The Central Power Purchasing Agency (Guarantee) has filed its projection of what the ex-WAPDA distribution companies will pay for electricity in calendar year 2027. NEPRA will hear it on 3 November 2026, and the result feeds the rebasing of consumer-end tariffs from 1 January 2027.
The base case is Rs27.309/kWh on sales of 130,081 GWh, for a total of Rs3.552 trillion. Six scenarios run from Rs25.883/kWh (the "ideal" case) to Rs29.255/kWh (the second worst case). The approved 2026 reference was Rs25.32/kWh on a total of Rs3.186 trillion, so the base case is 7.9% higher per unit and about 11.5% higher in total.
Those headline moves are the part most coverage stopped at. The more useful material is inside the scenario table, because it shows how little of the bill actually moves with the things the scenarios vary.
What the bill is made of
In the base case the components are:
- Capacity charges: Rs1.914 trillion, or Rs14.717/kWh
- Fuel: Rs1.319 trillion, or Rs10.141/kWh
- Variable O&M: Rs70.6 billion, or Rs0.543/kWh
- Transmission and market charges: Rs248.3 billion, or Rs1.909/kWh
Capacity charges are 53.9% of the total (1,914 ÷ 3,552). Fuel is 37.1%. The remaining 9% is O&M, transmission and the market operator.
Capacity charges are fixed payments to generators for being available, set by their contracts and indexed to the exchange rate, US inflation, KIBOR and SOFR. They do not fall when a plant runs less. That is the structural fact the rest of this note works from.
The scenarios move fuel, not much else
Across all six scenarios the capacity line takes only two values. It is Rs1.914 trillion in the base, ideal and RLNG-mix cases, and Rs1.962 trillion in both worst cases, where the rupee is assumed at Rs290 per dollar instead of Rs279. The gap is Rs48 billion, or 2.5%.
Fuel, by contrast, ranges from Rs8.714/kWh in the ideal case to Rs11.712/kWh in worst case II. That Rs3.0/kWh spread accounts for almost all of the Rs3.37/kWh gap between the cheapest and most expensive scenarios (29.255 − 25.883). The difference between a good and bad 2027 for the power purchase price is mostly a fuel-and-currency story. The contracted fixed cost is close to locked in whatever happens.
Our calculation: what a lost kWh is worth
The demand-sensitivity scenario is the one worth reading closely. It assumes demand growth of 0.7% instead of 2.6%, which cuts sales to DISCOs from 130,081 GWh to 127,937 GWh. That is 2,144 GWh lower, a drop of 1.65%.
Total cost in that case is Rs3,527.03 billion against Rs3,552.44 billion in the base. So the bill falls by Rs25.41 billion, or 0.72%.
Dividing one by the other: each kWh that is not sold reduces the total power purchase cost by about Rs11.85 (25.41 billion ÷ 2,144 million kWh). The average price of that kWh in the base case is Rs27.31. The other Rs15.46 or so does not disappear. It is re-spread across the kWh that are still sold, which is why the per-unit price in the low-demand case rises to Rs27.568, up Rs0.259 on the base.
Method: we used only the two totals and two sales figures in CPPA-G's scenario table as published in NEPRA's hearing notice and reported by Business Recorder and Mettis Global. We did not adjust for timing or for the small change in capacity per kWh (14.717 to 14.960) that the same scenario shows. The result is an average, not a marginal cost for any particular hour.
The Rs11.85 is close to the base-case fuel plus variable O&M figure (10.141 + 0.543 = Rs10.68/kWh), plus a little transmission. That is what one would expect if almost nothing fixed changes when sales fall.
Why this matters for net metering, captive power and grid exit
The ratio is a rough measure of how expensive grid defection is for the customers who stay. When an industrial user moves to captive gas or a commercial user adds rooftop solar and stops buying, the system saves roughly 43% of the average price (11.85 ÷ 27.31) and passes the remaining 57% to everyone else.
That arithmetic sits behind several current policy fights. The government's moves to rework net metering buyback rates, the levy on captive gas and the push for an Incremental Consumption Package to lift winter demand all rest on some version of it. The CPPA-G table now gives a number that can be checked against each proposal.
It cuts both ways. Demand that the system can add cheaply is worth about the same Rs15 or so per kWh in fixed-cost relief. The demand case is only an illustrative 0.7% scenario, and CPPA-G calls the whole set indicative.
How 2027 compares with 2026
Two figures from the 2026 determination are relevant. NEPRA approved Rs3.186 trillion at Rs25.32/kWh, which implies sales of about 125,830 GWh. The 2027 base-case sales figure of 130,081 GWh is 3.4% above that, which is higher than the 2.6% demand growth the base case assumes. Part of the gap may come from a different starting point or loss assumption, and the filing will need to explain it at the hearing.
Capacity charges for 2026 were approved at Rs1.923 trillion, or Rs14.86/kWh. The 2027 base case is slightly lower in total at Rs1.914 trillion and lower per unit at Rs14.717. The per-unit fall comes from more sales, not from a smaller fixed bill.
Monthly fuel adjustments as a check on the fuel line
The fuel component is the part of the 2027 projection that will be tested each month through fuel charges adjustments. The August 2026 adjustment, notified on 7 October, gives a recent reading. CPPA-G claimed an actual fuel cost of Rs8.8265/kWh against a reference of Rs7.0998, an adjustment of Rs1.7267. NEPRA allowed Rs8.2084, an adjustment of Rs1.1086. The regulator cut 35.8% of the claimed adjustment.
The 2027 base case puts fuel at Rs10.141/kWh as an annual average. The 2026 August actual, at Rs8.21, was in a high-demand month with hydro available. Winter months with more RLNG and furnace oil would run higher. Whether Rs10.14 is conservative depends heavily on LNG pricing, which has been volatile since the Middle East escalation earlier this year.
What to watch at the 3 November hearing
- Whether NEPRA accepts the 2.6% demand growth base, given that recent months have been uneven.
- The exchange-rate assumption of Rs279 per dollar, since every rupee of depreciation lifts the indexed capacity line.
- The RLNG-mix scenario (Rs26.441/kWh), which implies about Rs0.87/kWh of savings from a cheaper gas mix and is the only scenario that lowers fuel without assuming an ideal year.
- How the sales base is reconciled with the 2026 determination.
The filing also includes a separate petition for the CPPA-G fee for FY2026-27, admitted on 11 October.