The Megawatt Is Becoming an Economic-Development Metric

For generations, regions competing for major investment have advertised a familiar collection of advantages: available land, transportation access, labor supply, tax structure, proximity to customers, universities, industrial infrastructure and incentives. Deliverable electrical capacity is moving toward the front of that list as more projects depend on substantial amounts of power being available at a particular location. The relevant measure extends beyond a state's generating capacity or a utility's nameplate total to the additional electricity that can reach a proposed site at an acceptable price, with sufficient reliability and within the investor's development schedule.

Artificial intelligence is accelerating this change alongside semiconductor manufacturing, advanced factories, electrified industrial processes and other large loads. These investments are arriving as U.S. electricity demand emerges from a prolonged period of relative stagnation. In its September 2026 forecast, the U.S. Energy Information Administration projected record consumption in both 2026 and 2027, driven particularly by data-center development and increased manufacturing activity, with electricity sales reaching approximately 4,135 billion kilowatt-hours in 2026 and 4,211 billion kWh in 2027.

That growth changes the practical requirements of economic development. During much of the software era, regions could attract offices, research centers and professional-service firms without electrical infrastructure becoming a defining constraint on the next major employer. AI facilities and many of the industries developing around them require a different scale of provision, making the availability of megawatts an increasingly important measure of what a region can actually accommodate.

From Available Acres to Available Megawatts

Industrial site selection has always involved power, particularly in steel, chemicals, aluminum, refining and other energy-intensive industries. The change lies in the range of digital activities for which electrical capacity has become decisive: a large data-center campus may require hundreds of megawatts, while several campuses in one region can represent gigawatts of prospective demand. Semiconductor fabrication and advanced manufacturing can impose similarly concentrated requirements, bringing industries associated with technological advancement into direct competition for physical infrastructure.

The International Energy Agency estimates that around 20 percent of planned data-center projects globally could face grid-connection delays through 2030 because of electrical constraints. Its recommendation to locate facilities where both electricity supply and grid capacity are available reflects the difficulty of expanding transmission on a developer's preferred schedule. A hundred acres beside an interstate may appear attractive on a map, yet offer limited value to an electricity-intensive project if the nearest substation lacks capacity, network reinforcement will take years, or equipment and generation cannot be obtained economically.

An available industrial site therefore needs to be evaluated in terms of both its land and its electrical characteristics. Development officials should be able to establish how many megawatts are deliverable today, what could be added within three years, and whether transmission constraints, interconnection procedures, generation plans and electricity costs support the proposed investment. They also need to understand whether the system can accommodate later phases, because the capacity to energize an initial facility does not by itself establish a credible path to expansion.

Speed to Power

The interval between site selection and energization is becoming an increasingly valuable competitive measure, often described as speed to power. Businesses already evaluate time to market because capital committed to an unfinished facility produces little operating value. If one region can energize a site in eighteen months while another requires five years of studies and network upgrades, the difference can outweigh many conventional development incentives by changing when the investment begins producing returns.

The underlying mismatch in construction schedules is substantial. The IEA's 2026 electricity analysis reports that more than 2,500 gigawatts of generation, storage and large-load projects are stalled in grid-connection queues worldwide, while estimating that data centers can be constructed in roughly one to three years and new grid infrastructure can require five to fifteen years from planning through completion. A facility can consequently be ready for use long before the electrical system serving it is ready to supply the required load.

Lawrence Berkeley National Laboratory's 2026 Speed to Power study describes bottlenecks across forecasting, interconnection, resource procurement, market operations and cost allocation as data centers and other major loads expand. The study catalogued more than forty potential approaches being considered or implemented to accelerate connections, illustrating how many institutions and procedures affect the development timetable. Regions that can credibly shorten that timetable create an economic asset whose value, for some industries, can rival highway proximity or a favorable incentive package.

Power Availability Is Not the Same as Power Generation

A state can generate large quantities of electricity and still struggle to serve a particular new facility because power must reach the customer through transmission and distribution systems. Substation limits, congested lines, reliability requirements and the procurement schedules for transformers and switchgear all affect the amount that can be delivered. Evaluating generation alone therefore leaves out constraints that may determine whether a proposed investment can proceed.

The Department of Energy's 2026 draft National Transmission Needs Study identified data centers, domestic manufacturing and other large industrial loads among the drivers creating a pressing need for additional transmission infrastructure. Equipment availability adds another difficulty: the IEA reports that lead times for important grid components, including transformers and cables, have lengthened substantially, while deliveries of new gas turbines can also require several years. The ability to finance a project does not ensure that the necessary equipment will arrive within its construction schedule.

