Data Center Fiscal Impact Model

Adjust the parameters below to match a specific proposed facility. All revenue and cost calculations throughout the model will update based on these inputs. Default values represent a typical hyperscale data center project in a mid-sized Texas city as of 2026.

Before the fiscal arithmetic matters, the city must answer a prior question: Do we have the physical capacity to serve this facility — and if not, what does it cost to build it? This is the question citizens are asking. It is the right question. A data center that consumes the city's remaining water treatment capacity, or that requires electric grid upgrades at ratepayer expense, imposes real costs on existing residents regardless of how much property tax it generates.

This tab provides a transparent capacity assessment across every major municipal service category. Enter your city's current capacity and existing usage below. The model will calculate remaining headroom, show what the data center would consume, and flag any categories where capacity is insufficient.

"Citizens deserve to know: does this project fit within our existing infrastructure, or are we building new capacity to serve a private enterprise? And if we are building — who pays?"

The most frequent citizen objections to data center development fall into five categories. Each is legitimate and deserves a factual response — not dismissal.

Data center capital investment is the centerpiece of every industry pitch to local officials. The headline number — often exceeding $1 billion — is designed to impress. But the municipal fiscal analyst must decompose this figure to understand what actually generates taxable value and what does not. Not all investment dollars produce proportional revenue for the host city.

"A billion-dollar investment that is 75% abated generates less property tax than a $50 million shopping center at full value."

From the city's perspective, the critical distinction is between real property (land, building shell, site improvements — which are assessed by the appraisal district like any commercial building) and business personal property (servers, cooling equipment, UPS, generators, networking gear — which the operator self-reports and which depreciates rapidly). In many data centers, BPP at initial valuation exceeds real property value by 3-5x. However, BPP is the primary target of abatement agreements, meaning the largest component of taxable value is often the most heavily abated.

Property tax is the most significant long-term revenue source from a data center for Texas cities. It flows from two components: the real property (land + building) assessed by the county appraisal district, and business personal property (BPP) self-reported by the operator. The interplay between BPP depreciation, equipment refresh cycles, and abatement agreements creates a complex revenue profile that this section models year-by-year.

"The headline assessed value means nothing until you subtract the abatement. Model the NET taxable value."

Server equipment follows a steep depreciation curve. Under typical appraisal district schedules, a server purchased for $10,000 may be assessed at $7,000 in Year 1, $4,500 in Year 2, $2,500 in Year 3, and reach residual value by Year 5. However, data center operators continuously replace end-of-life equipment with new servers, creating a "treadmill" effect where aggregate BPP value stabilizes rather than declining to zero. The refresh rate — the percentage of equipment replaced annually — is the key variable. This model assumes a 70% annual refresh at steady state (meaning 70% of the initial BPP value is maintained through replacement cycles).

Sales tax from a data center has a distinctive two-phase profile: an enormous spike during construction and equipment installation, followed by a relatively modest steady-state flow during operations. Under current Texas law (Tax Code §151.359), qualifying data centers are exempt from the state 6.25% sales tax on equipment and operations, but local sales and use tax (up to 2%) continues to apply. This distinction is critical for municipal revenue modeling.

"The construction-phase local sales tax surge can be larger than a decade of operational collections. Don't let the spike distort your long-term projections."

The 89th Legislature is actively debating whether to repeal or cap the state exemption. If repealed, data center equipment purchases would become subject to the full 8.25% combined rate — a massive revenue increase for both state and local coffers. This model includes a scenario toggle for exemption repeal.

State-Exempt (6.25%): Servers, storage, networking equipment, cooling systems (CRAC/CRAH units, chillers), UPS systems, power distribution units, raised flooring, cable management, monitoring systems, and electricity (if qualifying facility is sole occupant).

Locally Taxable (1-2%): All of the above categories remain subject to local sales/use tax. Additionally, construction materials (concrete, steel, lumber, electrical conduit), furnishings, office equipment, and services not directly related to data processing remain fully taxable at both state and local rates.

Fully Taxable (8.25%): Tangible personal property incorporated into real estate (construction materials), services on exempt property, and rentals/leases of one year or less.

Water consumption is emerging as the most contentious issue in data center siting. A large data center using evaporative cooling can consume 3-5 million gallons per day — comparable to a small city. For Texas municipalities, particularly those dependent on surface water reservoirs or shared groundwater resources, this demand raises fundamental questions about infrastructure capacity, rate structure adequacy, and long-term water supply planning.

