NEPRA Grants Conditional Approval to ISP 2026-35: USD 900m Battery Storage Stripped, Transmission Trimmed, Legal Warning Issued to ISMO

Pakistan's National Electric Power Regulatory Authority (NEPRA) has issued a partly-concessional determination approving the Integrated System Plan 2026-35 (ISP) submitted by the Independent System and Market Operator (ISMO). On the headline alone, it reads like a firm nod. Read into the decision, and the picture changes completely: USD 900 million of battery energy storage (BESS) stripped out of the approved scope, a K-Electric transmission line removed, the transmission-investment envelope cut from USD 10.64 billion to USD 9.18 billion, and ISMO placed on notice that it could face legal proceedings if data discrepancies persist. A nod with three cuts and one threat attached — that is the real substance of the ruling.

Context: a national plan balanced on a single number

Pakistan runs its electricity system through a clear hierarchy. NEPRA is the regulator and holds the power to approve or reject. ISMO is the licensed system operator responsible for running the grid and preparing expansion plans. Below sit the distribution companies (DISCOs) and K-Electric, which must execute investment in line with the approved plan. Above the whole chain, the Council of Common Interests (CCI) functions as a ratification forum for assumptions tied to the "committed project" category.

The ISP 2026-35 is a long-horizon planning document built on the Indicative Generation Capacity Expansion Plan (IGCEP) and the Transmission System Expansion Plan (TSEP). Ultimately, the entire document rests on one foundational variable: the ten-year electricity demand forecast. Every capacity decision, every cost decision, every investment-priority decision flows from it.

And that foundational variable is precisely what is in dispute.

Core: the cost-optimisation engine still runs, but the inputs are being challenged

Technically, this is a standard least-cost optimisation exercise under normal IGCEP practice, not a methodological breakthrough. The programme's present-value cost is fixed at USD 47.08 billion for FY2025-35, marginally below the USD 47.13 billion recorded before the JCM hybrid was added.

The notable point lies elsewhere. NEPRA does not attack the novelty of the method; it attacks the quality of the data: demand forecasts, verifiability of figures, load-growth assumptions. The regulator directs ISMO to overhaul its methodology and rejects ISMO's request to disclaim responsibility for data it published itself. That is a significant governance tightening: a planner cannot file numbers and then declare itself not answerable for them.

One methodological inconsistency stands out most clearly around storage. The USD 900 million BESS project was marked "committed" across scenarios — meaning it implicitly existed in every calculation — yet was excluded from the optimisation exercise itself. Placing a USD 900 million item outside cost competition and calling it committed is a structure NEPRA identified and rejected. The result: BESS is excluded from the approval pending a fuller cost-benefit study.

By contrast, the Riali-II hydropower project is directed to be treated as committed on the basis of roughly 90% physical progress. That is sunk-cost logic — the money is already spent, and stopping would cost more than continuing.

The JCM hybrid (269 MW, comprising 95 MW solar and 174 MW wind) is the only component genuinely run through optimisation and revised on updated technical and financial parameters. Its inclusion lowered optimised wind capacity and slightly reduced total cost. This proves the optimisation engine works — but only as well as the data fed into it.

On transmission, the TSEP was trimmed by about USD 1.46 billion, from USD 10.64 billion to USD 9.18 billion. Of that, ongoing and already-planned works account for USD 4.30 billion and newly proposed works for USD 4.88 billion. The cut stems largely from a reduced demand forecast, which in turn pulled down required transmission capacity.

One detail is buried fairly deep: dropped projects are placed in "abeyance" rather than formally cancelled. That status creates off-balance-sheet limbo, politically useful for avoiding the write-down of sensitive projects.

Contrarian angle: the real worry is not the USD 900 million that was cut

The USD 900 million exclusion is the single largest financial intervention, but it is not the biggest risk. The cost reduction from USD 47.13 billion to USD 47.08 billion is roughly 0.1% — essentially cosmetic next to the two exclusions above it.

The underlying risk sits elsewhere: the plan still adds around 26 GW of new capacity to a system whose load demand is trending down. Rooftop solar, captive generation and behind-the-meter storage are eroding grid demand in ways the planning model does not sufficiently capture. When supply rises while baseline demand falls, the familiar consequence is overcapacity and fixed-cost burdens — capacity payments — landing on end-user bills.

NEPRA itself states in the document that costs will ultimately be borne by consumers. That is a systemic warning, not a passing remark.

One thing I must challenge in my own reading: inferences about political pressure around K-Electric, or about the depth of friction between ISMO and NEPRA, are largely speculative. The regulator's language — "unheard of and uncalled for" — signals frustration deeper than a routine determination, but it is not enough to conclude there is a hidden agenda. And the legal warning may simply be deterrent signalling to raise data quality in future submissions rather than a case being prepared.

Takeaway

This is a conditional, contested approval, not a clean bill of passage. Its greatest value lies in the governance precedent: a planner cannot disclaim responsibility for its own data, and the "committed project" category cannot be used to quietly remove old commitments through changed assumptions. If ISMO revises properly, BESS proves its case, and DISCOs align their investment plans, the whole value chain could be re-sized to a lower-demand reality — and that could be the best thing to happen to the power sector in years. If not, today's nod will be a technical liability that Pakistani consumers pay off for the next decade.