Data centers: what people hear vs. what the evidence shows
Most claims about data centers are half right. This briefing checks the eight claims Texas residents and councils hear most often, gives each a verdict, and shows what a city can do about it.
The goal is to decide whether a project fits the community. False and misleading claims need to be set aside, whichever side makes them. Real effects on water, noise, traffic, power and the budget need to be measured, understood and communicated accurately.
The claims, checked
The verdicts are not one-sided. Several claims made by supporters of data centers fail the same tests as claims made by opponents.
Data centers will drain our water supply.VerdictMisleading statewide. Can be true locally.
Data centers use under 1% of Texas water today. But one 100 MW campus can use 0.8 to 1.2 million gallons a day with evaporative cooling, or under 12,000 with dry cooling. The cooling design decides it.
The state tracks how much water data centers use.VerdictFalse.
State surveys drew limited responses, and a city utility’s smart-meter data is shielded from public records. Facility-level reporting has to be written into the city’s own agreement.
Data centers will raise your electric bill.VerdictNot settled. Not supported in Texas to date.
Where bills rose elsewhere, the state’s market and rate rules set the path. ERCOT has no capacity market, but Texas’s review of how transmission costs are divided is due December 31, 2026.
The World Health Organization says 40 decibels at night is the limit.VerdictMisleading.
That figure was set for aircraft noise in Europe. For data centers, the type of noise matters as much as the level: low-frequency hum and tones need their own testing.
Backup generators are the state’s job, not the city’s.VerdictTrue, with a gap to close.
Most backup generator fleets need no state registration, so the state cannot tell a city how many are planned. The site plan should show every one.
Data centers pay big local taxes. / Data centers pay no taxes.VerdictPartly true, both ways.
The state sales tax exemption costs Texas over $1 billion a year. Property tax terms are set locally, and the incentive agreement can carry PILOTs and community benefit payments or projects.
These deals are made in secret.VerdictPartly true before signing. False after.
Negotiations can be confidential, but incentive terms that cost public money become public once the agreement is signed.
Closed-loop cooling means the data center uses no water.VerdictMisleading.
“Closed loop” describes how the computers are cooled, not the whole system. If the outdoor heat rejection is evaporative, it still consumes water.
The numbers behind the verdicts
Community benefit arrangements
An incentive is an exchange, and the community’s side of it can include more than tax revenue. A property tax abatement, a Chapter 380 or 381 agreement, or a development agreement can carry payments or projects that answer a project’s real local effects and leave a lasting public benefit. These terms are negotiated; none is automatic.
Community benefit payment
A fixed or scheduled payment to the city or a designated fund.
Emergency services, parks, community programsPayment in lieu of taxes
A PILOT replaces part of the taxes abated or exempted.
Keeps city revenue closer to the unabated levelDeveloper-built infrastructure
Improvements built to city standards and dedicated to the city.
Water and wastewater lines, reclaimed water, roads, drainageService and capacity support
Equipment, training or staffing tied to the demand the project adds.
Fire and EMS equipment, battery storage hazard trainingWorkforce and education
Partnerships with local schools and colleges.
Training programs, internships, scholarshipsMake it enforceable
State the amount or scope, the schedule, who receives it, how it adjusts if the project changes, and the remedy if the developer does not perform. Payments that only cover costs the project creates are reimbursement, not benefit.
Questions that cut through the noise
Four questions to ask of any number
- Water taken, or water used up? Withdrawal and consumption are different measures.
- National, or local? A small national share can still be large for one town’s system.
- Today, or a projection? A 2040 projection should not sit beside today’s figures.
- Who produced it, and who paid for it? Funding does not make a number wrong, but it should be stated.
Eight questions before major commitments
- What is proposed at full buildout?
- Can the water system serve it in drought and at peak demand?
- Who pays for each kind of electric infrastructure?
- What is the net fiscal benefit to each public entity?
- Which jobs are commitments, and which are estimates?
- What will neighbors experience, and what protects them?
- Who independently checks the key claims?
- Who monitors performance after approval?
Get in front of the decision
This briefing draws on Engage Group’s full-scale process for evaluating data centers and other Advanced Technology Infrastructure Projects, from first contact with a developer through verification after approval. It is customized to each community’s water and power systems, development standards, taxing entities and priorities.
- Development standards and the studies they require
- Fiscal impact analysis by taxing entity, with abatement, Chapter 380/381, PILOT and community benefit strategy
- Water, power and noise review requirements, including testing protocols
- Development agreement terms and verification after approval
- Briefings for elected officials and the public
Prepared for public education as of November 2, 2026. Every figure in the full briefing is cited to its source; questions and recommendations are Engage Group, LLC’s. Research from other states is context, not a forecast for a Texas project. Check regulations and tax eligibility against current authority before relying on them.