⚠ 23 GW of clean energy was curtailed in 2025 — while coal ran at minimum loading.
⚠ Solar generators were paid ₹575–690 crore for energy they were prevented from delivering.
The grid is working exactly as the doctrine asks it to. The doctrine may no longer be defensible.
Curtailment & Duck Curve — why backing down clean MWh is an ethical choice
| Between May and November 2025, GRID-India curtailed approximately 23 GW of renewable energy — equivalent to roughly 18% of average monthly solar generation thrown away — while coal plants ran at minimum technical loading. This is the hidden carbon tax of operational conservatism. The 2025 numbers make it visible for the first time. |
|---|
01 · WHAT OPERATIONAL CONSERVATISM LOOKS LIKE IN 2026
23 GW of clean energy — curtailed, not delivered.
For decades, Indian grid operators have run on a defensible doctrine: when in doubt, protect the system. Renewable generation is intermittent; the grid is fragile; the safer dispatch decision is to keep dispatchable thermal plants running at minimum loading rather than ramp them down.
The 2025 data quantifies the cost. GRID-India curtailed approximately 23 GW of renewable energy between May and November 2025. Ember’s analysis puts the solar-specific number at 2.3 TWh curtailed between May and December 2025 — roughly four weeks of all-India solar generation thrown away.
The curtailment is geographically concentrated. RE penetration in Karnataka has touched 40%, Andhra Pradesh 22%, Tamil Nadu 21%. Gujarat’s Khavda solar parks recorded the highest individual-site curtailment levels in 2025. The states with the highest RE share are also the states bearing the highest curtailment burden.
02 · THE CURTAILMENT PARADOX
Three coupled facts that make curtailment an ethical indictment.
The ethical case against curtailment turns on three coupled facts:
▸ Clean MWh discarded. The 2.3 TWh of solar curtailed in 2025 was technically available, contractually committed, and economically cheaper than any alternative. It was discarded to protect the operating doctrine.
▸ Dirty MWh delivered in its place. Ember calculates the curtailed solar could have avoided approximately 2.1 million tonnes of CO₂ if it had displaced coal generation — roughly the annual emissions of 0.9 million Indian households.
▸ Compensation paid for non-delivery. Solar generators received an estimated ₹575–690 crore through emergency TRAS mechanisms for energy they were prevented from delivering. Consumers pay for the curtailment twice: once in the wasted clean capacity, once in the compensation bill.
Wind curtailment has a structurally different mechanic. Where solar curtailment is a midday-surplus problem driven by the Duck Curve, wind curtailment in 2025–26 was driven primarily by southwest-monsoon transmission congestion — excess generation coinciding with constrained evacuation corridors. Solar-wind hybrid projects with intra-day BESS smoothing manage this through anti-correlation, which is why the Part 09 hybrid price of ₹3.80/kWh depends on this logic working in practice.
03 · THE DUCK CURVE AND THE SUNSET SPIKE
Curtailment is the daytime symptom. The evening ramp is the other half.
India’s net load — total demand minus renewable generation — drops sharply at midday as solar overwhelms demand, then rises sharply at sunset as solar disappears and household demand peaks. The all-India evening ramp requires dispatchable capacity to spin up in under two hours.
This is the Duck Curve. The technical question is how to fill the belly with stored energy from the midday surplus. The operational question — what to do tonight, with the peakers available — is where the hidden carbon tax is actually levied. Gas plants that cost ₹11–15/kWh fill the ramp that curtailed solar could have stored at ₹4.64/kWh.
04 · PEAKER PLANTS AND THE TRUE COST OF STABILITY
The cost stack from cheapest to most expensive.
India operates three categories of peaking capacity. Their cost structures define where the carbon tax actually sits:
▸ Gas-based peakers (NTPC Anta, PPCL Pragati-III Bawana): marginal peak cost ₹11.25–₹14.75/kWh when running on spot RLNG. Plus cycling penalties — auxiliary consumption rises from 3% to 5% on hot starts; part-load heat rates increase 10–15%.
