Reading the monthly accounts

The figures are published monthly, free, with every identity intact. This chapter is about the four ways they are routinely misread, using the September 2026 release as the case.

The release gives you three columns, and most readers use one

Every line in the CGA table carries:

The Budget Estimate for the full year — the plan.

The actual up to the month — what happened.

Two percentages: the actual as a share of the estimate, and the same share at the corresponding period of the previous year, which the document labels COPPY and prints in brackets.

The third column is the one that does the analytical work and the one that is most often dropped. Without it, a percentage has nothing to be compared against except intuition, and intuition about government accounting seasonality is usually wrong.

Misreading one: ignoring pro-rata

September is the sixth month of a year that runs April to March. Half the year has passed, so a line running exactly to plan would read 50%.

Now reread the headline figure. The fiscal deficit was at 42.1% of its Budget Estimate.

That is 7.9 points below the pace required to hit the full-year target. A deficit at 42.1% at the halfway point is not a deficit in trouble; it is a deficit running under its annual allowance.

It was widely reported as "fiscal deficit hits 42.1% of full-year target", against 36.5% a year earlier, and framed as deterioration. Both figures are correct and the framing drops the only benchmark that makes either meaningful. The year-on-year move is real — the deficit is less front-loaded in the government's favour than it was — but "42.1% at the halfway point" on its own is a statement that the target is being met, not missed.

Check the two sides to see why. Total receipts were at 50.3% of estimate, marginally ahead of pace, and total expenditure at 47.7%, behind it. Receipts slightly ahead and spending slightly behind is, arithmetically, a deficit below pace. The headline percentage and its two components tell a consistent story and only the components make it legible.

Misreading two: treating pro-rata as the whole answer

Having just used it, the opposite error: 50% is a benchmark, not an expectation. Government flows are seasonal and some lines should be nowhere near it.

Chapter 3 found net tax revenue at 44.5% because Indian tax receipts are back-loaded, with the final advance instalment in March. Chapter 4 found capital expenditure at 53.6% because construction avoids the monsoon, and petroleum subsidy at 10% because a price moved.

So the sequence is: compare against 50% to orient yourself, then compare against the COPPY column to control for the seasonality, then ask what is structurally different about the line. Three steps, and most commentary performs none of them.

The clearest case in the September release is non-tax revenue at 72.2% of estimate — apparently spectacular, until the COPPY column shows 79.9% at the same point last year. A line 22 points ahead of pro-rata is running behind its own prior-year pace. Pro-rata alone would have told you the opposite.

Misreading three: quoting the wrong deficit

Chapter 2's point, in its practical form. Coverage quotes the fiscal deficit almost exclusively, and the fiscal deficit says how much was borrowed and nothing about what it bought.

For September 2026 the three readings diverge sharply:

At end-September, % of Budget Estimate	2026	Last year
Fiscal deficit	42.1%	36.5%
Revenue deficit	23.4%	5.2%
Primary deficit	23.4%	−1.7%

The fiscal deficit moved 5.6 points. The revenue deficit moved 18.2 points, from 5.2% to 23.4%. The primary deficit went from a surplus to 23.4%.

The headline captured the smallest of the three movements. The sharpest deterioration is in the measure of how much borrowing funded current spending, and in the measure of what this year's decisions cost — and neither was reported.

Misreading four: ignoring the note

Beneath the table sits the sentence that should accompany every quotation of these figures:

Fiscal deficit figure shown in monthly accounts during a financial year is not necessarily an indicator of fiscal deficit for the year as it gets impacted by temporal mismatch between flow of non-debt receipts and expenditure up to that month on account of various transitional factors both on receipt and expenditure side, which may get substantially offset by the end of the financial year.

The Controller General of Accounts is saying, on the document, that the monthly deficit is not a forecast of the annual one and that the mismatches producing it are expected to reverse.

Three further pieces of small print on the same page, all load-bearing:

Actuals are unaudited provisional figures. They will be revised, and the audited position comes much later from a different office.

COPPY means corresponding period of the previous year — the comparison is like-for-like in months, not a calendar-year comparison.

One crore is ten million, which the release spells out because its figures are in crore and its readers are not all Indian.

A release that states its own limits in advance is behaving the way the CPI and GDP releases did in the two previous subjects. By this point the pattern is clear enough to state as a rule: Indian statistical releases are, as a class, more honest about their weaknesses than the commentary built on them.

The two-sentence version the accounts support

What the exercise asks you to write, for September 2026:

At the halfway point of 2026-27 the Union government's fiscal deficit was ₹7,14,413 crore, or 42.1% of its full-year estimate — below the 50% pro-rata pace, with receipts slightly ahead of plan and spending slightly behind. About a fifth of that borrowing funded current spending rather than capital, but the revenue and primary deficits have both deteriorated sharply against the same point last year, and interest alone now absorbs half of net tax revenue.

Two sentences, every figure from one free document, and a reader who had only the headline has none of it.

The point

The monthly accounts give a Budget Estimate, an actual, a percentage and a prior-year percentage, and the four common errors are ignoring pro-rata, treating pro-rata as the whole answer, quoting only the fiscal deficit, and skipping the note. At end-September 2026 the fiscal deficit at 42.1% was 7.9 points below the pace needed to meet its target, which is close to the opposite of how it was reported — while the revenue deficit's 18.2-point deterioration, the larger movement, went unmentioned. The release itself warns that a monthly deficit is not an indicator of the annual one.