The Lift Line

It is easier to argue about the water a server drinks than about the water a sugarcane field drinks, and only one of them is ninety per cent of the problem.

Why This Editorial Matters for Your Exam

This is a model GS3 argument about misallocated attention, and the analytical move it makes, checking a claim against the distribution of the underlying quantity, is transferable to energy, emissions, subsidies and land. It also gives you the virtual water concept in a live Indian context rather than as a definition.

GS Paper 3: Conservation of water resources; cropping patterns and irrigation; infrastructure and industry; environmental degradation.

Concept Meaning Why it is testable
Virtual water Water embedded in the production of a traded commodity How a water-scarce region exports water without moving any
Water withdrawal against consumption Water taken from a source against water not returned to it Why cooling and irrigation are not comparable per litre
Volumetric pricing Charging by quantity used rather than a flat or nil rate The instrument the column proposes

Background and Context

The scarcity, in numbers. India’s per capita water availability has fallen from about 1,816 cubic metres in 2001 to an estimated 1,486 cubic metres in 2021, with projections of continued decline. The conventional thresholds are 1,700 cubic metres for water stress and 1,000 cubic metres for water scarcity, which places India in the stressed band and moving towards the scarce one.

The distribution of use. Agriculture takes roughly 90 per cent of India’s freshwater withdrawals. Industry and energy take most of the balance, and domestic use a small share.

Why the column exists. Its opening observation is that when data-centre water use starts appearing in the cartoons of financial dailies and the President of the United States stakes political capital on it, “an engineering footnote has become a public policy dispute”. The Indian debate is an import of an American one.

The new user. Data centre construction has expanded rapidly, and such facilities use water for cooling. WRI India has found that more than half of India’s data centres sit in water-stressed regions, with about three-quarters of facilities concentrated in five States: Maharashtra, Tamil Nadu, Karnataka, Telangana and Uttar Pradesh. That finding is what gives the counter-argument its force, and it should not be dismissed.

The wider pattern. Water-intensive activity, including ethanol production, manufacturing and data centres, is increasingly concentrated in drought-prone states, which is a misalignment between industrial policy and hydrology rather than a failure of any single industry.

The Analysis

1. The arithmetic settles the first question. If one sector takes nine-tenths of the resource, no efficiency programme confined to the other tenth can resolve scarcity. This is not an argument that data centres should be careless. It is an argument that they cannot be the solution because they are not the magnitude.

2. Agriculture’s problem is the incentive structure, not farming. Assured procurement for particular crops, electricity supplied free or nearly free for groundwater pumping, and canal water priced far below the cost of delivery combine to make water-intensive crops rational for the individual farmer in regions whose hydrology cannot support them. Paddy in the north-west and sugarcane in the semi-arid Deccan are the standing examples. The farmer is responding correctly to the signals given.

3. Virtual water makes the misallocation invisible. A tonne of rice grown in a water-scarce state and sold elsewhere exports the water used to grow it in embedded form. No water moves, so no one accounts for it. India is a substantial net exporter of virtual water through commodities such as rice, sugar, cotton and livestock products, which means a water-scarce country subsidises the water consumption of its trading partners.

4. Withdrawal is not consumption, and this is where the data centre argument is genuinely weak. Much cooling water is withdrawn and returned, sometimes warmer, whereas irrigation water is very largely consumed through evapotranspiration. Comparing the two by withdrawal volume flatters agriculture and penalises industry. A serious accounting framework distinguishes them, and India’s does not.

5. The local objection survives all of the above. A national share of a fraction of a per cent tells you nothing about what a single large facility does to one district’s aquifer. Scarcity is a basin-level phenomenon, not a national one. Both propositions are true at once: data centres are irrelevant to national water stress and can be decisive locally. An answer that holds both is a better answer than one that picks a side.

Data and Institutions Vault

Prelims-grade facts:

The distribution and the stress:

  • Agriculture accounts for roughly 90 per cent of India’s freshwater withdrawals.
  • Per capita water availability fell from about 1,816 cubic metres in 2001 to about 1,486 in 2021.
  • Below 1,700 cubic metres per person is conventionally water stress.
  • Below 1,000 cubic metres per person is conventionally water scarcity.
  • WRI India finds more than half of India’s data centres are located in water-stressed regions.
  • About three-quarters of India’s data centres are in Maharashtra, Tamil Nadu, Karnataka, Telangana and Uttar Pradesh.
  • Water-intensive sectors including ethanol and data centres cluster in drought-prone states.