Regional assessments need to distinguish among generation capacity, transmission capacity, connection timelines, upgrade costs and room for subsequent growth. A site may have access to substantial generation but inadequate transmission, adequate delivery infrastructure but insufficient new supply, or both resources behind an interconnection process too slow to meet the investor's needs. Even where an initial phase is feasible, expensive reinforcement or limited electrical headroom may weaken the case for expansion. The commercially useful measure is deliverable electricity at the required location, quantity, price and time, with each condition affecting the value of the others.

The New Infrastructure Arbitrage

Geographical differences acquire economic value when electrical capacity becomes a constraint. Consider two regions offering comparable land costs, transportation access, construction labor, business conditions and incentives, but with different power-development schedules. One can deliver 500 megawatts within two years and provide a credible route to one gigawatt as a campus expands; the other can initially deliver 100 megawatts but needs a new transmission line, additional generation, substation expansion and a six-year approval and construction process to reach 500 megawatts. Their conventional advantages may be similar, but they offer materially different investment opportunities.

The first region possesses electrical optionality, meaning that a business has room to expand without repeatedly confronting a binding power constraint. That flexibility has value both at the initial investment decision and when the company considers subsequent phases. Regions that maintained strong generation systems, built transmission capacity or preserved underused industrial infrastructure may consequently possess advantages that received less attention during the years of stagnant electricity demand.

This can give established industrial regions a new role in technological development. Transmission corridors, substations, rail access, industrial land, pipelines and utility expertise built around manufacturing can also support computational infrastructure and advanced production. The resulting opportunity depends on the condition and availability of those assets, but it allows the geography of future technology investment to draw on capabilities developed outside the previous generation of technology clusters.

AI Changes the Economic Meaning of Electricity

The connection to the larger Power → Compute → Intelligence framework begins with the uses to which electrical capacity can be put. Electricity supplied to a steel mill supports heat, mechanical work and manufactured material, while electricity supplied to a data center supports computation that can be applied to software development, scientific modeling, medical analysis, financial research, engineering optimization, logistics and machine reasoning. Although the outputs differ, both production processes depend on facilities and infrastructure in the physical economy.

The same dependence brings energy infrastructure into the systems that produce machine intelligence. Transmission lines connecting computational clusters, natural-gas turbines and nuclear reactors supplying reliable generation, and transformers linking campuses to the grid all support the delivery of computing services. Utilities, technology companies, infrastructure investors, semiconductor firms, energy producers, municipalities and industrial developers therefore participate in an increasingly connected production system, with the electrical network supplying an essential input to cognitive work.

Data Centers Make the Change Visible

The projected scale of data-center demand helps explain the urgency of these questions. Lawrence Berkeley National Laboratory's 2025 update, published in June 2026, estimates that data centers could account for approximately 11.8 percent of total U.S. electricity consumption by 2030 under its central estimate, with modeled scenarios ranging from 9.5 to 15.3 percent. EIA likewise describes data centers as an emerging dominant driver of long-term electricity-demand growth following roughly fifteen years of comparatively flat national consumption.

For regional strategy, the location of that demand is as consequential as its national total. Electricity available elsewhere provides little immediate benefit to a development site when the network cannot deliver it, and the IEA notes that approximately half of U.S. data centers under development are located in existing major clusters. Continued concentration can intensify local bottlenecks even where the broader electricity system appears capable of supporting additional consumption.

Constraints in established markets can create openings for regions able to offer a better combination of power and infrastructure. Some investment may move as developers encounter limits in transmission, generation, land or water availability, provided alternative locations can meet the project's other requirements. The opportunity thus extends to places capable of demonstrating that they can support the physical scale of advanced technology, including regions without a long-established identity as technology centers.

A New Site-Selection Scorecard

Economic-development organizations should examine electrical infrastructure with the specificity they already apply to transportation. A logistics assessment considers interstate connections, congestion, freight corridors, rail access, airport capacity and travel times rather than relying on the general observation that roads exist. An assessment of power should similarly establish which facilities the system can serve, under what conditions and on what schedule.

The first group of measures concerns available capacity, expansion capacity and time to connection: the load the current system can accommodate, the additional load that can realistically be added, and the date a facility can begin operating. These figures need to be considered alongside electricity cost and reliability, since technically available capacity may still be unsuitable if its price or resilience fails to meet the customer's operating requirements. Together, these measures describe whether a site can support the project's initial business case and subsequent growth.