"Water is both a revenue source (utility billing) and a cost driver (capacity expansion). The fiscal question is whether the rate structure covers the true cost of service."

Unlike residential or commercial customers whose water demand spikes dramatically in summer and drops in winter, a data center operates at near-constant load around the clock, every day of the year. This baseload characteristic is a significant advantage for municipal water utilities. Residential peaking factors (peak day ÷ average day) typically run 2.0x to 3.5x, meaning the city must build treatment and distribution capacity far above average-day demand. Data centers, by contrast, have a peaking factor of only 1.2x to 1.5x (the modest seasonal variation reflects higher cooling demand in summer). This means the city gets more revenue per unit of installed capacity — the infrastructure investment works harder.

"A data center consuming 500,000 gallons per day at a steady rate generates more net utility revenue than 2,000 residential homes consuming the same annual volume — because the homes require 2-3x the peak infrastructure capacity."

The industry standard metric for data center water efficiency is WUE, expressed in liters of water consumed per kilowatt-hour of IT energy. Leading operators report WUE metrics of 0.19-0.25 L/kWh. However, these figures often represent global averages and may not reflect the actual consumption at a specific Texas facility, where high ambient temperatures increase cooling demand significantly during summer months. Municipal water planners should model for peak summer demand, not annual averages.

Electricity is the lifeblood of a data center. A 100 MW facility draws as much power as approximately 80,000 Texas households — continuously, 24/7/365. The power infrastructure required (substations, transmission lines, distribution switchgear) represents one of the largest cost components of data center development, and the grid impact on surrounding ratepayers is a growing concern across ERCOT territory.

"For cities with municipal electric utilities, a data center can be the single largest revenue source. For cities served by investor-owned utilities, the revenue is limited to franchise fees — but the grid strain is shared by all ratepayers."

A data center's power demand profile is fundamentally different from every other customer class. Residential demand peaks in the afternoon and evening and drops overnight. Commercial peaks during business hours and collapses on weekends. Industrial may run 2-3 shifts. But a data center runs at near-peak load continuously — 24 hours a day, 7 days a week, 365 days a year. This is an extraordinary advantage for the utility provider and, by extension, for the city.

The metric that captures this is load factor — the ratio of average demand to peak demand. Most commercial customers have load factors of 40-60%. A data center typically achieves 85-95%. This means the utility's fixed infrastructure (substations, transformers, distribution lines) earns revenue nearly 100% of the time it is in service. For a municipal electric utility, a data center is the closest thing to a guaranteed revenue stream. For a city served by an IOU collecting franchise fees, the high load factor means maximum franchise fee revenue per MW of connected load.

"A 100 MW data center at 90% load factor generates as much utility revenue as 150 MW of typical commercial load — because the commercial load sits idle every night and weekend."

ERCOT estimates that cumulative data center demand in Texas will exceed 22,000 MW by 2030. This is an enormous addition to a grid that has already experienced reliability challenges during extreme weather events. For individual cities, the question is whether the local transmission and distribution infrastructure can handle the load without upgrades that impose costs on existing ratepayers. Substation construction — which can cost $15-50 million depending on capacity — is increasingly the pacing item for data center projects and often requires utility or municipal capital investment with uncertain cost recovery timelines.

Employment is typically the most overstated benefit in data center proposals. Industry advocacy groups cite figures that include broad NAICS sectors and temporary construction workers. Municipal officials need to distinguish between three categories: temporary construction employment (large but transient), permanent on-site employment (small), and indirect/induced employment (real but diffuse). An independent analysis by Food & Water Watch estimated that as few as 23,000 people nationally worked directly in data centers as of 2024 — across more than 5,000 facilities.

"A $1 billion data center that employs 30 people costs approximately $37 million per permanent job in capital investment. Compare that to the $50,000-$150,000 per job typical of most commercial development."

Unlike residential or mixed-use development, data centers impose a distinctive cost profile on municipal services. They generate minimal demand for population-driven services (schools, parks, libraries, recreation) because they employ so few people. However, they impose significant costs in three areas: water/wastewater infrastructure and service, fire protection and emergency response capability, and transportation infrastructure during construction.

