# Romania's GDP, quarter by quarter, 2020–2026: the first figure INS published and the current one

Romania's National Institute of Statistics (INS) publishes the first GDP figure 45 days after the quarter ends, updates it twice over the next two months, and the series is revised again after that. Follow how each of the 26 quarters changed, from the day of the first estimate to the current series.

- **0.1 pp**: median difference, in percentage points and without sign, between the first estimate and the second provisional version, at about 100 days (25 quarters, without Q2 2026, too recent)
- **0.3 pp**: median difference to the series of 7 September 2026, for 2023–Q1 2026: half the quarters changed less (Q1 2020–Q4 2022: 1.2 pp)
- **5 of 25**: quarters that went, between the first value and the current one, from growth to a fall or the reverse

> The figure of release day changes little in the first three months. Later revisions change it more, and in some quarters they turned growth into a fall or the reverse.

## The chart

### Forty-five days after the quarter, INS publishes the first figure

Each row is a quarter, from Q1 2020 to Q2 2026. The dot shows how much GDP grew against the previous quarter, seasonally adjusted, according to the first estimate INS published, which it calls the "signal estimate".

The latest row, Q2 2026: 0.0%. GDP was flat against Q1 2026.

### By the second provisional version, the figure changes little

INS publishes two provisional versions, about a month apart, the last about 100 days after the quarter. Between the first estimate and the last version, the median difference is 0.1 percentage points, in Q1 2020–Q4 2022 as well as from 2023 on (25 quarters). The largest difference: 0.8 points, in Q2 2023 (+0.9% → +1.7%).

### In the current series, the values of the same quarters look different

In the last position, the chart shows the series INS published on 7 September 2026. Against the first figure, the median difference is 0.3 points for the quarters from 2023 to Q1 2026 and 1.2 points for Q1 2020–Q4 2022.

INS releases point to possible sources of revisions: new data (agricultural and annual data, for instance), the seasonal adjustment recalculated every quarter and, in 2024, a major revision of the national accounts. The data do not attribute each difference to a cause.

### The largest revisions are from 2020–2022

Q2 2020, the quarter at the start of the pandemic: first figure −12.3%, current −8.2%. Q1 2022: +5.2% → +3.2%.

Six of the 12 quarters in Q1 2020–Q4 2022 were revised by at least one point. Among the 13 quarters since, none were.

### Sometimes the sign reverses: a fall instead of growth

Q1 2024: first figure +0.5%, current −0.2%. Q2 2024: +0.1% → −0.5%. In the releases of the time, the values for the two quarters showed no fall; in the current series both are negative.

Two quarterly falls in a row are the usual yardstick for a "technical recession". It is no official dating. It shows how fragile the yardstick is when the falls are a few tenths.

In all, for five of 25 quarters the current value shows a fall where the first showed growth, or the reverse: Q1 2020, Q1 2023, Q4 2023, Q1 2024, Q2 2024.

### On annual growth, the differences are of the same order

On the adjusted series, against the same quarter a year earlier, the median difference is 0.3 points for 2023–Q1 2026 and 0.8 for Q1 2020–Q4 2022. On the gross series: 0.2 and 0.8. The chart shows each of the three series.

### The revisions of Q2 2026 are only beginning

As of 2 October 2026, Q2 2026 has two published figures: the estimate of 14 August and the first provisional version of 7 September. Both give 0.0% against Q1 2026 (adjusted), −2.0% against Q2 2025 (adjusted) and −0.4% against Q2 2025 (gross series).

The history shows how far earlier quarters changed, not how far this one will. Its row is left out of the statistics.

## The size of the revisions, by series and period

Each dot below is one quarter: the distance between the first published value and the current one, in percentage points, without sign. The vertical line marks the median. There are 25 quarters: 12 in Q1 2020–Q4 2022 and 13 from 2023 on.

The differences have no steady direction. For quarterly growth, the current value is lower than the first in 14 quarters, higher in 10 quarters and equal in one quarter. For gross annual growth it is higher in 18 quarters and lower in seven quarters.

| Series | Period | Quarters | Median difference at 100 days | Median to the current series | Maximum | At least 0.5 points | Sign reversed |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Quarterly | Q1 2020–Q4 2022 | 12 | 0.2 | 1.2 | 4.1 | 8 | 1 |
| Quarterly | 2023–Q1 2026 | 13 | 0.2 | 0.3 | 0.8 | 5 | 4 |
| Annual, adjusted | Q1 2020–Q4 2022 | 12 | 0.3 | 0.8 | 2.1 | 9 | 0 |
| Annual, adjusted | 2023–Q1 2026 | 13 | 0.2 | 0.3 | 1.2 | 4 | 1 |
| Annual, gross | Q1 2020–Q4 2022 | 12 | 0.2 | 0.8 | 1.8 | 8 | 1 |
| Annual, gross | 2023–Q1 2026 | 13 | 0.2 | 0.2 | 1.8 | 4 | 0 |

## Method and limits

**The three versions.** The signal estimate appears about 45 days after the quarter, the first provisional version about 70 days, the second about 100. The values of each version come from the tables of the INS releases; the table below gives all of them, with the release link. For Q1 2021 no second version was found, so the first one stands. Q4 2019 is missing: its first release appeared only in March 2020, at 70 days.

