The Government Performance and Electoral Accountability Act

By Achuthan Panikath

Mon Aug 03 2026

Promises without denominators Nearly every consequential institution in American life is expected to state its goals in measurable form. Corporations publish targets and answer to them quarterly. Federal agencies, under the Government Performance and Results Act (GPRA) of 1993 and its 2010 modernization, publish strategic plans, set performance goals, and report results. Grant recipients define deliverables. Even small nonprofits are asked by their funders what success will look like and how it will be counted.

Electoral politics runs in the opposite direction. Candidates promise outcomes, cheaper healthcare, safer streets, a stronger border, revived manufacturing, in forms deliberately unfalsifiable: no baseline, no target, no deadline, no measure. The vagueness is rational for the candidate, since a promise without a denominator can never be broken. It is corrosive for the voter, who is asked to hire someone for the most consequential job in public life on the basis of commitments that cannot be evaluated afterward even in principle. Election seasons then relitigate the past through dueling statistics, with no agreed record of what was promised or what was delivered.

The Government Performance and Electoral Accountability Act would build the missing record: a voluntary, standardized Candidate Public Commitments Registry, through which candidates for federal office may publish measurable commitments before an election, and an independent institution that tracks delivery afterward.

The anatomy of a registrable commitment A commitment entered in the registry would carry six components: a baseline, the current measured state of the thing to be changed; a target, the intended state; a deadline; a measurement method, specifying the data source and calculation; the responsible institution, identifying who would actually have to act; and dependencies, the external conditions on which delivery relies. In practice: reduce average federal permitting time for major infrastructure projects from its published baseline to a stated number of days by a stated year, measured by a named agency dataset, dependent on enactment of specified legislation.

The registry would classify each commitment by the candidate's actual span of control, and this classification is the intellectual core of the design. Directly controllable commitments lie within the powers of the office sought. Shared-control commitments require cooperation from other branches, states, or agencies. Legislative commitments require Congress to act. Aspirational commitments describe outcomes, such as macroeconomic conditions, that no official controls and every official campaigns on anyway.

The classification exists because the naive version of political accountability, treating every unmet promise as a broken one, is both unfair and analytically useless. A president cannot guarantee outcomes controlled by courts, markets, foreign governments, or a hostile Congress, and an accountability system that pretended otherwise would punish honesty and reward the vaguest possible promising, recreating the original problem with extra steps. Classifying control up front lets evaluation ask the right question for each commitment type.

Evaluation After each election, a nonpartisan Public Performance Office, structured for independence on the model of the Congressional Budget Office (CBO) and the Government Accountability Office (GAO), would track registered commitments of officials who won. It would publish annual progress reports; completion status against each baseline, target, and deadline; the data and methodology behind every assessment, drawing on official statistics wherever they exist; and, critically, the official's own explanation of any missed target alongside the office's analysis of why the deviation occurred.

Two things the office would never publish: a single composite politician score, and any judgment expressed in the vocabulary of praise or blame. Reducing a public career to one number would be false precision in service of headlines, and evaluative rhetoric would spend the office's credibility on exactly the commodity it cannot afford to sell. The dashboard states what was promised, what was measured, what happened, and what explanation was given. Interpretation belongs to voters, journalists, and opponents, who will supply it enthusiastically.

Why voluntary participation is not fatal The registry's obvious vulnerability is that it is optional, and the candidates most allergic to accountability simply will not use it. True, and anticipated. The design wagers on competitive dynamics rather than compulsion. A candidate who registers measurable commitments acquires an argument: my promises are on the record and checkable; my opponent's are fog. Once one credible candidate in a race uses the registry, declining it becomes a message in itself. Debate moderators and journalists gain a standing question with a factual answer. The mechanism resembles the way voluntary financial disclosure, candidate debates, and tax-return release became near-mandatory through norm pressure rather than statute, and where norm pressure fails, the registry still delivers value through whoever does participate. Compulsion, by contrast, would raise First Amendment compelled-speech problems and poison the institution's legitimacy at birth.

A deeper criticism deserves a direct answer: politics is not a spreadsheet, and governing is judgment under uncertainty, not target-hitting. Correct, and the design already concedes it. The system evaluates commitments, not character, and its most important output for a missed target is not the miss but the explanation. Circumstances change; a pandemic, a war, or a court ruling can invalidate any plan. The accountability question the registry actually poses is the one mature institutions pose internally: when reality diverged from the plan, did the official report the divergence honestly and adjust course, or pretend the promise had never been made? That question is answerable, fair, and currently never asked with any rigor at all.

Summary of the proposal Establish a voluntary, standardized Candidate Public Commitments Registry for federal candidates, with each commitment specifying baseline, target, deadline, measurement method, responsible institution, and dependencies, classified by degree of the official's control. Create a nonpartisan Public Performance Office, modeled on CBO and GAO independence, to publish annual, methodologically transparent progress reports on elected participants' commitments, including officials' own explanations of deviations, with no composite scoring and no evaluative rhetoric.

Key references Government Performance and Results Act of 1993 and GPRA Modernization Act of 2010. Congressional Budget Office and Government Accountability Office institutional design and independence practices. Performance.gov and federal agency priority-goal reporting. Comparative work on election-pledge fulfillment, including the multi-country political-science literature measuring campaign-promise delivery rates.