Our Data & Methodology
How we source, process, and present economic data for all 27 EU member states — with rigour, transparency, and independence.
Data Sources
Where Our Data Comes From
All economic data presented on Eunomist is drawn from official statistical authorities operating across Europe and internationally. These institutions compile, verify, and publish authoritative datasets covering economic output, labour markets, prices, fiscal positions, trade, and social indicators across EU member states.
We do not rely on commercial data vendors, unofficial estimates, or secondary sources when primary statistical releases are available. Where data from multiple official sources is available for the same indicator, we use the source that applies the most consistent methodology across member states to support fair cross-country comparison.
All data is attributed at the indicator level. The specific source used for each indicator can be identified through the dataset and methodology flags displayed alongside each figure.
Data Collection
Data Collection Process
Eunomist fetches data directly from official statistical APIs maintained by Eurostat, the European Central Bank, the International Monetary Fund, and the World Bank. The primary format used by Eurostat — the principal source for EU-level economic indicators — is JSON-stat 2.0, a standardised JSON schema designed for the compact representation of multidimensional statistical datasets. JSON-stat 2.0 encodes dimension labels, unit metadata, status flags, and observation values in a single structured payload, eliminating the ambiguity that can arise from manual spreadsheet handling.
The data collection pipeline operates on a scheduled basis, querying source APIs for updated releases as official publication calendars indicate new data is available. Eurostat's publication schedule varies by indicator: quarterly national accounts data (for GDP and growth rates) is typically released six to eight weeks after the reference quarter ends, while annual indicators such as government debt ratios follow a longer release cycle aligned with the European Semester fiscal surveillance process. The pipeline is designed to detect when source values have been revised — a routine occurrence in official statistics — and update stored values accordingly, with the data vintage date recorded alongside each figure.
Validation checks are applied at three stages. At ingestion, values are checked against expected ranges for each indicator — implausible outliers (for example, a reported inflation rate of 500% for a country not experiencing a hyperinflationary episode) are flagged for manual review before publication. At the cross-reference stage, values for the same indicator and time period are compared across multiple available datasets to detect series breaks or unexplained discrepancies. At the display stage, status flags embedded in the source data — provisional estimates, forecast values, breaks in series, data under embargo — are preserved and surfaced on the platform rather than suppressed. Readers should treat flagged values with appropriate caution.
-
Automated Ingestion via Official APIs
Data is fetched directly from Eurostat, ECB, IMF, and World Bank APIs in JSON-stat 2.0 and SDMX formats on a regular schedule, ensuring figures reflect the most recently published official values.
-
Cross-Reference Validation
Values are cross-referenced against multiple releases for the same indicator period to detect revisions, anomalies, or breaks in series. Discrepancies are flagged for review before publication.
-
SDMX Quality Flags Preserved
Where the source data carries quality flags (provisional, estimated, break in series, forecast), these are preserved and displayed alongside the figure on the platform — never suppressed.
-
Normalisation Without Adjustment
Indicator values are stored in standardised units to support consistent comparison. Values are not adjusted, modelled, or imputed beyond what the originating statistical authority has published.
Indicator Definitions
What Each Indicator Measures
Eunomist tracks 18 key economic indicators across all 27 EU member states. Each indicator is drawn from an official source and defined in line with the methodology used by that source. The definitions below explain what each indicator measures, why it matters for understanding a country's economic position, and how to interpret it correctly — including common misreadings to avoid.
GDP (Current Prices)
Source: Eurostat — nama_10_gdp
Gross Domestic Product at current market prices measures the total monetary value of all goods and services produced within a country's borders in a given year, expressed in euros without any adjustment for price level differences across countries. This is the headline measure of economic size. It is most useful for assessing a country's weight within the EU economy — Germany's €4.1 trillion GDP accounts for roughly a quarter of total EU output — but it should not be used to compare living standards, since it takes no account of differences in price levels or population size.
GDP (Purchasing Power Standards — PPS)
Source: Eurostat — nama_10_gdp
GDP expressed in Purchasing Power Standards removes the distortion caused by price level differences across EU member states. A euro buys significantly more in Bulgaria or Romania than in Denmark or Luxembourg, which means nominal GDP comparisons overstate the economic gap between high-price and low-price countries. PPS-adjusted GDP converts national figures into a common artificial currency that accounts for these differences. This is the correct measure to use when comparing economic size across countries on a like-for-like basis.
GDP Growth Rate
Source: Eurostat — nama_10_gdp
The annual rate of change in real GDP — that is, GDP stripped of the effects of inflation so that only genuine changes in economic output are measured. A positive growth rate means the economy produced more goods and services than the previous year; a negative rate indicates contraction. Growth rates matter because they determine whether a country is converging toward or diverging from the EU average over time. A consistent 3% growth rate in a lower-income member state, sustained over a decade, produces meaningful improvements in living standards and EU cohesion.
