Follow TexasMoneySupport

How we verify this data

Every figure Follow Texas Money publishes can be traced back to a document a government itself published, and reproduced from it by anyone. Here is exactly how.

From official bytes to a published figure

  1. Archive. We retrieve the official file from the publishing agency and store the original bytes unchanged. Re-fetching creates a new archived copy; nothing is overwritten.
  2. Verify. We record a SHA-256 checksum of the retrieved bytes, so a reader can confirm they are looking at the same file we parsed.
  3. Parse deterministically. A versioned parser reads the file the same way every time. The same input and the same parser version always produce the same output — no randomness, no locale-dependent parsing, no floating-point money.
  4. Reconcile. Every value is reconciled against the archived file before it can be published, and an independent reviewer can reproduce each published number from that file.

Three kinds of number, kept distinct

Follow Texas Money keeps three categories visibly separate so you always know what you are looking at:

  • Sourced fact — a value taken directly from an official document, shown with its exact amount and a link to that document. Almost everything on this site is a sourced fact.
  • Calculated metric — a value we compute from sourced facts using an approved, documented formula. Two are published: reported principal across displayed pledge categories for debt, and the derived revenue and spending totals for cities and counties. Both are labelled wherever they appear, and both are described below.
  • Potential indicator — a statistical observation that may merit attention. It is never a finding of wrongdoing, and none are published yet.

The one calculated metric we publish

Reported principal across displayed pledge categories is the sum of principal outstanding across the pledge categories shown for a single government, within a single stated fiscal year. Nothing else is added to it.

It is deliberately not called total debt. The Bond Review Board compiles this data from what issuers report, and it excludes obligations of less than one year, commercial paper, Build America Bond subsidies, and conduit debt. A government’s complete obligations are not knowable from this source, so we do not claim to publish them.

The same sum taken across many governments — the statewide figure on Explore — adds each government’s own latest reported year. Governments report in different years, so that figure and the FY2025-only figure are shown separately and are not interchangeable.

City and county revenue and spending

Revenue and spending for Texas counties and incorporated cities come from the U.S. Census Bureau’s State and Local Government Finances survey, using the individual-unit files that report one government at a time.

The Census Bureau’s own caution comes first, because it changes how these figures should be read. The Bureau states that these individual-unit records are inputs used to produce its published estimates. They are not audited financial statements for an individual government, and the Bureau does not review them as a time series for any single government. A figure here tells you what the Bureau recorded for that government in that survey; it is not the government’s own audited annual report, and two years of it should not be read as a trend the Bureau stands behind.

Totals are calculated, not quoted. The Census file publishes no “total revenue” or “total expenditure” line. Every total shown here is the sum of a fixed, published list of Census item codes, and that list is recorded rather than assembled on the fly. Spending is also shown two different ways — by kind of spending and by function it was spent on. Those are two views of the same money and are never added together.

Not every figure was reported by the government. The Census Bureau marks where each value came from, and we carry that mark through rather than flattening it:

  • Reported — the government reported this value to the Census Bureau.
  • Census compiled — the Bureau took the value from another source, typically the government’s own annual financial report, rather than from a survey response.
  • Census corrected — a Census analyst corrected the reported value.
  • Census estimate — the Bureau imputed this value. The government did not report it and the Bureau has estimated it. This is not a reported amount and we never present it as one.

A total made only of reported values is labelled derived. A total containing any Census estimate is labelled an estimate, because a single imputed component makes the whole figure one. A total containing compiled or corrected values, but no estimate, says so too.

Each government is shown in its own fiscal year. The Census survey year is not a fiscal year: one survey covers fiscal years ending across a twelve-month span, so two governments in the same survey can be reporting periods five months apart. We show the fiscal year end each government actually reported. Where the Bureau published no fiscal year end, we withhold the observation rather than guess which of two years it belongs to.

A government with no figure is not a government with no money.The Bureau surveys every local government only in census years; other years are a sample. A city absent from a sampled year was not surveyed. We show nothing for it, never $0.