▸ HP-DAM (High-Price Day-Ahead Market) — introduced March 2023. Emergency procurement channel; clearing prices up to ₹20/kWh in peak hours.
▸ FDRE (Firm and Dispatchable Renewable Energy) — renewable generation co-located with storage, configured for fast ramp. NTPC’s most recent tender discovered ₹4.64/kWh (ABC Cleantech). The clean peaking alternative.
Gas Asset Comparison — CERC 2024–29 Tariff Framework
| Parameter | NTPC Anta | PPCL Pragati-III Bawana |
|---|---|---|
| Installed Capacity | 419 MW | 1,371 MW |
| Annual Fixed Charge | ₹235–₹260 crore | ₹1,150–₹1,280 crore |
| Design Heat Rate | 2,450–2,500 kCal/kWh | 1,850–1,950 kCal/kWh |
| Base ECR (APM/Mixed Gas) | ₹4.35/kWh | ₹4.85/kWh |
| Peak ECR (Spot RLNG) | ₹11.50–₹14.00/kWh | ₹10.50–₹12.00/kWh |
| CERC Peak Incentive (6–10 PM) | ₹0.75/kWh | ₹0.75/kWh |
| Total Marginal Peak Cost | ₹12.25–₹14.75/kWh | ₹11.25–₹12.75/kWh |
India’s Power Procurement Cost Stack — 2026 Grid Economics
| Source / Market Segment | Cost Range (₹/kWh) | Grid Profile |
|---|---|---|
| Utility-Scale Solar | ₹2.60–₹2.80 | Intermittent daytime base |
| Pithead Coal | ₹3.50–₹4.20 | Conventional baseload |
| FDRE (Firm & Dispatchable RE) | ₹4.64 | Clean baseload alternative |
| Gas Peakers | ₹11.00–₹15.00 | Fast-ramping balancing cushion |
| HP-DAM (High-Price Day-Ahead) | Up to ₹20.00 | Emergency merchant clearing |
The hidden carbon tax of operational conservatism is the difference between what would clear if FDRE were the marginal peaking resource, and what actually clears today. At ₹11–15/kWh versus ₹4.64/kWh, that tax runs to thousands of crore per year.
05 · HISTORICAL CURTAILMENT AND THE 2030 PROJECTION
The 2025 headline sits on top of a decade of structural curtailment.
The 2.3 TWh solar figure is not an anomaly. It sits on top of years of routine wind curtailment that the data infrastructure was too weak to surface. Tamil Nadu’s wind belt has been losing roughly 15–25% of its annual output for over a decade. The 2025 solar headline is only marginally larger in absolute volume than the routine annual TN wind loss.
[DATA] Historical wind and solar curtailment — 2012 to Q1 2026 Two structural observations: wind curtailment is older, deeper, and more geographically concentrated than solar — Tamil Nadu's wind belt has been losing roughly 15–25% of annual output for over a decade. The geographic pattern of curtailment flipped in 2026 as transmission caught up in the south and Rajasthan/Gujarat solar ramped up. The 2030 trajectory — without BESS at scale — projects curtailment rising to 5–8 TWh annually. |
|---|
The geographic pattern of curtailment flipped in 2026 as transmission caught up in the south and Rajasthan/Gujarat solar came online at scale. The constraint is now the ISTS capacity in the north-west corridor — not a temporary problem, but a structural one.
06 · INSTITUTIONAL ACCOUNTABILITY — THE 55% MTL STORY
The single most important regulatory action of this decade.
The IEGC 2023 mandate requires thermal generators to operate down to 55% Minimum Technical Loading (MTL), down from the previous 70% floor. CERC’s projection: this single change creates 17–19 GW of additional evening flexibility — enough to absorb the Duck Curve ramp without gas peakers in most hours.
