The Full Chain

Accountability is usually delivered in pieces. A watchdog group files a records request and publishes a finding. A consulting firm, hired by the government itself, recommends efficiencies. An advocacy organization runs an awareness campaign. A financial advisor, retained by the same officials under scrutiny, restructures the debt. Each piece is real, and each piece stops at the edge of its mandate. The citizen who started with a suspicion is handed from one incomplete process to the next, and the record never travels intact from the first question to the last dollar.

LGAI is built as one continuous process instead. We call it the full chain: six stages, one organization, one standard of evidence. What follows is what LGAI does at each stage, and what the citizen receives in return.

The chain begins with a person, not a press release. Citizens, neighborhood groups, and civic organizations engage LGAI because something about their local government does not add up — a tax increase without a visible service, a contract that always seems to go to the same firm, a debt figure that keeps climbing. Intake is the moment that concern becomes a case, and it is handled with deliberate discipline.

LGAI conducts a preliminary records review before committing to a full investigation: a first pass through the adopted budget, the most recent annual financial report, and the public disclosures already on the record, to determine whether the fiscal signal warrants the depth of a Seven-Pillar Audit. Most concerns are genuine; not all of them are investigable at scale, and saying so honestly at intake is part of the service.

What the citizen receives is a serious, unbiased hearing. LGAI brings no political assumptions to intake. Every case opens the same way — no position, no target, no assumption about who is responsible for anything — and the same analysis is run on every jurisdiction regardless of who governs it. Neutral at the start, honest at the end →

When the preliminary review shows sufficient signal, LGAI applies the Seven-Pillar Audit in full. The framework examines a local government across seven dimensions simultaneously — fiscal trajectory, headcount, service-demand alignment, procurement and contracting, technology adoption, disclosure quality, and stakeholder engagement and impact — because single-dimension audits miss the patterns that only appear when the dimensions are read together.

Every claim is sourced to a primary document. Budgets, payroll, board minutes, contracts, and campaign-finance filings are pulled into machine-readable form and benchmarked against peer jurisdictions, so that a figure is never presented in isolation but always against what comparable communities spend and staff. The methodology is published, not proprietary in the sense of being hidden: any competent citizen can follow it. Read the Seven-Pillar Audit Framework →

What the citizen receives is a documented record where there was once a suspicion — every number traceable to its source, every comparison defensible, every conclusion reproducible.

Where the record shows waste, abnormal spending growth, ballooning debt, or contract patterns that cluster in ways the public would not choose, the findings do not stop at the pattern. They identify the officials whose votes, signatures, and administration produced it. This is not editorializing. It is the arithmetic of self-government: every dollar in a public budget was proposed by someone, voted on by someone, signed by someone, and administered by someone.

Public spending is not weather. It does not simply happen. When a debt load doubles or a sole-source contract recurs, there is a decision behind it and a decision-maker behind the decision. Attaching accountability to names is not an act of bias against those names; it is the entire point of accountability research, and refusing to do it would be the failure.

What the citizen receives is clarity about responsibility — not a diffuse sense that “the government” is inefficient, but a specific, sourced account of which decisions, by which officials, produced which outcomes.

A finding that no one reads changes nothing. Stage 4 puts the record in front of the community that pays for it: published findings written in plain language, public dashboards that let residents explore the data themselves, citizen briefings, and civic-education campaigns that explain what the numbers mean and how local budgets work.

This stage is educational, and deliberately so. LGAI is a nonpartisan research institute; its role is to inform the public about its own government's fiscal record, not to tell anyone how to vote. The record is placed in the open, organized and sourced, and the citizens who own it decide what to do with it.

What the citizen receives is knowledge that was previously locked inside four-hundred-page PDFs and closed-session decisions — the fiscal reality of their own community, made legible.

Exposure diagnoses the problem; Stage 5 prescribes the remedy. Through the Citizens Savings & Efficiency Platform, LGAI turns forensic findings into a concrete, jurisdiction-specific plan: savings ranges derived from the community's own budget, payroll, contracts, and capital plans, sequenced into first-cut, second-cut, and third-cut priorities so that the easiest and least disruptive savings come first and the structural changes follow. See the Citizens Savings & Efficiency Platform →

The prescription includes technology-adoption plans benchmarked not to vendor marketing but to documented government deployments — the real, government-disclosed results already achieved by cities and counties that have automated permitting, records processing, and back-office workflows. LGAI does not promise what has not been done somewhere; it shows what has. Fifty Cities Already: the documented deployments →

Underneath the prescription is the Savings Engine, a variable-input model in active development that translates a jurisdiction's specific figures into modeled savings rather than generic benchmarks. What the citizen receives is not a complaint but a plan — a sequenced path from documented waste to recoverable dollars.

