On 27 August, summing up the first half of the year at a government meeting, Mishustin said:
“Thanks to these figures, budget revenue is also growing. It has exceeded 18 trillion 640 billion roubles. That is substantially above the forecast. What matters is that its structure continues to change: non-oil-and-gas receipts already account for more than 80% of the total. Our economy is undergoing serious diversification.”
Where the share comes from
The figures come from Minfin's own preliminary estimate: first-half revenue was 18,622 bn roubles (+5.8% y/y), of which oil and gas provided 3,661 bn (−22.7%) and non-oil sources 14,961 bn (+16.3%). The non-oil share was indeed 80.3%; a year earlier it was 73.1%.
But a share is a fraction, and either term can change it. Break down the +7.2 percentage-point shift. Had non-oil revenue stayed flat, the fall in oil and gas revenue alone would have added +4.8 pp. Had oil and gas revenue not fallen, non-oil growth would have added +2.9 pp. Together these add up to more than 7.2 because the two effects interact, but the proportion is clear: two thirds of the “diversification” comes from the denominator collapsing, not the numerator growing. In the limiting case, oil and gas revenue falling to zero would push the share to 100% — and calling that diversification would be a stretch.
The control case: July
For January–July, the share was 79.2% — below the “80% mark”. What happened? Oil and gas revenue rebounded to 934 bn in July, and the “diversification” went into reverse. A metric that deteriorates in a good month for oil is measuring the oil price, not structural policy.
“Substantially above the forecast” is also unclear: the full-year revenue plan is 40.28 trn, half of which is 20.1 trn, and the actual 18.64 trn falls short of that. The transcript does not say what exactly Mishustin was comparing it with.
How to measure it honestly
We have seen this trap twice before. Oil & gas budget revenues rose +18.6% y/y in July — while standing at 57% of the pre-war level and scoring 14 out of 100, the second worst of the eight. The rate looked like a recovery; the level remained near its low. The share creates the same trap with a different fraction.
That is exactly why our indicator measures the real level against 2021: neither the rate trick nor the share trick can move it. Its denominator is fixed at the pre-war year, and no collapse in another revenue line can “improve” it. The other indicators are on the monitor.
What we do not know
Mishustin says 18.64 trn; Minfin's preliminary estimate says 18.622. The gap most likely means the government was looking at a fuller report, but we could not find a published copy. Nor is there a breakdown of non-oil revenue by source, so these figures cannot separate the contribution of the VAT rise to 22% from 1 January from real economic growth.