This does not replace anything. Debt still comes from the Texas Bond Review Board; the Census file contains debt figures and we do not publish them. School district revenue and spending still come from TEA PEIMS. Debt, school finance, and city and county finance are separate sourced sections, on different reporting bases, and are never combined into one number.

Public money and political influence: seven measures, never one total

Money that moves between governments, associations, lobbyists and campaigns is reported by several different agencies, under several different laws, and those reports overlap. A city’s dues payment to an association and that association’s own reported lobbying expense can describe the same dollars seen from opposite ends. Adding them would count much of that money twice while every individual figure stayed correct.

So Follow Texas Money keeps seven measures and never combines them. There is no “total influence spending” anywhere on this site, and there is no field in our data or our public API that could hold one.

  • Confirmed direct lobbying — money a government’s own accounting system labelled as lobbying.
  • Confirmed association dues — payments identified by the payer as dues or membership fees.
  • Payments to organisations that lobby — money a government paid to an organisation that retains lobbyists. A payment to an organisation that lobbies is not a payment for lobbying: a city’s premium to its regional risk pool is insurance, and the pool also retains lobbyists. Both are true and the second does not colour the first.
  • Organisation-reported lobbying — what a recipient told the Internal Revenue Service on its own return.
  • Registered lobby compensation — the bracket a lobbyist filed with the Texas Ethics Commission. TEC publishes a range, never a figure, so we show the range. We never convert it to a midpoint, a minimum or a maximum, because that number would be ours rather than the State of Texas’s.
  • Federal lobbying — income or expense reported under the federal Lobbying Disclosure Act. A different jurisdiction, kept separate.
  • Candidate finance — contributions and expenditures reported to the Texas Ethics Commission. Campaign money is not public spending.

Relationships between governments, organisations, lobbyists and candidates are shown only where an agency published the connection in a filing we archived. None is inferred from a similar name, a shared address, or two things appearing near each other. A registered relationship, a payment, a lobbying registration and a contribution are all sourced facts, and none of them is a finding of wrongdoing.

What we collected, and what we did not

A statewide figure is only honest if it says what is behind it. Every influence total on this site carries its collection state, and where a source was not collected we say which one and why rather than letting the gap disappear into a smaller number.

  • Collected. Texas Ethics Commission lobby registrations and activity, IRS Form 990 returns, federal Lobbying Disclosure Act filings, TEC campaign finance, PEIMS objects 6214 and 6495, and public payment registers published by governments on their own verified domains.
  • Not collected: official election results. The Secretary of State’s results are served behind a bot-detection challenge. We do not bypass bot detection, so no election outcome appears anywhere on this site. That is a collection failure on our side, not a statement about any election or any candidate.
  • Not published yet. Everything acquired for this milestone is withheld pending an explicit, recorded publication decision. Where you see “no figures are published”, the data may well exist in our archive — it has not been cleared for publication, and that is deliberate.

Where a figure is incomplete we write “at least” in front of it. A floor presented as a total is a false completeness claim, and a missing value is never rendered as $0 — those are different facts about a government and we keep them apart.

How address lookup works

“Find my governments” answers one question: which governments have jurisdiction over a location. It works in four steps, and the separation between them is deliberate.

  1. The address you type is sent to the U.S. Census Bureau’s geocoding service, which converts it to a map location. Follow Texas Money does not save your address in its database. We cannot speak for what the Census Bureau logs.
  2. We check which published government boundaries contain that location, using official boundary data from the U.S. Census Bureau (counties, incorporated places, school districts) and the Texas Commission on Environmental Quality (water districts).
  3. Each boundary is linked to a Follow Texas Money government record only by an exact official identifier — a Census geographic identifier, or the federal school-district identifier that TEA publishes alongside its own district number. We never link a boundary to a government by matching names. Two Texas school districts are both named “Northside Independent School District”, in different counties, and only their identifiers tell them apart. A boundary we cannot link exactly is shown as boundary identification only and carries no financial data at all.
  4. Financial figures come separately, through the same publication controls that govern every other page here. Finding a government at an address does not publish anything about it.