As of mid-2026, the rollout is fragmented:
▸ Central ISGS plants (NTPC, NHPC) have technically achieved 55% MTL capability — but many units maintain a 60–65% operational buffer in practice to avoid flame instability and tube damage.
▸ State Gencos are lagging. Older 210 MW units in West Bengal, Maharashtra, and Uttar Pradesh struggle with coal-quality variations that make 55% MTL risky without prohibitively expensive oil support. NTPC reported over 690 boiler tube leakages between 2021 and 2025 from frequent ramping.
▸ April 2026 — new DSM rules. CERC changed the Deviation Settlement Mechanism denominator from “Available Capacity” to “Scheduled Generation.” Staying high on MTL during solar peaks is now financially lethal for thermal generators. Market design is finally aligned with grid physics.
▸ June 2026 — 40% MTL deadline. The next step beyond 55% is already on the regulator’s clock for a select list of units. The 40% frontier is where the real evening-ramp flexibility lies.
07 · THE INTERGENERATIONAL BURDEN OF INACTION
₹145 billion in BESS support approved. Less than 1% commissioned.
The structural fix to curtailment is storage that absorbs midday solar surplus and dispatches it through the sunset ramp. That fix is sanctioned and funded:
▸ VGF approved: ₹91 billion (43.2 GWh, Tranche 1) + ₹54 billion (30 GWh, Tranche 2) = ₹145 billion in BESS viability gap funding.
▸ Pipeline end-2025: 92 GWh of BESS projects in development — up from 19 GWh in 2024.
▸ Actually commissioned end-2025: approximately 0.8 GWh. Just over 1% of the pipeline. The gap between sanctioned and commissioned is, in operational terms, why curtailment is still happening.
The IEA estimates India needs over 230 GWh of battery storage and 60,000+ km of additional transmission by 2030. The timeline gap — between when storage is needed and when it will be commissioned — is being paid for today in curtailed clean MWh and gas peaker tariffs. That cost is being levied disproportionately on the current generation of consumers and on communities that cannot afford backup power when the grid fails.
08 · THE VERDICT
Operational conservatism made sense for a thermal-dominant grid. Not for this one.
In 2026, with 50% non-fossil installed capacity and Khavda-class solar parks producing surplus midday power, the doctrine of “when in doubt, cut the renewables” is no longer a technical necessity. It is a choice — with a price tag of ₹575–690 crore in wasted compensation, 2.1 million tonnes of avoidable CO₂, and a growing structural curtailment trend.
The three legs of the fix are each partially underway: MTL flexibility (55% mandate, April/June 2026 DSM rules), storage deployment (₹145 billion VGF, 92 GWh pipeline), and dispatch protocol reform (FDRE tender pipeline). The question is not whether the policy framework exists. The question is which leg is most behind in your operating region — because that is where the carbon tax is being levied today.
| The 55% MTL framework is law. The 40% MTL deadline is June 2026. The VGF for BESS is sanctioned. The question for grid operators and RE developers in 2026 is which leg of the implementation triangle — flexibility, storage, or dispatch protocol — is most behind in your operating region. That is where the carbon tax is being levied today. |
|---|
→ · COMING UP IN PART 5
The Green Paradox — when going greener makes the grid less reliable. Part 5 introduces the Pavagada vs. Rewa contrast, examines how large-scale solar introduction can destabilise distribution networks, and prices the hidden costs of the transition onto the communities that can least afford them.
#EthicsofGridStability #RenewableEnergy #IndianRESector #GridOperations #Curtailment #DuckCurve #BESS #DISCOMReform #CarbonTax #EnergyJustice
Series Roadmap
Prelude:-----https://www.linkedin.com/feed/update/urn:li:activity:7465638847238922241/
Part-3----https://www.linkedin.com/pulse/ethics-load-shedding-equity-outages-who-stays-dark-mundoli-0s3dc/
Part-4----- https://www.linkedin.com/pulse/operational-conservatism-hidden-carbon-duck-curve-why-mundoli-w9npc/
Part-5 to 13 : coming soon