Police Patrol vs. Fire Alarm Oversight

Reference. Reports what the sources say. States no conclusion of ours.

Edition label: 2026-11

Not everything on this page is footnoted. Our citation check covers the main text, but not headings, the lists of related entries and sources, or open questions, which may state facts without a footnote. Check the footnoted text or the source before relying on them.

Definition: Typology of congressional oversight set out by Mathew D. McCubbins and Thomas Schwartz in “Congressional Oversight Overlooked: Police Patrols versus Fire Alarms,” American Journal of Political Science 28(1), Feb 1984.1 The original distinction2:

McCubbins and Schwartz argued that fire-alarm oversight is rationally preferred by legislators because it conserves their scarce time and attention while allowing third parties to bear the search costs.5

Application to Intelligence Oversight

The original McCubbins-Schwartz distinction was refined by Ogul & Rockman (1990) (Table 1, p. 14), which, following Aberbach (1987), treats the original framework as confusing two distinct dimensions and builds a matrix from them:6

Active Reactive
Centralized A — empirical NULL SET (Police patrol as McCubbins-Schwartz defined it; no actual oversight body fits this cell)7 B — Special select committees in the aftermath of public scandal (Ogul and Rockman mention the Iran/Contra investigatory committee elsewhere as an example of chamber-wide oversight)8
Decentralized C — Police patrol as modified here (Aberbach Appropriations subcommittees fit this cell)9 D — Fire alarm as McCubbins-Schwartz defined it (still the most frequent form of oversight)10
  1. The original “police patrol” cell is empirically empty.11 No actual congressional body conducts truly centralized + active oversight; what exists in practice is decentralized + active police-patrol activity in committee/subcommittee settings (cell C — Aberbach’s Appropriations subcommittees with rich intelligence networks).12
  2. Cell B (Centralized + Reactive) is occupied by special select committees.13 Ogul-Rockman mention the Iran/Contra investigatory committee as an example of oversight that reflects chamber-wide forces; the cell-B text itself names no example.14
  3. Fire-alarm (cell D) remains the most frequent form even after the 1970s-80s growth in oversight activity.15 Reactive review is especially prevalent when information is uncertain or difficult to obtain and when overseers are risk-averse.16

Loch Johnson’s intelligence-oversight application. Per Johnson (2004b), the standard McCubbins-Schwartz framework needs further adaptation for intelligence oversight specifically — the “fire alarm” mechanism is unreliable because few interest groups exist in the intelligence domain, those that do can rarely air grievances in public given the classified nature of their work, and the agencies’ “high walls” make reporting very difficult.17 Johnson’s intelligence-specific claim is that the agencies’ “high walls” (his quotation of Dana Priest), while protecting them from foreign spies, also isolate them from the normal processes of legislative accountability.18

Examples of intelligence oversight:

Open Questions

Sources