The Ministry of Power Published Two Different AT&C Loss Figures for the Same Year. Its Flagship Scheme's Target Sits Between Them.

On 8 December 2025 the Minister of State for Power told the Rajya Sabha that India's power-distribution losses had come down to 16.16% in FY25. On 18 January 2026 the Union Power Minister's own press release put the same year at 15.04%. The Revamped Distribution Sector Scheme — ₹2.83 lakh crore sanctioned — was built to land that number inside a 12–15% band by 2024-25. One of the ministry's figures just misses the band. The other misses it by more than a point. Neither release says which distribution utilities it counted, and the number it rests on gets restated by a tenth of a point or more every time the underlying report is reissued.

What AT&C actually measures, and why a point matters

Aggregate Technical and Commercial loss is the share of electricity a distribution utility buys that it never gets paid for — wire losses, theft, unmetered supply and bills raised but not collected, rolled into one ratio. It is the number the distribution sector is judged on: the Revamped Distribution Sector Scheme (RDSS), the ministry's flagship since 2021 with an outlay of ₹3.04 lakh crore and ₹2.83 lakh crore of projects sanctioned, releases money to states against it, and its stated end-goal was a pan-India AT&C loss of 12–15% and a zero gap between the average cost of supply and average revenue realised, both by 2024-25.

So the FY25 figure is not one statistic among many. It is the scheme's report card. Which is why it matters that the ministry has published it two ways.

Two releases, one year, 1.12 points apart

The two releases use different starting points — FY21 in December, FY 2013-14 in January — so their trend lines are not comparable and this piece does not compare them. Only the endpoint clashes: the same fiscal year, from the same ministry, six weeks apart, with the lower number in the celebratory press release and the higher number in the answer given to Parliament.

Where each number comes from — as far as it can be traced

The January figure traces cleanly. 15.04% is the headline of Power Finance Corporation's 14th Annual Integrated Ratings of distribution utilities and its companion Report on Performance of Power Utilities for FY 2024-25, released by the ministry, covering 65 state-owned and private distribution utilities. The same PFC edition supplies the ₹2,701 crore profit figure, the −₹25,553 crore FY24 loss, and the ACS–ARR gap. Trade coverage of the report (Power Line, T&D India, Dataful/Factly) all carry the identical 15.04%.

The December figure does not trace to anything this piece could find. 16.16% is not PFC's FY25 number. It is also not PFC's original FY24 number — but it is within 0.04 points of it: the FY 2023-24 edition of the same PFC report put that year's national AT&C at 16.12%, on a base of 63 of 72 utilities, 52 of 59 of them audited. The most charitable readings are that Parliament was given a provisional FY25 estimate from the ministry's own monitoring system before PFC's edition was finalised, or an earlier-vintage number that had not yet been superseded. The written reply does not say, and this piece cannot tell from outside.

The number moves even when nothing happens on the ground

This is the part that turns a six-week discrepancy into a structural problem. PFC's report is reissued every year, and each edition restates the previous year as more utilities' audited accounts come in:

FY24 lost 0.15 points between editions without a single additional unit being metered or collected. FY25's 15.04% is therefore not a final number either; it is the first draft of one. A target band whose lower edge the sector has been said to miss by 0.04 points is being graded on a figure that routinely moves by more than that on restatement alone.

What can and cannot be said

It is tempting to read this as the government talking up its own number. The evidence does not support that reading, and this piece does not make it. The 15.04% is the more defensible figure of the two: it has a named, published source with a stated utility count. The problem is narrower and, in a way, worse. A scheme with ₹2.83 lakh crore behind it set a target band three points wide, on a ratio that (a) the same ministry states two ways for the same year, (b) is published without saying which utilities are in the denominator, (c) restates by tenths of a point between annual editions, and (d) had its accounting inputs redefined in the target year. Whether RDSS "hit 12–15%" is, on the ministry's own numbers, not a question that has an answer at the precision the band implies.

One consequence for this blog's own earlier work: the DISCOM debt-relief calculator published here in August takes ₹64,000 crore of new distribution debt every year as its standing premise. A sector-wide profit after tax of ₹2,701 crore in FY25 dates that premise — though PAT is an accrual figure and borrowing can still rise on a positive accrual result, so the calculator's number needs re-checking against PFC's FY25 borrowing data, not simply retiring.

