Improving Models for Agency Appellate Review

An ACUS study examines how agency review models differ across agencies.

Hundreds of agencies across the federal bureaucracy adjudicate millions of cases each year. Nearly all their hearing-level decisions are subject to internal agency appeal under systems established by statute or, more commonly, internal agency regulation.

Despite the widespread use of appellate systems in administrative adjudication, little comparative research has been undertaken to examine their role and workings throughout the federal government. Nearly all the studies of appellate systems have been limited to particular agencies. Ronald Cass’s 1983 study for the Administrative Conference of the United States (ACUS) was a notable exception, although it was limited, as most studies have been, to appeals of hearing-level decisions under the Administrative Procedure Act (APA). Cass’s study built on a 1969 ACUS study by James Freidman.

Several years ago, ACUS returned to the subject when it asked us to undertake a large-scale comparative study of agency appellate systems. This study, unlike its two predecessors, covered systems that review both APA decisions and the much larger number of non-APA hearing-level decisions. Scholars have called this the new world of agency adjudication.

We identified appellate systems at twelve well-known agencies. They included the most commonly used review structures: direct appellate review by the agency head (usually a board or commission) resulting in a final agency decision, as at the Securities and Exchange Commission; direct appellate review by an agency-head delegate resulting in a final agency decision unreviewable by the agency head, as at the Social Security Administration; appellate review by a statutorily created appellate adjudicator (often a multi-member board) resulting in a final decision unreviewable by the agency head, as at the Board of Veterans Appeals; and intermediate appellate review (often by a multi-member body) resulting in a decision subject to final agency-head review, as at the Justice Department’s Executive Office of Immigration Review.

For each agency appellate system, we used legal materials and other publicly available information to prepare a detailed overview of the system’s structure, composition, procedures, and other attributes. We then conducted interviews with at least one high-ranking official at each agency—in most cases, the head of the agency’s appellate program. We followed a carefully designed script, but we also asked follow-up questions based on answers the official gave during the interview and the extensive agency overview we had created before the interview. Although our study design had obvious and acknowledged methodological limitations, it represents an important contribution to the literature on agency adjudication and has served ACUS well.

From the agency-specific overviews, interviews, and an accompanying literature review, we reached numerous findings concerning agency appellate systems today. Our findings address important legal issues such as the statutory or regulatory basis for appellate review; the sources of procedural law; the selection of cases for appellate review, when review is discretionary; and the standards of review. Our findings also address the management, administration, and bureaucratic oversight of appellate systems, as well as the public disclosure of decisions and other materials on agency websites. Based on our findings and proposed recommendations, ACUS adopted Recommendation 2020-3, Agency Appellate Systems, at its December 2020 plenary session. The recommendation offers agencies more than two dozen best practices that, when implemented, will substantially improve their appellate-review systems.

Our findings are too extensive to summarize here, but one bears emphasis: The judicial model of appellate review associated with federal court review of agency decisions is not the predominant model in agency review of hearing-level decisions, although it plays an important role in many programs. We observed other models of review at work—many with bureaucratic features distinct from federal court review.

The main reason agency review models vary is that agency review can serve numerous—and sometimes competing—objectives. We identified at least seven objectives:

Error correction or prevention;
Inter-decisional consistency;
Policymaking;
Political accountability;
Management of hearing-level adjudication;
Operational effectiveness and systemic awareness; and
Reduction of federal court litigation.

Agencies’ institutional and procedural choices are driven by their chosen objectives. An appellate system with the main objective of policymaking—as opposed to, for example, error correction or prevention—would be unlikely to grant review in most cases, especially in a high-volume system with scarce resources. The system would instead intervene selectively in cases with contested policy questions. To focus on policymaking, moreover, the system might provide a deferential standard for reviewing factual findings or forego certain factfinding review altogether.

Surprisingly, we found that few agencies have fully identified their objectives for appellate review, much less tailored their procedures and practices to realize those objectives. We responded to this shortcoming in our report to ACUS by recommending, first and foremost, that agencies explicitly identify the objective, or objectives, of appellate review—not only internally, but also in their publicly disseminated rules of practice—and then structure their processes and procedures accordingly. ACUS, for its part, adopted our recommendation.

Our study is not the final word on agency appellate systems, but we hope it will encourage more scholarly research—and, in particular, more cross-agency and comparative studies—on agency adjudication at the appellate and hearing level. To promote this scholarship, we identified several promising areas for future study.

More research is needed, for example, on quality assurance systems in high-volume adjudication programs. In an important article last year, Daniel Ho and David Marcus suggested that quality assurance programs may “operate as a more systematic replacement for a laborious, underutilized system of appellate review.” Recently, ACUS launched a project to study, as Jerry Mashaw studied for ACUS almost fifty years ago, quality assurance in adjudication.

In sum, when it comes to agency appellate review, much more work needs to be done—empirically, doctrinally, and theoretically—on issues large and small. With how important agency appellate review has become for administrative adjudication, we highly doubt that it will be another four decades before ACUS returns to the subject.