The gap between demanding and tracking
“Hold them accountable” is one of the most common phrases in political life. It appears in campaign speeches, activist manifestos, editorial pages, and social media posts across every issue and every political tendency. It is invoked after scandals, after broken promises, after disasters, and after elections. It is almost never followed by a plan for how the tracking will actually work.
The result is accountability as sentiment — a widely shared feeling that institutions should face consequences for failing to meet their commitments, disconnected from any structure that would actually do the holding. The phrase functions as an expression of anger and expectation. It does not function as a mechanism.
This is not a minor distinction. The gap between demanding accountability and having the infrastructure to track it is the difference between pressure that changes institutional behavior and pressure that dissipates after the news cycle ends. Outrage cycles are real, and they produce real effects — sometimes immediately. But institutional behavior over time is not primarily shaped by moments of outrage. It is shaped by whether anyone is watching consistently, whether the record of what was promised is maintained, and whether non-delivery carries a durable political cost.
What accountability tracking actually requires is not complicated in principle. It requires someone who is watching. It requires a record of what was promised, in language specific enough to evaluate. It requires a method for comparing promises to outcomes. And it requires that the comparison be publicly accessible and persistent — available after the news cycle that produced the original commitment has long since moved on. None of these elements are exotic. None of them happen by default.
What happens after the win
Civic energy follows a recognizable pattern. A problem becomes visible through a triggering event — a crisis, a death, an investigation, a leaked document. Attention accumulates. Organizing happens. Pressure is applied. A result follows: a law passes, a commitment is made, an official announces a new policy, a regulatory change is finalized. The cycle completes with a win that is real.
And then the pressure that produced the win dissolves. The coalition that built it disperses. The organizations involved move on to the next campaign, which is where the energy and the funding are. The media that covered the story follows the next story. The public officials who made commitments under pressure return to operating under the ordinary conditions of low scrutiny and diffuse expectations.
What is left: the text of a law that may or may not be implemented as written; the announcement of a commitment that may or may not be fulfilled; and no systematic structure for comparing what was promised to what actually happened.
This pattern repeats across issues and across political eras. The gap between environmental legislation and EPA enforcement is a structural feature of American regulatory policy, not an anomaly — the Environmental Integrity Project’s 2025 reporting{target=”_blank”} documents how deep cuts to state and federal budgets have left regulators unable to fully implement the protections the laws require. The gap between healthcare coverage expansions and actual access to care is well documented. The gap between police reform commitments made after 2020 and what departments actually implemented is now generating its own body of research: a 2025 Johns Hopkins study{target=”_blank”} found that while 48 states passed police accountability laws following the protests of that year, the most effective accountability measures — certification and decertification of officers — were adopted in only 13 states, and implementation challenges were reported across jurisdictions as a consistent finding.
The pattern is structural, not specific to any one issue or political alignment. It holds in states with Democratic governors and states with Republican ones. It holds for environmental law, housing policy, criminal justice reform, and healthcare access. The specifics vary; the structure does not.
The institutional actors who were targets of civic pressure know this pattern too. They have learned — because the pattern teaches — that commitments made under sufficient pressure tend to relieve the pressure. They have also learned that follow-through is rarely tracked with sufficient persistence to create a meaningful political cost for non-delivery. That lesson shapes their behavior in ways that are rational from their perspective and corrosive to accountability as a functional concept.
Why accountability fades — the structural reasons
Three structural features explain why accountability pressure dissipates after a win, even when the organizations that produced the win remain nominally active.
The first is attention economics. News cycles move forward, not backward. A commitment made under public pressure is news — it represents a change, a response, a visible result. Its non-fulfillment, months or years later, is rarely news. The original story has no audience by then. The partial implementation, the delayed rulemaking, the quietly revised enforcement guidance — these require context that takes time to acquire, and they generate none of the dramatic narrative that drove initial coverage. The people with the context to evaluate whether a commitment was actually met are rarely the ones in front of a camera or driving traffic. The structure of media attention, especially in an era of accelerating news cycles and shrinking newsrooms, systematically disadvantages accountability tracking.
The second is organizational discontinuity. The coalitions and organizations that produce civic wins are often held together by the energy of the campaign. A campaign has a defined goal, a timeline, a shared enemy, and a clear measure of success. When the campaign achieves its stated goal, the coalition has formally succeeded. It has no obvious mandate to continue doing the unglamorous, low-urgency work of monitoring implementation. That work does not generate the engagement that sustains organizations through funding cycles — it does not produce the mobilization moments, the rally photographs, or the demonstrable victories that organizational communications require. So it does not get done, not because the organizations do not care, but because the incentive structures that sustain organizations do not reward it.