**The current value.** It is the value in the series INS published on 7 September 2026. For the seasonally adjusted series it was compared with Eurostat (version of 1 October 2026): the largest difference is 0.05 percentage points, and the display uses Eurostat's rounding. The gross series comes from INS's Excel record.

**The difference.** It is the absolute difference between the current value and the first published value, both rounded to one decimal as the public reads them. "Sign reversed" means one is positive and the other negative; a 0.0 is not counted.

**What it does not show.** There are 25 quarters, not a randomly drawn sample, and the two periods are not independent. Recent quarters have had fewer rounds of revision, so their revision histories are less complete. A revision brings new data; it does not prove the first value was wrong, or that the current one is final. Q2 2026 stays out of the statistics. The Q1 2020–Q4 2022 period includes the pandemic and the major revision of 2024; the gap between the two periods does not show that INS became more precise.

**What this page does not do.** It does not predict how a future quarter will be revised and recommends no financial decision. Past differences do not bound the next one.

## Every quarter, every version

Change in real GDP, %, in each published version. The first column of each group is the signal estimate; the last is the current series. The Q2 2026 row has no second version yet.

| Quarter | Quarterly: first estimate | Quarterly: provisional 1 | Quarterly: provisional 2 | Quarterly: current | Annual, adjusted: first estimate | Annual, adjusted: provisional 1 | Annual, adjusted: provisional 2 | Annual, adjusted: current | Annual, gross: first estimate | Annual, gross: provisional 1 | Annual, gross: provisional 2 | Annual, gross: current |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Q1 2020 | +0.3 | +0.3 | +0.3 | −2.4 | +2.7 | +2.7 | +2.7 | +2.1 | +2.4 | +2.4 | +2.4 | +2.5 |
| Q2 2020 | −12.3 | −12.3 | −11.9 | −8.2 | −10.5 | −10.5 | −10.2 | −8.5 | −10.5 | −10.5 | −10.3 | −9.9 |
| Q3 2020 | +5.6 | +5.6 | +5.8 | +3.8 | −6.0 | −6.0 | −5.7 | −5.4 | −6.0 | −6.0 | −5.7 | −5.0 |
| Q4 2020 | +5.3 | +4.8 | +4.8 | +4.9 | −1.7 | −1.8 | −1.8 | −2.5 | −1.5 | −1.4 | −1.4 | −1.3 |
| Q1 2021 | +2.8 | +2.8 | – | +1.3 | 0.0 | 0.0 | – | +1.3 | −0.2 | −0.2 | – | +0.2 |
| Q2 2021 | +1.8 | +1.8 | +1.9 | +1.0 | +13.6 | +13.6 | +14.4 | +11.5 | +13.0 | +13.0 | +13.9 | +14.8 |
| Q3 2021 | +0.3 | +0.4 | +0.4 | +0.2 | +8.0 | +8.1 | +8.2 | +7.7 | +7.2 | +7.4 | +7.4 | +6.1 |
| Q4 2021 | −0.5 | −0.1 | −0.1 | −0.1 | +2.7 | +3.9 | +3.9 | +2.5 | +2.2 | +2.4 | +2.4 | +2.4 |
| Q1 2022 | +5.2 | +5.2 | +5.1 | +3.2 | +6.5 | +6.5 | +6.4 | +4.4 | +6.5 | +6.5 | +6.4 | +5.8 |
| Q2 2022 | +2.1 | +2.1 | +1.8 | +0.2 | +5.3 | +5.3 | +5.0 | +3.6 | +5.3 | +5.3 | +5.1 | +4.5 |
| Q3 2022 | +1.3 | +1.3 | +1.2 | +0.4 | +4.7 | +4.7 | +4.6 | +3.8 | +4.0 | +4.0 | +3.8 | +3.2 |
| Q4 2022 | +1.1 | +1.0 | +1.0 | +1.2 | +5.0 | +4.9 | +4.8 | +5.0 | +4.6 | +4.6 | +4.5 | +3.8 |
