The history of the statistical office dates back to the early years of the European Coal and Steel Community (ECSC). An auxiliary statistical service was founded in 1952 to provide statistical information to the Community, especially regarding the member states’ coal and steel sector, in order to take measures on that sector. The statistical office has been reorganised several times. The first major organisational change took place in 1958, after the ECSC member states had signed the Treaty of Rome to establish the European Economic Community (EEC). At that point, the statistical division was reorganised to a statistical office for the executive body of the EEC, the Commission, and given its current status as a Directorate General. Later organisational changes have taken place in accordance with the emergence of new policy areas, the enlargement of the Community/Union and a new Commission taking office in Brussels. From the EEC’s first enlargement in 1973, the statistical office was renamed Eurostat.
The initial difficulty was that each member state had a statistical system of its own based on different historical trajectories, despite the long tradition of international statistical collaboration beginning with the first international congress in Brussels in 1853. The statistical office struggled to standardise—or, in EU parlance, harmonise—methods and concepts used in data collection and statistics compilation. Standardisation itself is often a coercive act. In the context of compiling statistics, this means that some ways of seeing the world are replaced by others. Because the statistical office is dependent on the data collected by national statistical institutes, the availability of commensurate data is crucial. This was already guaranteed by the ECSC Treaty, which obliged member states to supply the High Authority with data.
During the first decades, Eurostat played a secondary role in the international statistical community. As the European Community’s political power grew, Eurostat gained an increasingly important role in statistical cooperation. Since the 1960s, Eurostat has exported its statistical know-how as a part of development cooperation, for instance, to former French colonies in Africa. An important means was the training of statisticians, which was furthered by the establishment of the European Training Centre for Economist-Statisticians from the Developing Countries (CESD) in 1962. The centre was funded by the Statistical Office and the French Statistical Institute (INSEE). This can be understood as an effort to promote democracy and good governance but also as a form of statistical colonialism, entrenching European statistical practices (involving methods and concepts) in Africa, EU candidate countries and, more recently, Asian states, such as Georgia, Kazakhstan and Uzbekistan.
Indicatorisation and juridification of statistics since the 1990s
Following the Single Act and the Maastricht Treaty, European integration intensified, and negotiations on economic and monetary union were afoot. Statistics were increasingly used to monitor member states’ performance and the economic convergence. Economic integration, especially, put pressure on Eurostat, as the euro convergence criteria and later the Growth and Stability Pact were built on statistical indicators. The relationship between economic integration and economic statistics has been co-constitutive: it would have been impossible to advance the single currency without statistics, but the single currency simultaneously strengthened economic statistics.
Since the turn of the millennium, Eurostat has invested considerable work in indicators. Indicators are a kind of number often compiled from several variables and given an accessible and simplifying name to allow communication on complex topics. In politically tense processes, indicators are attractive, as they can be used to depoliticise goal-setting and move it from traditional political arenas into the hands of expert groups. Indicators can be seen as a form of soft law, linked and intertwined with other forms of soft law, such as reports and comparisons. Indicators have also been pivotal in the Open Method of Coordination (OMC), itself a form of soft law, launched as an instrument for implementing the Lisbon strategy. Instead of binding legislation, the OMC aims at harmonizing policies through the setting of common objectives and the use of measuring instruments and benchmarking. For instance, agreement on so-called Laeken indicators for social inclusion in 2001 was an important step towards the development of EU social policy and, as such, was a major achievement.
Since the 1990s, legislation has been an important means of guaranteeing the availability of comparable and commensurate statistics crucial for the expanding EU. The European Statistical System is consolidated by a regulation passed by the European Parliament and the Council, and most EU statistics are based on legislation. In addition to legislation, Eurostat has used more persuasive forms of soft power to harmonise statistics, launching a staff exchange system in the 1980s and a training programme for European statisticians in the early 1990s. These schemes, still in operation, have familiarised numerous statisticians from national statistical institutes with Eurostat and the EU’s statistical system.
Balancing professional independence and political relevance
The legitimacy and credibility of statistics, fundamental to their usability, are maintained by separating the statistical and politico-administrative spheres, which are distinct but indispensable to each other. Professional independence, seen as guaranteeing the quality of statistics, is the first principle in the European Statistics Code of Practice. However, the relationship between policy and statistics has been paradoxical. On the one hand, statistics are expected to be relevant for the Union’s policymaking; on the other hand, they are presumed to be independent.
In the aftermath of the global economic crisis in 2008, a crisis also emerged over Greek economic statistics, stemming from excessively close relations between government politics and statistics. In 2009, the budget deficit and public debt figures first released by Elstat, the Greek national statistical institute, were too low. The corrected higher figures published after the new director-general Andreas Georgiou took office led to a rapid deterioration of Greece’s economic situation. Georgiou was accused of undermining the national interest and became a scapegoat in Greece. He resigned in 2015 and faced a criminal investigation in 2016, receiving a two-year suspended sentence. The international statistical community and the European Commission condemned the sentence as a form of political persecution.
Despite the difficulties, Eurostat has succeeded in making things that hold together, whether in terms of inflation, public deficit, employment or population size, so that the EU can act on them. Such a statistical capacity has become indispensable for the political institutions. Statistics produced by Eurostat are used for various purposes in EU governance. Statistics are used to calculate the financial contributions of member states and to distribute regional and social funds. In policymaking, strategic targets are set via statistical indicators, and the indicators are, in turn, used to monitor member states’ performance against the targets. Indeed, statistics have shaped how the EU, as a political entity and community, is known and imagined. Statistics are even seen as furthering the formation of European identity, as they enable people to learn more about their fellow citizens.
Alastalo, Marja, “Eurostat: Making Europe comparable and commensurable,” In Heiskala, R. & J. Aro (eds.), Policy Design in the European Union. An Empire of Shopkeepers in the Making? (London: Palgrave Macmillan, 2018, 87-110).
Atkinson, Anthony. B., Marlier, Eric, Nolan, Brian, “Indicators and targets for social inclusion in the European Union,” JCMS: Journal of Common Market Studies 42 (2004).
Desrosières, Alain, The Politics of Large Numbers: A History of Statistical Reasoning (Cambridge, Massachusetts: Harvard University Press, 1998).
Desrosières, Alain, “Measurement and its uses: Harmonization and quality in social statistics,” International Statistical Review 68 (2000).
Penissat, Étienne, Jay, Rowell, “The Creation of a European Socio-economic Classification: Limits of Expert-driven Statistical Integration,” Journal of European Integration 37 (2015).