The third is the complexity of implementation. A law or a commitment is a political document. Implementation is a technical and administrative process that unfolds over years through rulemaking, agency guidance, budget allocations, and enforcement decisions. Following that process — understanding what the regulatory language means, tracking whether the agency is conducting the rulemaking the law requires, monitoring enforcement actions, evaluating whether the announced policy is being applied as written — requires sustained technical engagement that most civic organizations are not built to provide over long time horizons. The people who have this capacity are rarely organized to maintain a public-facing accountability function.
The cumulative effect of these three features is that institutions learn, correctly, that accountability pressure is temporary. That lesson does not produce outright defiance — it produces something subtler and more durable: the systematic management of the appearance of compliance while minimizing actual obligation. Commitments are made in language that cannot be evaluated. Implementation timelines are extended. Metrics are revised when prior metrics show failure. The institutions are not cynical in any simple sense; they are adaptive, in ways that the structure of civic accountability makes rational.
What real accountability infrastructure looks like
Accountability infrastructure is not a new concept. It exists in several domains, has a documented track record, and provides a clear model for what the work requires. The question is not whether it can be built but why civic life so rarely builds it deliberately.
Journalism at its best performs accountability tracking. Beat reporters who cover the same institutions over time — who build context, who know the history, who can evaluate whether this year’s announcement is consistent with last year’s commitment, who maintain sources inside agencies and can report on the gap between stated policy and actual practice — represent accountability infrastructure. The capacity to evaluate a commitment requires prior knowledge of what was promised, which requires someone who was paying attention when the promise was made. The decline of that capacity{target=”_blank”}, particularly in local journalism, has had direct accountability consequences. Newspaper newsroom employment fell 57% between 2008 and 2020{target=”_blank”}, from roughly 71,000 jobs to about 31,000. The reporters who built institutional knowledge about local agencies, school boards, regulatory bodies, and municipal governments largely do not exist anymore. When no one is watching consistently, the cost of non-delivery drops.
Inside government, inspector general offices, GAO reports, and legislative oversight mechanisms perform a version of this function: documenting what agencies said they would do, tracking what they did, and making the comparison publicly available. These mechanisms are imperfect, under-resourced, and subject to political pressure. But they are built around the right structural logic — sustained, specific, comparative documentation. They exist because legislatures recognized, at some point, that the executive branch required a watching function that was separate from the executive branch itself. That logic does not become less relevant at the civic level; it becomes more relevant.
Some advocacy organizations have built accountability infrastructure for specific domains. FollowTheMoney.org{target=”_blank”} maintains comprehensive 50-state data on campaign contributions, independent spending, and lobbying — tracking who funds which candidates and with what amounts over time. GovTrack{target=”_blank”} has built a legislative tracking database covering bill progress, voting records, and missed votes. ProPublica’s Trump Town project{target=”_blank”} maintained a searchable database of 3,855 administration appointees, documenting their prior employment, lobbying histories, and conflicts of interest — information that would otherwise have required individual FOIA requests to assemble and would have been inaccessible to most people doing oversight work. These are examples of accountability infrastructure: specific, comparative, persistent, and publicly accessible.
What these examples have in common is worth stating explicitly. They are built around specific, answerable questions — what did this person promise, what did they vote for, who paid for their campaign, where did they work before they took this regulatory position? They are maintained over time rather than produced once and abandoned. They are designed for retrieval by people who did not produce them, rather than held as organizational assets accessible only to insiders. And they make comparison possible: you can look at what was stated and look at what happened and draw a conclusion that does not require taking anyone’s word for it.
The commitment-tracking problem specifically
One of the most common and least examined failure modes in civic accountability is the commitment made under pressure in language that cannot be evaluated.
“We will prioritize affordable housing.” “We are committed to environmental justice.” “We will work to ensure accountability in our department.” These formulations generate a news story at the moment of announcement. They satisfy the immediate demand for a response. They do not generate a measurable standard against which follow-through can be assessed, because they contain no measurable standard. What would it mean to fail to prioritize? How would anyone determine that the commitment to environmental justice was not fulfilled? The language is constructed to foreclose that question.
Functional accountability tracking requires that commitments be specific enough to evaluate. This is actually a civic capacity — the ability to push for specific, measurable commitments rather than accepting vague language that sounds like a win but cannot be tracked. “We will prioritize affordable housing” is not a commitment. “We will permit 1,500 new units of affordable housing in the following zoning categories within 18 months” is a commitment. The first produces a headline. The second produces a benchmark.