| Q1 2023 | +0.1 | +0.1 | +0.2 | −0.2 | +2.8 | +2.8 | +2.9 | +1.6 | +2.3 | +2.3 | +2.4 | +2.7 |
| Q2 2023 | +0.9 | +0.9 | +1.7 | +1.0 | +2.7 | +2.7 | +2.6 | +2.4 | +1.1 | +1.1 | +1.0 | +1.0 |
| Q3 2023 | +0.4 | +0.9 | +0.9 | +0.9 | +2.1 | +2.9 | +2.9 | +2.9 | +0.2 | +1.1 | +1.1 | +2.0 |
| Q4 2023 | −0.4 | −0.5 | −0.5 | +0.4 | +1.0 | +1.1 | +1.1 | +2.2 | +2.9 | +3.0 | +3.0 | +3.2 |
| Q1 2024 | +0.5 | +0.4 | +0.7 | −0.2 | +1.8 | +1.8 | +2.2 | +2.1 | +0.1 | +0.1 | +0.5 | +0.6 |
| Q2 2024 | +0.1 | +0.1 | +0.3 | −0.5 | +0.8 | +0.8 | +0.9 | +0.5 | +0.8 | +0.8 | +0.9 | +0.9 |
| Q3 2024 | 0.0 | −0.1 | 0.0 | +0.6 | −0.2 | −0.3 | −0.1 | +0.3 | +1.1 | +1.1 | +1.2 | +1.7 |
| Q4 2024 | +0.8 | +0.8 | +0.6 | +0.9 | +0.7 | +0.7 | +0.5 | +0.8 | +0.7 | +0.7 | +0.5 | +0.5 |
| Q1 2025 | 0.0 | 0.0 | 0.0 | −0.4 | +0.5 | +0.6 | +0.6 | +0.7 | +0.2 | +0.3 | +0.3 | +0.3 |
| Q2 2025 | +1.2 | +1.2 | +1.2 | +0.9 | +2.1 | +2.1 | +2.3 | +2.1 | +0.3 | +0.3 | +0.3 | +0.4 |
| Q3 2025 | −0.2 | −0.2 | −0.2 | 0.0 | +1.4 | +1.4 | +1.5 | +1.4 | +1.6 | +1.6 | +1.7 | +1.7 |
| Q4 2025 | −1.9 | −1.9 | −1.8 | −1.9 | −1.6 | −1.5 | −1.4 | −1.4 | +0.1 | +0.2 | +0.2 | +0.2 |
| Q1 2026 | −0.2 | 0.0 | 0.0 | −0.1 | −1.5 | −1.1 | −1.1 | −1.1 | −1.7 | −1.2 | −1.2 | −1.2 |
| Q2 2026 | 0.0 | 0.0 | – | 0.0 | −2.0 | −2.0 | – | −2.0 | −0.4 | −0.4 | – | −0.4 |

## Sources

- [Press releases on quarterly GDP](https://insse.ro/cms/ro/tags/comunicat-pib-trimestrial), INS: For every quarter in 2020–2026: the signal estimate and the two provisional versions, as PDF. The figures of each version come from their tables; each row of the table below links to its three releases.
- [Quarterly gross domestic product: the full series](https://insse.ro/cms/ro/content/produsul-intern-brut), INS: The Excel record of the series as of 7 September 2026 (provisional 1): gross and seasonally adjusted, 1995–Q2 2026. This is where the current value of the gross series comes from; for the adjusted series it was checked against Eurostat.
- [namq_10_gdp: quarterly GDP, Romania](https://ec.europa.eu/eurostat/databrowser/view/namq_10_gdp/default/table?lang=en), Eurostat: Chain-linked volumes, seasonally adjusted, quarterly and annual change, version of 1 October 2026. The figures match the INS series to within 0.05 percentage points; the display uses Eurostat's rounding.
- [Major revision of the national accounts, 2024](https://insse.ro/cms/files/statistici/pib%20trim/info/revizuirea-majora-a-conturilor-nationale-2024.pdf), INS: INS note on the revision of a long series under the European harmonised revision policy. It is one reason older quarters carry a different value today.
- [INS release for Q4 2025, section "Revizuiri" (revisions)](https://insse.ro/cms/sites/default/files/com_presa/com_pdf/pib_tr4r2025_1_0.pdf), INS: An example of a reason INS gives: unadjusted values for Q1–Q3 2025 were revised after data sources were updated (provisional agricultural data and annual data).

---

Marius Comper · INS and Eurostat data · analysis: 2 October 2026. [mariuscomper.uk](https://mariuscomper.uk/).