There's a disrepancy between KSH and Eurostat figures – Why is that?
It is not uncommon for the Hungarian statistical office to publish data that differs from the same data in the Eurostat database. This is somewhat strange, given that Eurostat works with data from national statistical offices. The reason for the discrepancy is usually methodological and often serves to ensure comparability between countries or, at least, data reporting according to a uniform methodology. The most notable example of this is the difference between inflation indicators, where several methodological differences between the national 'consumer price index' (CPI) and the EU 'harmonised index of consumer prices' (HICP) cause the two indicators to diverge.
This is less common in retail sales data. In the December 2025 retail sales data release, several discrepancies were found in the Eurostat data series compared to the previous reports of the KSH.
The differences are small, at just 0.1 percentage points, but the perception of a 0.1% increase versus a 0.2% increase is still different.
In the monthly data series, the figures for October and December 2025 did not align.
| Change in retail sales on a monthy basis (%) | ||
| Eurostat | KSH | |
| July 2025 | -0.4 | -0.4 |
| August 2025 | 0.8 | 0.8 |
| September 2025 | 0.0 | 0.0 |
| October 2025 | 0.6 | 0.5 |
| November 2025 | 0.1 | 0.1 |
| December 2025 | 0.1 | 0.2 |
| Sources: Eurostat, KSH | ||
There is a discrepancy in the annual data series in one place, in July.
| Change in retail sales on an annual basis (%) | ||
| Eurostat | KSH | |
| July 2025 | 1.8 | 1.7 |
| August 2025 | 2.4 | 2.4 |
| September 2025 | 3.0 | 3.0 |
| October 2025 | 3.1 | 3.1 |
| November 2025 | 2.5 | 2.5 |
| December 2025 | 3.5 | 3.5 |
| Sources: Eurostat, KSH | ||
We asked the KSH about the reason for the discrepancy, and the office replied that,
in accordance with the relevant regulations, it uses unadjusted, calendar-adjusted and seasonally and calendar-adjusted monthly averages for the 2021 = 100% volume index time series sent to Eurostat. These figures are rounded to one decimal place. The data is uploaded to the database after validation by Eurostat. Eurostat calculates and publishes data for comparison with the previous month and the same period of the previous year. The KSH performs the same calculations, but uses unrounded data.
In other words, according to the KSH, the discrepancies are due to inaccurate rounding. Although domestic statistics are calculated from accurate data (to at least several decimal places), Eurostat calculates further indicators from the KSH index rounded to one decimal place. Therefore, the KSH data are more accurate.
A practical example
The office supported its explanation with specific calculations. For example, in October 2025, the indicator value calculated using the 2021 base was 103.1866. When converted to a monthly basis, this became 100.5473. Rounding this to one decimal place gave 100.5.
However, Eurostat was given a figure of 103.2, as the value of 103.1866 was rounded up. Eurostat then used this figure to calculate a monthly index of 100.5847, which, when rounded to one decimal place, became 100.6 — one tenth higher than the value published by the KSH.
What can be done about this?
Although the different calculation methods do not produce significantly different results, it is confusing that official institutions use two different figures for the same statistical concept. There are two ways to solve this problem.
- Eurostat could, on the one hand, request more accurate data from national authorities. It is unclear why it is content with data accurate to one decimal place, given that this leads to discrepancies in basic indicators such as the monthly and annual change indices.
- Alternatively, if Eurostat does not want to calculate these indices itself, it could simply ask countries to provide more accurate ones.
Portfolio asked Eurostat for their opinion on the discrepancy in the data, but they did not provide a substantive response. They wrote that discrepancies can sometimes occur due to the rounding of decimal places, but that the office strives for maximum accuracy. This includes improving their methodology.
Cover image (for illustration purposes only): Getty Images