This specificity is often opposed by institutional actors precisely because it creates trackable accountability. Vague commitments relieve pressure without creating obligation. Specific commitments do both. The demand for specificity — for numbers, timelines, defined metrics, and named responsible parties — is therefore not a technical nicety but an exercise of civic power. It is the difference between a promise that can be monitored and a promise that cannot.
Tracking whether a specific commitment was met requires, in turn, a record of what was promised, a method for measuring what was delivered, a timeline against which both can be compared, and a structure that maintains the comparison after it is no longer newsworthy. The first three elements are relatively tractable if the commitment was made in specific language. The fourth — persistence beyond the news cycle — is the one that most consistently fails.
Accountability and power — why institutions resist it
Accountability is not politically neutral, and the resistance to accountability infrastructure is not incidental.
Institutions that benefit from low accountability have structural incentives to resist the development of systems that would hold them to their commitments. This resistance rarely takes the form of outright opposition to the concept of accountability — that would be a politically untenable position. Instead it is expressed through the forms described above: a preference for vague commitment language, the release of information in formats that are difficult to analyze, the use of bureaucratic framing that requires expert context to evaluate, the underfunding of oversight mechanisms, and the revision of performance metrics when prior metrics show failure. Each of these is defensible in isolation. Together they constitute a systematic management of accountability as a political risk rather than a substantive obligation.
The watch changes what happens. An institution that knows it is being observed by a persistent, technically capable, publicly accessible accountability structure behaves differently than one subject to periodic outrage with no follow-through. Police departments that know use-of-force data will be publicly reported behave differently than those that know it will not. Regulatory agencies that know their enforcement records will be tracked against their statutory mandates behave differently than those operating with de facto opacity. This is the mechanism by which accountability infrastructure produces effects — not through punishment after the fact, but through the anticipation of scrutiny that shapes behavior in advance.
Which means that those with interests in the current distribution of outcomes have structural reasons to oppose the development of that scrutiny. This connects directly to the power analysis in Core Idea #4: the problem is not simply that officials make promises and fail to deliver through inattention or incompetence. It is that the structure of accountability is itself contested terrain. Who watches, what they watch, what they can access, and whether their findings reach a public audience are all questions over which power is exercised. Building civic accountability infrastructure is not a neutral administrative function. It is a form of civic power that organized interests will predictably contest.
Naming this clearly matters because it changes the analysis of why accountability fades. It is not primarily because civic organizations are distracted or under-resourced, though they often are. It is not primarily because journalists have short attention spans, though news cycles are genuinely constraining. It is also because the reduction of accountability capacity serves the interests of institutions that would otherwise face accountability — and those institutions have resources, legal access, and structural position that civic organizations lack.
What America’s Plan treats as accountability infrastructure
Accountability is the fourth stage of this project’s pipeline — sentiment, plan, pressure, accountability — for reasons that follow from everything described above. The sequencing is not rhetorical; it reflects what civic work actually requires. Sentiment without a plan stays diffuse. A plan without pressure stays theoretical. Pressure without accountability tracking produces wins that do not hold. The fourth stage is not an endpoint but a structural requirement for the earlier stages to have lasting effect.
The theory of change underlying this project treats accountability tracking as a distinct form of civic work, separate from advocacy and from electoral organizing, with its own requirements and its own time horizon. What the project is working toward in that domain: the capacity to track specific commitments against specific outcomes, make that tracking publicly accessible, and maintain it long enough to be useful across the time horizon over which institutional behavior actually changes. That time horizon is typically years, not news cycles.
This infrastructure does not yet exist in fully built form here. The commitment-tracking tools, the comparative databases, the public-facing records — these are the direction, not the current state. Being clear about that is part of what accountability requires. An accountability project that overstates its own capacity would be doing the thing it is criticizing.
What distinguishes accountability infrastructure from accountability as slogan is finally reducible to four elements: specificity (what exactly was promised), persistence (the record exists after the news cycle ends), accessibility (anyone can find it without organizational access or inside knowledge), and comparison (what was delivered can be set against what was promised, publicly). Without those four elements, “hold them accountable” is an expression of how people feel about institutions that have failed them. With those four elements, it is a structure that has a chance of changing what institutions do.
The feeling is real and warranted. The structure is the part that does not build itself.
This article was researched and drafted with AI assistance under human review. See our full AI and editorial practices.