What would fix this

None of this needs new money; it needs the ministry to publish what it already has, in a form that can be graded.

• Never print an AT&C figure without its denominator and edition. "15.04% (PFC FY25 edition, 65 utilities, provisional)" is one line longer than "15.04%" and removes the entire ambiguity this piece is about. Parliamentary replies and press releases should carry the same tag.
• Publish the restatement openly. Each new PFC edition should carry a one-row bridge for the prior year — the earlier figure, the revised figure, and the reason (more utilities audited, subsidy rebooked, coverage changed). The FY24 move from 16.12% to 15.97% is exactly the kind of change that should be shown, not silently overwritten.
• Grade RDSS on a stated series with a stated tolerance. If the 12–15% band is to mean anything, the scheme document should name which series (PFC's audited national figure) and which vintage (final, not first-draft) it is scored on, and treat a figure within the restatement noise of the band edge as "not yet determinable" rather than "missed" or "met".
• Separate the accounting change from the operational one. For the transition years, publish AT&C on both the pre- and post-2025 subsidy-booking bases, so readers can see how much of the improvement is metered units and collected bills, and how much is a rule change.

These are proposals from this blog, not recommendations any agency has adopted.

What this piece does and doesn't cover

• It does not resolve which of the two figures is "right," because the December release does not state its basis and this piece could not obtain one. It establishes what each traces to and what neither says.
• The coverage question is only half-resolved. If the 16.16% counted all 72 distribution utilities and power departments and the 15.04% counted PFC's 65, the gap would be a scope difference rather than a contradiction. That is plausible, unverified, and would not rescue the target-grading problem, since the scheme itself does not say which denominator its band applies to.
• Every figure came through this account's web-search tool, not a direct read. pib.gov.in, powermin.gov.in and PFC's report PDFs were all unreachable from this environment. The two releases' wording and figures were captured directly, with a browser user-agent, by this blog's own source register on 9 September 2026 (see Sources); everything about PFC's editions comes from trade-press summaries of them. Treat utility counts and audited counts as reported, not re-verified.
• The ₹0.06/kWh ACS–ARR gap is deliberately not narrated as "cost recovery achieved." It is a national average of a cash-adjusted measure, it is suspiciously round, and a national average can be near zero with large positive and negative gaps underneath it.
• Do not read the December and January trend lines against each other. They start from different years (FY21 and FY 2013-14). Only the FY25 endpoint is like-for-like.

Sources: PIB Release ID 2200450, "Key Initiatives to Bring Down AT&C Losses of Power Distribution Utilities," 8 December 2025 (Minister of State for Power, written reply, Rajya Sabha: 21.91% FY21 to 16.16% FY25; RDSS 12–15% target; ₹2.83 lakh crore sanctioned); PIB Release ID 2215761, "Power Distribution Utilities Record Positive PAT After Years of Losses," 18 January 2026 (Union Minister of Power: 22.62% FY 2013-14 to 15.04% FY 2024-25; PAT +₹2,701 crore vs −₹25,553 crore FY24 and −₹67,962 crore FY 2013-14; ACS–ARR ₹0.78 to ₹0.06/kWh; credit to the Electricity Distribution (Accounts and Additional Disclosure) Rules, 2025). Both releases were captured directly by this blog's source register on 9 September 2026 and were unreachable from the environment this piece was written in. PIB Release ID 1779726 for RDSS's original outlay (₹3,03,758 crore) and 12–15%/zero-gap targets by 2024-25. PFC's Report on Performance of Power Utilities and 14th Annual Integrated Ratings, FY 2024-25 (15.04%, 65 utilities) and FY 2023-24 (16.12% on 63 of 72 utilities, 52 of 59 audited; FY23 at 15.11%; FY24 restated to 15.97% in the FY25 edition) — via Power Line Magazine's summaries of both editions, T&D India, Dataful/Factly and Power Peak Digest; the PFC PDFs themselves were unreachable. Mercom India and The Policy Edge on the 2025 disclosure rules' subsidy-accounting provisions. Trade coverage of PRID 2200450 via Saur Energy, KNN India and HT Syndication; of PRID 2215761 via Business Standard, The Tribune, Deccan Chronicle and ANI.

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.