"The cost question is not just 'How much does it cost to serve?' but 'What infrastructure capacity must the city build or reserve that cannot serve other development?'"

Data centers present unique fire risks. They contain massive quantities of lithium-ion batteries (UPS systems), diesel fuel storage (backup generators), and dense electrical equipment operating at high voltages. While modern data centers are equipped with sophisticated suppression systems (clean agent, inert gas, or pre-action sprinkler), the host city's fire department must still maintain response capability. This may require specialized training, equipment (foam systems, hazmat capability), and potentially a dedicated or upgraded fire station if the facility is in a growth area. Annual cost of maintaining this capability is estimated at $150,000-$400,000 depending on existing department resources.

This is the tab that answers the fundamental question: Does the city come out ahead? By netting all identified revenues against all identified costs for each year of the 20-year analysis period, we arrive at the annual and cumulative net fiscal impact on the city's general fund and enterprise funds. A positive number means the data center generates more revenue than it costs to serve. A negative number means the city is subsidizing the facility.

All figures are presented in both nominal (undiscounted) and net present value (NPV) terms. The NPV figures discount future cash flows to Year 0 dollars using the discount rate specified on the Assumptions tab. The NPV is the number that matters for fiscal decision-making — it reflects what the 20-year revenue stream is actually worth to the city today.

"A positive net fiscal impact does not automatically mean the project should be approved. It means the fiscal arithmetic works. The city must still evaluate opportunity cost, community impact, water supply sustainability, and long-term strategic fit."

Each year's net fiscal impact (revenue minus cost) is multiplied by a present value factor: 1 / (1 + r)^n, where r is the discount rate and n is the year number. At a 4% discount rate, $1 received in Year 10 is worth only $0.676 today; $1 received in Year 20 is worth only $0.456. This discounting has a dramatic effect on the perceived value of long-term fiscal benefits, particularly the full-value property tax revenue that begins flowing only after a 10-year abatement period expires.

Revenue growth (default 1.5%) models modest annual increases from property value appreciation and utility rate adjustments. Cost inflation (default 2.5%) models the reality that municipal service costs — particularly labor and materials — tend to rise faster than revenues. Together with the discount rate, these three parameters determine whether the 20-year NPV is materially different from the nominal cumulative total.

This section presents three pre-configured scenarios to illustrate the range of possible fiscal outcomes depending on the abatement package offered, facility characteristics, and policy environment. Each scenario uses the same facility size but varies the key fiscal drivers.

None of these scenarios addresses the opportunity cost of the land. If 80 acres of commercially-zoned land in a growing Texas city were developed as a mixed-use center (retail, office, multifamily), what property tax, sales tax, and population-driven revenue would it generate? What employment density would it support? This "but for" analysis is beyond the scope of this model but should be part of any comprehensive evaluation. A rough rule of thumb: commercial/retail development generates 10-50x more permanent jobs per acre than a data center, with corresponding payroll tax, sales tax, and consumption-driven economic activity.

The fiscal impact model in this tool is not theoretical. The following case studies document actual data center abatement proposals in Texas cities as of April 2026. These cases illustrate the standard deal structure, the arguments made by developers to justify abatements, the citizen opposition that has emerged statewide, and the fiscal arithmetic that cities must evaluate. Each case provides a benchmark against which this model's outputs can be calibrated and validated.

"The best validation of a fiscal model is whether it reproduces the numbers that cities are actually negotiating. These cases prove it does."

These case studies confirm that the model built in this tool addresses every dimension that Texas cities are currently grappling with. The Fort Worth case validates the fiscal arithmetic — our model produces results in the same range as the city's published figures. The Temple case demonstrates why the Capacity Audit tab exists — without enforceable, public utility commitments, the city has no accountability mechanism. And the statewide legislative debate confirms that the era of automatic, unscrutinized data center tax breaks is ending.

For municipal clients considering a data center proposal, this tool provides what the developer's consultant will not: an independent, city-perspective fiscal analysis with NPV discounting, infrastructure capacity verification, honest employment benchmarking, and transparent abatement cost quantification. The question is not whether data centers are good or bad. The question is whether this specific deal, on this specific site, with this specific abatement structure, produces a positive net fiscal impact for the city's taxpayers after accounting for all costs and discounting future revenues to present value. That's what this model answers.