The concepts:

  • Virtual water is the water embedded in producing a traded commodity.
  • Withdrawal is water taken from a source; consumption is water not returned to it.
  • Cooling water is largely withdrawn and returned; irrigation water is largely consumed.
  • India is a substantial net exporter of virtual water through farm commodities.

The institutional frame:

  • Water is Entry 17 of the State List, subject to Entry 56 of the Union List.
  • The Central Ground Water Board functions under the Ministry of Jal Shakti.
  • The Central Ground Water Authority regulates groundwater extraction in notified areas.
  • The National Water Mission is one of the missions under the National Action Plan on Climate Change.
  • Atal Bhujal Yojana is the scheme for participatory groundwater management.
  • The Composite Water Management Index was published by NITI Aayog.

⚠️ Watch the trap: Do not compare sectoral water use by withdrawal alone. A thermal plant or a data centre withdraws a large volume and returns much of it; irrigation consumes most of what it takes through evapotranspiration. Two figures with the same unit can describe entirely different impacts on a basin, and using them interchangeably is the commonest error in water answers.

The Debate

FOR (the focus on data centres is misdirected): Agriculture takes nine-tenths of withdrawals, cropping incentives push water-intensive crops into semi-arid regions, much cooling water is returned rather than consumed, and no framework exists to compare uses across sectors.

AGAINST (new users still have to answer for themselves): Data centre demand is new, concentrated and often located in already stressed districts, so its local marginal impact can be severe. Agriculture’s dominance is a reason to reform agriculture, not a licence for any new user to add unpriced demand.

Balanced verdict: The disagreement is about scale of analysis, not about facts. Nationally the column is right; locally the objection is right. The framework that resolves it is the one the column proposes, applied to everyone: measure, price, disclose and recycle. If a data centre must publish its water balance, so must a sugar mill and, in aggregate, an irrigation command area. Only then does the question of who the villain is become answerable rather than rhetorical.

How to Think About This

Before accepting that any sector is the cause of a resource problem, ask for the distribution of use. Public argument gravitates to the most visible user rather than to the largest one, because visibility, not magnitude, drives attention. This is a general failure mode and it recurs across debates on emissions, on electricity subsidy, on food waste and on plastic. The corrective is boring and reliable: get the denominator first.

Diagram-in-Words

National scale: the column is right Agriculture, about 90 per cent industry, energy and domestic share the rest Basin scale: the objection is right one facility, one stressed district over half of data centres sit in stressed regions a national average conceals a local aquifer Water is unmetered and unpriced, so uses cannot be compared Measure Price Disclose Recycle
Both scales are correct at once, which is why the argument never resolves. The missing common unit, not the villain, is the actual policy gap.

Takeaway Box

Lift line: It is easier to argue about the water a server drinks than about the water a sugarcane field drinks, and only one of them is ninety per cent of the problem.

Prelims hooks: Agriculture at about 90 per cent of withdrawals; per capita availability 1,816 cubic metres in 2001 to about 1,486 in 2021; stress at 1,700 and scarcity at 1,000 cubic metres; virtual water; Central Ground Water Board and Authority; Atal Bhujal Yojana; NITI Aayog’s Composite Water Management Index.

Mains keywords: virtual water, withdrawal against consumption, volumetric pricing, basin-level stress, incentive misalignment.

Ethics and interview angle: Pricing water would allocate it efficiently and would fall hardest on those least able to pay. What sequencing makes the reform defensible?

PYQ linkage: Connects to past UPSC Mains questions on water resource management, on cropping pattern distortions caused by subsidies, and on industrial siting in ecologically stressed regions.

Sources: Business Standard

Source: Data Centres Are Not the Villain: India's Water Argument Is Aimed at the Wrong Sector — Ujiyari.com | Free UPSC & State PCS Editorial Analysis