Live on XaptiqUGC NET Economics (Dec 2026) and IIT JAM Economics (EN) test series are now live on Xaptiq.View test series
ExplainedInfrastructure > Data CentresIndian EconomyInfrastructureIndustry

NITI Aayog wants grid and water checks before states clear data centres

A NITI Aayog framework would make states approve data centres on proven electricity and water availability and measurable economic benefit, rather than on the size of the investment promised.

What happened

A framework prepared by NITI Aayog's Frontier Tech Hub proposes that states assess data-centre projects on electricity and water availability, environmental safeguards and measurable economic benefits, rather than on investment commitments alone. Developers, distribution companies (DISCOMs) and state transmission utilities would jointly prepare a Power Readiness and Ramp-up Plan before approval. For projects needing more than 20 megawatts (MW), it proposes mandatory automated demand-response readiness and integration with State Load Dispatch Centres, which balance electricity supply and demand across state grids. Additional load would be approved only after the necessary infrastructure is commissioned.

Context

India's data-centre capacity stood at about 1.6 gigawatts in mid-2026 and could reach 6 GW by 2029, requiring an estimated $110 billion in investment, according to real estate consultancy JLL. Google has announced a $15-billion artificial intelligence infrastructure investment in Visakhapatnam, while Amazon and Microsoft are expanding. On October 14, Minister of State for Electronics and IT Jitin Prasada said India was working on a dedicated data-centre policy, and said data centres in some countries were taking electricity away from consumers, who were consequently paying more. "We don't want to get into that," he said.

Why it matters

The framework's central move is a cost-allocation rule. Developers would bear the cost of infrastructure built exclusively for their projects, while the cost of shared grid upgrades would be distributed among beneficiaries. That decides who pays when one very large user needs the network extended - the user, or every other consumer on the same distribution licence. On the incentive side, states would calculate a project's net economic contribution by quantifying land, electricity, infrastructure, duty and capital subsidies separately, then setting them against fiscal contribution, employment, local procurement, research and skilling. Incentives would be released against measurable milestones and could be suspended if commitments are not met.

Impact for India

  • State governments: Approval would rest on a Power Readiness and Ramp-up Plan agreed with DISCOMs and transmission utilities, not on announced investment value.
  • Electricity consumers: Assigning project-specific network costs to the developer is designed to prevent the outcome the IT minister warned about, where ordinary consumers pay more.
  • Data-centre operators: Projects above 20 MW would need automated demand-response readiness and State Load Dispatch Centre integration, and incentives would be milestone-linked and revocable.
  • Water-stressed regions: States would prepare a 20-30-year water availability assessment, and allocations must not reduce dependable supply to existing users.
  • Local communities: Large campuses would face community impact assessments covering water, traffic, emissions, noise, waste and local employment, with MeitY running a centralised compliance audit.

Key data

Demand-response threshold: 20 MWAbove this, automated readiness mandatoryWith State Load Dispatch Centre integration
India's data-centre capacity: About 1.6 gigawattsCould reach 6 GW by 2029Mid-2026 estimate by JLL
Investment implied by that growth: $110 billionEstimated requirementTo reach 6 GW by 2029
Water availability assessment horizon: 20-30 yearsProposed for statesCovering groundwater, surface water, municipal supply and treated wastewater
Amazon's proposed Thane facility: 422 MWPlanned electricity requirementOn 53 acres, with 193 backup diesel generators

Concepts

Demand responseAn arrangement in which a large consumer reduces or shifts electricity use when the grid is stressed, in exchange for a tariff benefit or as a condition of supply.The framework would make automated readiness mandatory for data centres drawing more than 20 MW.
State Load Dispatch CentreThe body that balances electricity supply and demand across a state's grid in real time.Large data centres would have to integrate with it so their load is visible to the system operator.
Beneficiary-pays cost allocationA principle under which the cost of shared infrastructure is divided among those who use it, rather than socialised across all consumers.Developers would fund infrastructure built only for their project, while shared grid upgrades are split among beneficiaries.
Colocation data centreA facility that rents out physical space, power and cooling, as distinct from one selling cloud or computing services on top.The framework asks states to treat these differently from cloud, GPU computing and managed hosting providers when designing incentives.

Theory lens

Negative externality and internalisation - when one user's demand imposes costs on a shared resource, efficient pricing makes that user face the cost instead of spreading it across everyone else. The framework applies this twice: developers fund network extensions built only for them, and water allocations must not reduce dependable supply to existing users. Minister Jitin Prasada's warning that data centres elsewhere left consumers paying more describes exactly the externality the rule is meant to prevent.

Global context

The resistance is not unique to India. The article notes that protests against data centres in Thane and Visakhapatnam echo opposition in parts of the United States and Europe, where communities have questioned the impact of such facilities on electricity grids, water resources and local infrastructure.

Stakeholders

Gains: Electricity consumersFacility-specific grid costs would fall on the developer rather than being spread across the distribution network.
Gains: State load despatch systemProjects above 20 MW would be integrated with State Load Dispatch Centres and carry automated demand-response readiness.
Gains: Existing water usersAllocations to data centres should not reduce dependable supplies for households and agriculture.
Gains: State exchequersIncentives would be weighed against a project's net economic contribution after accounting for subsidies given.
Pressure point: Data-centre developersApprovals, additional load sanctions and incentive disbursal would all become conditional and reversible.
Pressure point: Projects in water-stressed locationsThey would face independent validation of water sources, metering and periodic reporting.

Practice Question — from the story

With reference to the NITI Aayog Frontier Tech Hub framework on data centres, consider the following statements:

  1. It proposes mandatory automated demand-response readiness for data centres with electricity requirements above 20 MW.
  2. It recommends that states prepare a 20-30-year assessment of water availability before clearing such projects.

Which of the statements given above is/are correct?

  1. 1 only
  2. 2 only
  3. Both 1 and 2
  4. Neither 1 nor 2
Reveal answer & explanation

Answer: Both 1 and 2

  • Statement 1: Correct because for data centres with electricity requirements above 20 MW the framework proposes mandatory automated demand-response readiness along with integration with State Load Dispatch Centres.
  • Statement 2: Correct because the framework asks states to prepare a 20-30-year assessment of water availability covering groundwater, surface water, municipal supply, treated wastewater and competing demands.

Therefore, the correct answer is Both 1 and 2.

Concept check — test your understanding

The framework says a developer should bear the cost of grid infrastructure built only for its own project, while shared upgrades are split among those who benefit. Which economic principle is being applied?

  1. Marginal cost pricing, under which every user is charged a price equal to the cost of supplying the last unit.
  2. Cross-subsidisation, under which large industrial users are charged more so household supply can be priced lower.
  3. Internalising an externality, by assigning a cost to the party that causes it instead of spreading it across other consumers.
  4. Peak-load pricing, under which users pay a higher rate for consumption at times of highest system demand.
Reveal answer & explanation

Answer: C

  • A single very large load can force a network extension whose cost would otherwise be recovered from all consumers on that licence. Charging it to the developer makes the decision-maker face the full cost of its own choice, which is what internalising an externality means.
  • Option A describes a pricing rule for the marginal unit of electricity, not for who funds a specific capital asset.
  • Option B is a genuine feature of Indian power tariffs, but it deliberately shifts cost between consumer classes - the opposite of what this rule does.
  • Option D is a real instrument and the framework does touch demand response, but peak-load pricing addresses when power is used, not who pays for building the connection.