 139 S. Ct. 2400 (2019).

[80] See Kisor v. Shulkin, 880 F.3d 1378, 1379–80 (Fed. Cir. 2018) (O’Malley, J., dissenting from denial of en banc hearing).

[81] See CAVC’s treatment of Kisor on remand, in which the Federal Circuit reached the opposite conclusion on ambiguity from its prior decision.

[82] Michael P. Allen (now Chief Judge, U.S. Court of Appeals for Veterans Claims), Commentary on Three Cases from the Federal Circuit and Court of Appeals for Veterans Claims, 5 Veterans L. Rev. (2013) (using the term “hamster wheel” to describe the remand cycle).

[83] See U.S. Dep’t of Veterans Affairs, AMA Congressionally Mandated Report (Aug. 2024) (reporting that 86% of veterans who disagreed with initial decisions chose the faster VBA review lanes rather than Board appeals, and that Legacy inventory continued to decline); see also James D. Ridgway, Why So Many Remands?: A Comparative Analysis of Appellate Review by the United States Court of Appeals for Veterans Claims, 1 Veterans L. Rev. 113 (2009) (documenting the pre-AMA remand rates that motivated reform).

[84] See Byron v. Shinseki, 670 F.3d 1202 (Fed. Cir. 2012) (limiting CAVC reversal authority); see also Allen, supra note 79 (discussing the tension over the scope of CAVC’s authority to reverse rather than remand).

[85] Now is the Time, supra note 2, at 392.

[86] James D. Ridgway, Barton F. Stichman & Rory E. Riley, Not Reasonably Debatable: The Problems with Single-Judge Decisions by the Court of Appeals for Veterans Claims, 27 Stan. L. & Pol’y Rev. 1 (2016).

[87] Id. at 23–41 (reviewing over 4,000 single-judge opinions and finding significant variation in outcome and repeated violation of the Frankel criteria).

[88] Id. at 1.

[89] Frankel v. Derwinski, 1 Vet. App. 23, 25–26 (1990).

[90] Ridgway, Stichman & Riley, supra note 83, at 23–41.

[91] Now is the Time, supra note 2, at 392.

[92] Id. at 394.

[93] U.S. Ct. of Appeals for Veterans Claims, Notice: Death of the Honorable William S. Greenberg (Mar. 16, 2026).

[94] See U.S. Ct. of Appeals for Veterans Claims, About the Court, https://www.uscourts.cavc.gov/about.php (last visited Mar. 25, 2026); Cong. Research Serv., IF11365, U.S. Court of Appeals for Veterans Claims: A Brief Introduction (2021) (“Seven judgeships are permanently authorized; two additional seats are authorized under a temporary expansion that will expire in 2026.”).

[95] Judges Pietsch and Bartley were both appointed in 2012 for fifteen-year terms. See supra note 5.

[96] Four vacancies out of nine authorized seats (seven permanent, two temporary expansion).

[97] See Judges, supra note 5. Chief Judge Allen and Judge Meredith were appointed in 2017; Judges Toth, Falvey, Laurer, and Jaquith were appointed between 2019 and 2020. Six appointments to a nine-judge court in a single term.

[98] Gaffney, supra note 68.

[99] Cong. Budget Office, Cost Estimate: H.R. 2137, at 3 (2025).

[100] Now is the Time, supra note 2, at 380.

[101] About Us, CAVC Bar Ass’n, https://www.cavcbarassociation.org/ (last visited Mar. 25, 2026).

[102] See, e.g., Harvard Law School Veterans Legal Clinic; Puller Veterans Benefits Clinic, William & Mary Law School; Notre Dame Law School Veterans Legal Clinic.

[103] Texas State Bar Military and Veterans Law Section, Proposal to the Texas Board of Legal Specialization: Veterans Disability Law as a Board-Certified Specialty (2026) (on file with author).

[104] As of this writing, the three identified former VLJs in private veterans disability practice are the author, James Ridgway (Bergmann & Moore), and Jim March (Bergmann & Moore).

[105] Estimate based on practitioner knowledge of the CAVC clerkship alumni community. Published estimates vary, but the number of former clerks actively practicing in the private veterans benefits bar is smaller than is sometimes assumed.

[106] Now is the Time, supra note 2, at 380.

[107] Id. at 394.

[108] Id. at 395.

[109] Id. at 396.

[110] See 38 U.S.C. § 7101A(b)(1); compare 5 U.S.C. § 3105.

[111] Cf. Appeals Regulations: Title for Members of the Board of Veterans’ Appeals, 68 Fed. Reg. 6,621, 6,622–23 (Feb. 10, 2003).

[112] Now is the Time, supra note 2, at 394.

[113] Id. at 399.

[114] This committee could be modeled on the ABA’s Standing Committee on the Federal Judiciary, adapted for the CAVC’s specialized context.

[115] Denis Steven Rutkus, Cong. Research Serv., R43762, The Appointment Process for U.S. Circuit and District Court Nominations: An Overview 8 (2014).

[116] Now is the Time, supra note 2, at 394.

[117] Id. at 399.

[118] Id. at 394.

[119] Michael P. Allen, Veterans’ Benefits Law 2010–2013: Summary, Synthesis and Suggestions, 6 Veterans L. Rev. 1, 7 n.33 (2014).

[120] Now is the Time, supra note 2, at 400.