The final stage is the one no watchdog reaches, because it requires a different discipline. LGAI's team includes veterans of investment banking and merchant banking — people who have spent careers restructuring corporate balance sheets — and Stage 6 applies that same discipline to municipal ones. Debt portfolio review, refinancing analysis, capital re-prioritization, and reserve-policy design are the tools by which a community moves from “we found the waste” to “the tax line went down.”

A municipal balance sheet is not sacred or mysterious. It has obligations that can be refinanced, capital plans that can be resequenced, and reserves that can be governed by policy rather than habit. The same analysis that lowers a company's cost of capital can lower a county's, and the beneficiary is the taxpayer rather than a shareholder.

What the citizen receives at the end of the chain is the thing the whole chain exists to produce: lower taxes, consolidated obligations, and a government whose finances have been brought back into the service of the people who fund them.

A six-stage chain that ran on manual labor alone would investigate one community at a time and never catch up. What lets LGAI hold a single standard of evidence across every stage — and across many jurisdictions at once — is a customized-AI technology spine that runs the length of the chain. It is worth being precise about what that means, and equally precise about what it does not.

At the front of the chain, large-language-model tooling performs large-scale document ingestion: annual financial reports, adopted budgets, and board minutes — the four-hundred-page PDFs that have historically made citizen oversight impractical — are read, structured, and cross-referenced at a speed no manual team could match. LLM-assisted extraction pulls the relationships that matter across contracts, campaign-finance filings, and voting records, surfacing the patterns that Stage 2 then verifies and Stage 3 attributes.

The pipelines are strongest where the underlying data is already machine-readable. Florida's first-in-the-nation XBRL mandate is the model: under House Bill 1073 and Fla. Stat. §218.32, local financial statements are becoming machine-readable by statute and filed through the state CFO's LOGER-X portal. Where every other state's watchdogs hand-extract figures from PDFs, LGAI's pipelines can read the raw material of accountability directly — and the federal Financial Data Transparency Act of 2022 points every state the same direction.

Feeding the far end of the chain is the variable-input Savings Engine, which takes a jurisdiction's specific figures and models sequenced savings rather than quoting generic benchmarks. And through all of it runs a single, non-negotiable commitment: every model-assisted finding remains reproducible by a human from the public record. The AI accelerates the reading; it does not replace the reader, and it does not have the last word. Every output is analyst-reviewed before it is published. Our editorial and verification standards →

The framing is deliberate. Private firms are using this exact class of AI tooling to flatten their own cost structures — the workforce and productivity shifts LGAI has documented in named corporate actuals. LGAI applies the same class of tooling to the oversight of local government, so that the productivity revolution reshaping the private sector is turned, for once, toward the taxpayer's benefit rather than away from it. Productivity Parity: the corporate actuals →

Kendall Square is often called the densest square mile of innovation in the world, and the description is close to literal: a concentration of research institutions, technology firms, and data-science talent unmatched anywhere. LGAI has opened an office at 245 Main Street, inside the Cambridge Innovation Center, to place the model this paper describes inside that ecosystem.

The reason is methodological, not decorative. LGAI's peer benchmarking — the comparison of like jurisdiction to like jurisdiction that gives every finding its context — draws on the published Fiscally Standardized Cities methodology of the Lincoln Institute of Land Policy, which normalizes municipal finances so that cities with different service responsibilities can be compared honestly. That methodology is the reference standard behind LGAI's cross-jurisdiction work.

LGAI is also joining the disclosure-modernization community around machine-readable government data — the XBRL US State and Local Government effort to make public financial statements structured and comparable by default — and engaging the region's civic-technology community, including Code for Boston. Around all of it sits a deep fiscal-research ecosystem, from the Boston Fed's New England Public Policy Center to the government-performance and data-innovation programs at Harvard's Kennedy School, whose published work on municipal performance measurement informs the standards LGAI holds itself to.