A second group examines how credible those commitments are in practice. The generation pipeline and transmission strength indicate whether supply and delivery can expand as promised, while equipment availability and regulatory execution affect whether transformers, substations, approvals and interconnection work will arrive on time. A technically grounded assessment connects each forecast of available megawatts to the infrastructure and decisions required to make it real.

These measures join the existing considerations of talent, capital, transportation, taxes, housing, education and institutional quality. For electricity-intensive investment, however, power can operate as a threshold requirement: incentives, workforce training, improved roads and assembled land cannot make a 500-megawatt facility function with a 100-megawatt supply. Development strategy must recognize that physical limit before attempting to compensate for other disadvantages through financial or administrative support.

Municipalities Need to Understand What They Are Selling

Municipalities pursuing technology investment need a detailed understanding of the electrical systems serving their principal development sites. An organization seeking data centers, advanced manufacturing, robotics or semiconductor-related investment should know the available capacity and connection constraints of its industrial areas almost as well as it knows their acreage and zoning. General assurances about energy availability become less useful as prospective customers require specific commitments about the scale and timing of service.

Understanding those resources also improves the decision about which investments to pursue. A hyperscale data center can require extensive infrastructure while directly employing fewer people than many conventional industrial facilities, making water consumption, grid costs, tax incentives, land use and the allocation of electrical capacity important parts of the economic assessment. The merits will vary by project and locality, so the ability to accommodate a facility should inform a comparison of its benefits and costs rather than create an automatic presumption in its favor.

Electrical capacity is consequently a measured economic-development resource with competing uses. If a region has 400 MW of near-term excess capacity and allocates 300 MW to one project, it reduces the capacity immediately available for manufacturing facilities, housing growth, another technology campus or future industrial expansion. Evaluating those opportunity costs requires officials to consider the long-term economic value generated by different uses of the same constrained infrastructure.

The Utility Is Becoming Part of the Development Team

Utilities have long worked with major industrial customers, but the expanding range of large electrical loads deepens their role in regional development. Planning horizons, generation portfolios, transmission programs and interconnection procedures can directly influence whether a location can attract and accommodate investment. Coordination with economic-development organizations therefore needs to begin early enough for infrastructure conditions to inform recruitment and site selection.

The participants bring different knowledge to the same investment decision: a municipality understands its available sites, grid planners can determine whether a nearby corridor can accommodate another 200 MW, and the utility can assess upgrade schedules and equipment procurement. Bringing those perspectives together allows a development proposal to connect its land, infrastructure and power commitments to a feasible construction timetable. Regions presenting that kind of technically grounded plan offer investors a stronger basis for action than general assurances that electricity will be available when needed.

Abundance Creates Optionality

Infrastructure capacity gives regions room to accommodate economic opportunities as they emerge. Transportation networks support additional commerce, water systems support population and industry, telecommunications support digital businesses, and electrical headroom allows new facilities to connect as demand develops. The value of that capacity includes uses that planners cannot identify precisely when the infrastructure is first built.

Transmission planners twenty-five years ago could not have known the eventual electrical requirements of frontier AI models, yet the systems they built may now help accommodate those loads. Capacity that appears excessive under one set of economic conditions can become valuable when technologies and industrial requirements change. Planning exclusively around a static view of demand can therefore leave a region poorly positioned to respond to opportunities outside the assumptions of its original forecast.

Recent U.S. demand growth illustrates that risk. EIA reported that electricity demand increased only around 0.1 percent annually between 2005 and 2019, compared with approximately 1.7 percent annually between 2020 and 2025, with data centers now contributing substantially to the acceleration. Infrastructure planned with room for expansion provides more flexibility when growth returns than a system whose available capacity and development procedures assume prolonged stagnation.

Economic Development Is Becoming Energy Strategy

Advanced economic development increasingly requires an understanding of the production system beneath the industries a region hopes to attract. AI depends on computation delivered through data centers, whose chips, networking and cooling systems require electricity from generating facilities supported by fuel, equipment, capital and permitting. Transmission lines, transformers and substations must then deliver that power to the machines performing the work, making each stage a potential constraint on investment.

These connections call for closer coordination between development agencies, utilities, infrastructure providers and private investors. The practical task is to ensure that physical systems can respond when businesses identify opportunities, with credible plans for the capacity, approvals and construction required. Regions can improve their competitive position by removing obstacles to viable investment and aligning infrastructure development with the needs of emerging industries.