The Congressional Bureaucracy

Publication Date

2020

Volume

168

Document Type

Article

Abstract

Congress has a bureaucracy.

This Article introduces the concept of the “congressional bureaucracy,” and theorizes what it means for Congress to have an internal workforce of more than 4,000 nonpartisan, highly specialized, and long-serving experts, without which the modern Congress could not function. These experts—not elected Members or their political staffs—write the text of the laws, audit implementation, research policy, estimate bills’ economic effects, decide which committees control legislation and which amendments can be made, edit and rearrange already-enacted (!) legislation into the law as we see it in the U.S. Code, and much more. The congressional bureaucracy furthers internal and external separation of powers, revives theories of Congress as a rational actor, and supplies key insight for statutory interpretation. But Courts, lawyers and legal scholars have almost entirely ignored their existence.

This project is based on two years of confidential interviews with high-level staffers in Congress’s nine nonpartisan legislative institutions—the Office of the Law Revision Counsel; the Offices of the Legislative Counsels; the Congressional Research Service; the Government Accountability Office; the Parliamentarians; the Congressional Budget Office; the Joint Committee on Taxation; MedPAC and MACPAC—and additional interviews with partisan staff. The project furthers a new line of legislation scholarship about the value to theory and doctrine of understanding how Congress actually works. Courts cannot claim the doctrines of statutory interpretation are democratically linked to Congress, as virtually all judges do, without understanding how it writes legislation.

Our research reveals that the congressional bureaucracy serves purposes previously unimagined by legal scholarship. Classic bureaucracy literature posits that Congress loses power when it delegates. But the congressional bureaucracy was explicitly founded so that Congress could reclaim and safeguard its own powers against an executive branch that was encroaching on the legislative process. The bureaucracy also safeguards Congress’s own internal separation of powers, the salutary decentralization of law-producing responsibilities among a collection of nonpartisan actors, preventing any one aspect of the lawmaking process from coming under undue political or centralized control.

Understanding the congressional bureaucracy’s work also provocatively deconstructs the concept of a “statutory text.” The words Congress enacts are the result of a highly dialogic process that is triggered by and includes assumptions about critical inputs from the bureaucracy. Members and partisan staff focus on the substance of legislation at the macro level, not the specific words chosen at the micro level—that is the bureaucracy’s job. What we see when we open the statute books often is not even what Congress enacted or how Congress arranged it, because OLRC reorganizes and edits the laws after passage. So conceived, the concept of a “statute” is much more capacious than merely the “text” at the moment of the vote. None of this is illegitimate; Congress has set itself up this way. All of these inputs are part of the “text” as Congress intends it to be understood.

Together, these institutions paint a picture of a Congress that is not as irrational as the public considers it to be. They also have on-the-ground lessons for statutory interpretation, highlighting critical inputs that courts miss and numerous statutory cues—from code placement to consistency of language to CBO scores—some of which courts dramatically overread, others of which should be attractive even to textualists because they result from formalist, objective, collectively congressional action. The field is now engaged in emerging debates about whether doctrine can absorb this kind of detail about legislative process; understanding the congressional bureaucracy is a critical new piece of this account.

Share

COinS