A word on language, because precision here is a matter of integrity. LGAI is engaging this ecosystem, joining these communities, surrounded by these institutions, and drawing on their published methodologies. It is not partnered with, affiliated with, or endorsed by any university or research body named here. No institution in Cambridge has any relationship with LGAI beyond the ordinary one of published scholarship being available for anyone to learn from. We are neighbors in a research district, and we say only that.

What Cambridge offers, in the end, is proximity to the disciplines LGAI's model already runs on — fiscal standardization, machine-readable disclosure, civic technology, and government performance measurement — gathered in one place. The move does not change the model. It sharpens it.

The model is engineering, but the reason for it is older than any spreadsheet. Government is instituted to serve the people who consent to it and pay for it. When that service is well documented and openly disclosed, self-government works. When the record goes dark, the relationship inverts, and the machinery of the state begins to serve itself.

Thomas Jefferson understood the stakes of that darkness. Writing to Edward Carrington in 1787, he judged the informed public more essential than the government itself:

“Were it left to me to decide whether we should have a government without newspapers, or newspapers without a government, I should not hesitate a moment to prefer the latter.”

— Thomas Jefferson to Edward Carrington, 1787

Jefferson's point was never a preference for chaos; it was that a public able to see its government is the precondition for any government worth keeping. The collapse of local news has created exactly the darkness he feared — a nationwide oversight vacuum in which thousands of local governments now operate with no sustained scrutiny at all. LGAI exists to help rebuild the seeing, with the discipline of a research institute in place of the vanished newsroom. The Lag-Lead Trap: the oversight vacuum →

The inversion has an engine, and Frédéric Bastiat named it in 1848. Once the state becomes a channel through which organized interests draw more than they contribute, every party is tempted to reach for the same lever:

“The state is the great fiction through which everybody endeavors to live at the expense of everybody else.”

— Frédéric Bastiat, 1848

Bastiat's fiction is not inevitable; it is what happens when no one is watching the ledger. The incentive to live at everyone else's expense is checked only by a public that can see who is drawing what, and at whose cost. That is precisely the ledger LGAI's six stages are built to keep. In the tradition of political economy that runs through writers such as Murray Rothbard, the danger is understood plainly: an institution that raises its revenue by legal compulsion rather than by voluntary exchange carries no automatic discipline of the market, and so must be disciplined instead by transparency and by an informed citizenry.

And the remedy Jefferson prescribed for the state itself was restraint. In his first inaugural address in 1801 he described the government a free people should want:

“A wise and frugal Government, which shall restrain men from injuring one another, shall leave them otherwise free to regulate their own pursuits of industry and improvement, and shall not take from the mouth of labor the bread it has earned.”

— Thomas Jefferson, First Inaugural Address, 1801

“The bread it has earned” is the whole of it. Public money is taken from labor under legal compulsion, and that is exactly why it carries a higher duty of care than money entrusted voluntarily. A donor who dislikes how a charity spends can stop giving; a taxpayer cannot. LGAI holds to the stewardship standard articulated in its methodology work: funds taken by compulsion must be accounted for to a stricter standard than funds given by choice — because the people who provided them had no option to withhold. The stewardship standard →

The Cambridge office opens with a defined first-year program. LGAI will expand its publication of machine-readable datasets, so that the structured records behind its findings are available for citizens, journalists, and researchers to inspect and reuse. The Savings Engine will progress from an internal model toward a public tool, putting jurisdiction-specific savings estimates within reach of the communities that need them. And the investigation docket will grow beyond the seven communities currently active, as citizen referrals and documented fiscal signal warrant.

None of it works without the first stage, and the first stage is a citizen. The chain begins when someone who lives in a community decides that its books deserve a serious reading. The invitation is open, and it is concrete: submit a tip, request the methodology, replicate the framework. Every stage of this model is documented precisely so that it can be checked, reused, and carried into places LGAI has not yet reached. Submit a tip →

That is the resolution of the inversion this paper began with. The relationship between a community and its government is not fixed in whatever shape it has drifted into. It can be set right, deliberately, with a chain of work that runs from a citizen's first question to the last dollar returned to the people who earned it.

The government works for the people. Where that has been forgotten, LGAI exists to remind it — with records, not rhetoric.

— Local Government Accountability