GDP per Capita (Nominal)
Source: Eurostat — nama_10_pc
Total GDP divided by the resident population, expressed in euros at current market prices. GDP per capita is the standard measure of average economic output per person and is widely used as a proxy for living standards, though it has well-documented limitations: it says nothing about distribution, excludes non-market activity, and can be distorted by very large financial sectors (as in Luxembourg, where GDP per capita is inflated by cross-border financial flows that do not reflect the living standards of Luxembourg residents).
GDP per Capita (PPS)
Source: Eurostat — nama_10_pc
GDP per capita adjusted for purchasing power differences across countries, expressed as an index where the EU average equals 100. A country with a PPS per capita index of 130 has average purchasing power 30% above the EU norm; a country at 70 is 30% below. This is the most useful single indicator for comparing living standards across EU member states, though it remains an average that conceals significant intra-country inequality. Eurostat uses this indicator as the primary criterion for Cohesion Fund eligibility.
Unemployment Rate
Source: Eurostat — une_rt_a (ILO definition)
The proportion of the labour force that is without work, available to work, and actively seeking employment — measured according to the International Labour Organization definition to ensure cross-country comparability. The ILO definition requires all three conditions to be simultaneously true, which means people who have given up looking for work (discouraged workers) are excluded, and people working as few as one hour per week are counted as employed. This can make unemployment rates appear lower than the true degree of labour market slack, particularly in countries with large part-time or precarious work sectors.
Employment Rate
Source: Eurostat — lfsa_ergan (age group 15–64)
The share of the 15–64 age group that is in employment, expressed as a percentage. The employment rate is often more informative than the unemployment rate because it captures labour market participation directly rather than measuring the residual of those actively seeking work. A country can have both a low unemployment rate and a low employment rate if many people are neither working nor looking for work. Sweden and the Netherlands consistently lead the EU with employment rates above 80%, while several Southern European member states remain below 65%, reflecting structural differences in female labour force participation, early retirement norms, and youth employment.
Inflation Rate (HICP)
Source: Eurostat — prc_hicp_aind
The annual rate of change in the Harmonised Index of Consumer Prices — a standardised measure of consumer price inflation designed to be comparable across EU member states. HICP is the ECB's primary inflation target variable and covers a basket of goods and services consumed by households, including food, energy, housing costs (excluding owner-occupied housing in the standard measure), and services. Because the HICP methodology is harmonised across member states, it is the correct measure to use for cross-country inflation comparisons, in preference to national CPI measures which use varying methodologies.
Government Debt (% of GDP)
Source: Eurostat — gov_10dd_edpt1 (Maastricht definition)
General government gross debt as a percentage of GDP, measured according to the Maastricht Treaty definition used in the EU's Stability and Growth Pact framework. This covers the consolidated gross debt of central government, state government, local government, and social security funds — but excludes financial sector liabilities even where the government holds a majority stake. The Maastricht Treaty's reference value for sustainable debt is 60% of GDP; member states above this threshold are subject to the EU's Excessive Deficit Procedure. As of the latest data, twelve EU member states exceed this threshold, with Greece at over 160% of GDP at one extreme and Estonia below 20% at the other.
Government Deficit (% of GDP)
Source: Eurostat — gov_10dd_edpt1
General government net borrowing (deficit) or net lending (surplus) as a percentage of GDP, on an accruals basis following ESA 2010 accounting standards. A deficit means the government is spending more than it is receiving in revenues in a given year; a surplus means the reverse. The Maastricht reference value for the deficit is 3% of GDP. The deficit is a flow measure (what happened this year), whereas government debt is a stock measure (the accumulated total). A country can run a small deficit while its debt ratio is still falling if nominal GDP is growing faster than the new borrowing added each year.
Current Account Balance
Source: Eurostat — bop_eu6_q / IMF Balance of Payments Statistics
The current account records a country's transactions with the rest of the world — exports and imports of goods and services, income flows (such as wages paid to cross-border workers and investment returns), and current transfers. A surplus means the country is earning more from the rest of the world than it is spending; a deficit means the reverse. Persistent large surpluses — Germany has run a current account surplus exceeding 5% of GDP for over a decade — are a point of EU macroeconomic imbalance surveillance, because they can contribute to demand deficits elsewhere in the currency union.
Population
Source: Eurostat — demo_pjan
The total resident population as of 1 January of the reference year, as published in Eurostat's demographic statistics. Population figures are used as the divisor for all per-capita calculations on the platform. Population trends — particularly declining working-age populations in Eastern and Baltic member states driven by emigration and below-replacement fertility — are a key structural factor shaping long-run growth potential and fiscal sustainability across the EU.