Three distinctions that matter in Texas

A postal city is not a municipality. The city name in a mailing address is a postal routing convention, not a statement of government. Much of Texas is unincorporated, and an address that reads “Houston” may lie outside the City of Houston. We report a municipality only when the location falls inside its published boundary, and we never substitute the nearest city.

A water service area is not a water district. The utility that delivers water to an address may be a private company, a city department, or a cooperative. We use TCEQ’s political water-district boundaries, which are governments. Several water districts can legitimately cover one address; that is a feature of Texas governance, not an error, and we list all of them.

A school district is not an attendance zone. We identify the school district that governs and taxes a location, not the campus a student would attend.

What the boundaries are, and are not

Published GIS boundaries are informational. They are not a legal, survey, or engineering determination, and the agencies that publish them say so. They also reflect a point in time, so recent annexation or district changes may not appear yet; each result shows the date its boundary data was retrieved. Near a boundary line, a result may be wrong in a way this data cannot detect — if a location falls on a shared edge we list every jurisdiction it touches rather than choosing one.

How current each source is

Every source below is collected from the publisher on a schedule we set from that publisher’s own release terms. This table says when we last collected each one and whether that is inside the schedule. It describes our collection, not the figures: a figure reported for a 2023 fiscal year is still that year’s figure however recently we fetched it, and nothing here withdraws or changes a published number.

Each source we collect, when we last collected it, and whether that is within the publisher’s own release schedule.
SourceLast collectedExpected scheduleStatus
Texas Bond Review BoardDebt Outstanding By Local Government2026-07-20122 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
Texas Bond Review BoardDebt Outstanding By State Issuer Searchable by CIB CAB VAR TRB2026-08-04107 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
U.S. Census BureauState and Local Government Finances - Individual Unit Files2026-08-10101 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
Texas Comptroller of Public AccountsLocal Hotel Occupancy Tax ReportingNeverAbout every 400 daysNot yet collectedWe have never collected this source, so nothing from it is published.
Texas Comptroller of Public AccountsProperty Tax Rates and Levies (city and county)2026-08-03108 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
Texas Comptroller of Public AccountsLocal Sales Tax Allocation Payment DetailNeverAbout every 45 daysNot yet collectedWe have never collected this source, so nothing from it is published.
Texas Comptroller of Public AccountsSpecial Purpose District Public Information Database (SPDPID)2026-08-03108 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
Texas Comptroller of Public Accounts, Data Analysis and Transparency DivisionTexas State Expenditures by County2026-08-04107 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.
Texas Comptroller of Public AccountsMonthly State Revenue Watch — All Funds (Excluding Trusts)2026-08-02109 days agoAbout every 45 daysOlder than expectedMore time has passed since we collected this than the publisher’s own release schedule. It does not mean the published figures are wrong — a figure for a 2023 fiscal year is still that year’s figure — it means a person should check for newer data.
Texas municipalities and counties (individually)Official annual financial reports (AFR / ACFR) and official financial transparency statements2026-08-1398 days agoNone establishedNo schedule establishedThis source has no fixed publisher schedule, so there is no interval it could be overdue against. Its collection date is still shown.
Texas Education AgencyDistrict 2024-2025 Financial Actual Data2026-08-1299 days agoAbout every 400 daysCurrentCollected within the period we expect this publisher to release on.

What we do not do

Apart from the single labelled metric above, we do not combine figures the source reports separately — general obligation, revenue, and limited pledge are always shown as distinct categories. We do not estimate missing values, we do not adjust for inflation or population, and we do not present an outlier as proof of misconduct. Where a government has no published figure, we say so — it never means the figure is zero.

We also do not add together the governments that share an address. A county, a city, a school district, and one or more water districts report on different fiscal years, pledge different revenue, and overlap one another. A single “total debt affecting your address” would look authoritative and would not mean anything, so we show each government’s own reported figure and leave it at that.