Last updated: April, 2026. Download as PDF.
Government document readability describes how easily a member of the public can understand, navigate, and act on information from an agency. The Plain Writing Act of 2010 requires covered federal documents to use plain writing, while state and sector-specific requirements vary. Adult literacy data and federal usability guidance both show why agencies should design information for people with a wide range of reading skills.
The Readability Gap in Numbers
The scale of the problem is documented across multiple independent studies, and the findings are consistent: government documents are dramatically more complex than their intended audience can handle.
The National Center for Education Statistics' PIAAC results measure adult literacy through proficiency levels, not school-grade equivalents. The results show that many adults have difficulty synthesizing, evaluating, or navigating dense text. That is a stronger and more accurate basis for plain-language practice than claiming that all adults read at one grade level.
The Center for Plain Language's Federal Report Card publishes annual agency assessments and its evaluation approach. Its findings repeatedly identify issues such as jargon, passive voice, organization, and unclear calls to action.
Readability formulas can help identify text that deserves review, but the Federal Plain Language Guidelines emphasize that audience, organization, word choice, information design, and testing all affect whether content works.
Why Government Writes This Way
Government documents are not difficult to read by accident. Five structural forces consistently push government writing toward complexity.
Legal caution
Government writers operate in a legal environment where every word can be challenged, litigated, or audited. The instinct to qualify every statement, add disclaimers, and use precise legal terminology creates documents that protect the agency from legal risk while failing to communicate with the public. A single federal notice may pass through legal review, policy review, and program review — each layer adding complexity, caveats, and jargon. By the time a document reaches the public, it often reads like it was written for lawyers rather than citizens.
Institutional inertia
Many government documents are revised from earlier templates. Agencies may be reluctant to overhaul legacy documents because revision requires time, legal review, policy review, and stakeholder coordination.
The curse of expertise
Government employees who write public-facing documents are typically subject-matter experts. They understand their program's regulations, eligibility criteria, and processes intimately. This expertise creates a cognitive bias known as the "curse of knowledge" — the assumption that the reader shares the writer's familiarity with technical terms, acronyms, and bureaucratic processes. An eligibility specialist at CMS may use the phrase "modified adjusted gross income" without realizing that most Americans cannot define any of those three words individually, let alone the compound concept.
No feedback loop
In most government agencies, there is no systematic mechanism for the public to report that a document was confusing. The Plain Writing Act requires agencies to maintain a feedback mechanism, but in practice these are email addresses that receive little traffic and generate even less follow-up. Without feedback, writers have no data on which documents are failing and no incentive to improve specific content. Compare this to private-sector digital products, where A/B testing, user research, and customer support tickets create a continuous stream of usability data.
No institutional measurement
The most fundamental barrier is the absence of institutional measurement. Most agencies do not know how many public-facing documents they maintain, what reading level those documents are written at, which documents generate the most confusion calls to their contact centers, or how their writing quality compares to the Federal Plain Language Guidelines. You cannot improve what you do not measure.
What the Readability Gap Costs
The cost of unclear government writing is not abstract. It manifests in measurable economic harm, increased administrative burden, and reduced access to benefits and services.
Direct administrative costs
Unclear instructions can shift work from self-service channels to call centers, correspondence teams, and field offices. The federal government tracks paperwork burden through the Office of Information and Regulatory Affairs; however, those totals should not be presented as the cost of unclear writing without a study that isolates causation.
Benefit non-participation
Complex application processes can contribute to people failing to complete or renew benefits. The Office of Evaluation Sciences publishes federally conducted evaluations of ways to reduce administrative burdens and improve access to public programs.
Compliance errors
When people cannot understand what government requires, errors and missed deadlines become more likely. Federal paperwork estimates measure time spent responding to information collections, but they do not isolate how much of that burden is caused by writing quality.
Health outcomes
In healthcare, unclear communication can affect a person's ability to find, understand, and use information. The U.S. Department of Health and Human Services' Health Literacy Online guidance translates health-literacy research into evidence-based practices for digital content.
Why It Has Not Been Fixed
The Plain Writing Act became law in 2010. Over 15 years later, the average federal document still scores a C for writing quality. Three structural factors explain the persistence of the problem.
First, the law has no enforcement mechanism. Section 6 of the Plain Writing Act explicitly bars judicial review of compliance. No citizen or organization can sue an agency for failing to write clearly. Compliance is enforced only through annual reporting and public accountability, which has proven insufficient to drive sustained improvement.
Second, no vendor market exists. For web accessibility, the existence of WCAG scanning vendors (Level Access, Deque, Siteimprove, etc.) created a market dynamic where agencies could purchase compliance as a service. For plain language, no equivalent vendor ecosystem has emerged. Agencies have no one to call, no RFP to issue, and no line item to budget. The absence of a vendor market is both a symptom and a cause of the compliance gap.
Third, plain language has been framed as a culture and training problem rather than a technology and compliance problem. Agencies invest in workshops and style guides — which are valuable but insufficient at institutional scale. Training changes individual behavior; compliance infrastructure changes institutional outcomes.
The Technology Shift
The emergence of large language models in 2023 fundamentally changed what is possible. Before LLMs, evaluating documents for plain language quality beyond simple readability scores required human expert review. Scoring a single complex document for jargon density, logical flow, audience appropriateness, and actionability could take a trained reviewer 30 to 60 minutes. Scoring an agency's full corpus of 10,000 to 50,000 documents was economically infeasible.
AI-assisted auditing can perform a first-pass review across a large corpus and surface documents for human attention. Agencies should validate automated findings, especially for legal, benefits, safety, and health content, rather than treating model output as a compliance determination.
Technology can support a plain-language program, but accountable governance, subject-matter review, usability testing, and continuous editorial work remain essential.
Primary sources
- Plain Writing Act of 2010, Public Law 111-274 (GovInfo)
- Federal Plain Language Guidelines
- NCES Program for the International Assessment of Adult Competencies
- Office of Information and Regulatory Affairs paperwork reporting
- Office of Evaluation Sciences project library
- Health Literacy Online, U.S. Department of Health and Human Services
gpt.us.org is building the vendor ecosystem that plain language compliance has never had. See your agency's compliance score →