AI-Assisted Analysis
How Our Narrative Analysis Is Generated
The written analysis that accompanies data on Eunomist — the contextual paragraphs on country pages, indicator explanations, and topic deep-dives — is generated using Claude, Anthropic's AI assistant, operating under detailed editorial instructions that specify the analytical standard, tone, and content requirements for each page type.
For each analysis, the model receives structured data inputs: the indicator values for the country or topic in question, the EU-wide average for the same indicator, relevant peer country comparisons, and historical trend data where available. The model is instructed to produce analysis that contextualises every figure against at least one reference point, avoids generic claims that could apply to any country, and maintains the editorial standard of a serious financial publication — authoritative, clear, and precise without being condescending. The model is explicitly instructed not to produce placeholder text, hedged non-answers, or analysis that defers substance to an unspecified future.
We use AI-assisted analysis for two reasons. First, scale: Eunomist covers 27 countries across 18 indicators, producing hundreds of data points that all require contextualisation. Human-only writing at this scale would either require a large editorial team or produce inconsistent quality across the coverage set. Second, consistency: AI-generated analysis, operating from the same data inputs and the same editorial instructions, applies the same analytical framework to every country and indicator — eliminating the uneven treatment that can arise when different human authors cover different countries with varying depth and attention. All output is reviewed against content standards before publication.
Indicators
How Indicators Are Calculated
Eunomist does not calculate or model primary indicator values. All values for individual country indicators are sourced directly from official publications. Where EU-wide averages are displayed, these are computed from the available country data in our dataset using either simple mean or weighted mean methods, as appropriate for the indicator type.
EU ranking positions are calculated from the set of EU member states for which data is available in the same reference year. Rankings may differ from those published by source authorities where those rankings use different reference years, coverage, or weighting methods.
Year-on-year comparisons: Trend indicators (shown as up/down arrows on indicator cards) compare the most recent available year against the immediately preceding year. The difference shown is the absolute change in the indicator's unit, not a percentage change of the value.
Limitations & Caveats
What the Data Doesn't Capture
Official economic statistics are among the most carefully constructed quantitative measures in the social sciences — but they have real limitations that any serious reader of economic data should understand. GDP, the most-cited indicator on this platform, measures the value of market transactions but excludes non-market activity: unpaid care work, household production, and the informal economy. The informal economy is particularly significant in some EU member states — estimates for Southern and Eastern European countries range from 15% to 30% of GDP — meaning that official GDP figures systematically understate true economic activity in countries with larger informal sectors. Cross-country comparisons of GDP levels are therefore comparisons of what is measured, not necessarily of what exists.
EU averages can mislead in two related ways. First, the EU is a union of very unequal economies: Luxembourg's GDP per capita is more than six times that of Bulgaria. An EU average that spans this range contains less information than the distribution itself. Second, the average can move for reasons unrelated to genuine economic change — when new member states join, or when statistical methodologies are revised, the EU average shifts without any underlying change in economic conditions. We try to contextualise averages with the relevant distribution, but readers should treat EU averages as reference points rather than benchmarks.
Wealth distribution is almost entirely absent from our current indicator set. GDP per capita tells you nothing about how output is shared between citizens. A country with high GDP per capita and extreme inequality may offer worse material conditions for the majority of its population than a lower-income country with a more equal distribution. Similarly, unemployment rates tell you nothing about employment quality — the prevalence of zero-hours contracts, involuntary part-time work, or in-work poverty. Where these distributional dimensions are directly relevant to the interpretation of an indicator, we note them in the accompanying analysis. We are working to expand coverage of distributional and quality-of-work indicators as data availability improves.
The interpretive context provided alongside indicators on country pages is editorial in nature and intended to aid understanding. It does not constitute financial, investment, tax, or legal advice. Always consult qualified professionals for decisions involving specific country situations.
Update Schedule
When Data Is Refreshed
Data on Eunomist is updated on a rolling basis as new official statistical releases become available from source authorities. The reference year displayed on each indicator card identifies the vintage of data shown — this is the year the data refers to, not the year it was published. In most cases, the most recent available data is one to two years behind the current calendar year, reflecting the time statistical agencies require to compile, validate, and publish national accounts and survey data.
Different indicators are published on different schedules. Quarterly GDP flash estimates are released six to eight weeks after the reference quarter ends; final national accounts follow six to nine months later. Inflation data is released monthly with minimal lag. Government debt and deficit data, which must be compiled and reported by member states under the European Semester process, is published annually, typically in April and October. Population data is released annually, generally in the first half of the following year. To check the vintage of a specific figure, refer to the data source attribution displayed alongside each indicator card.
Contact
Questions or Corrections
If you identify an error in the data or methodology, or have questions about a specific indicator, please contact us through the about page. We take data accuracy seriously and will investigate and correct verified errors promptly.