Date: 12 December, 2018


Quality statistics are fundamental to economic policy. Policymakers have to rely on official statistics all the time. They have to, for example, decide on government spending based on, among many factors, an accurate picture of tax revenues. When policymakers take decisions on borrowing, they trust that the official statistics report accurately on debt and other necessary information. But it is not only policymakers who rely on official statistics. It is also citizens, financial market participants, companies and many other stakeholders who have a right to expect that statistical offices provide an accurate picture of the economic phenomena they describe.

In the European Union, economic surveillance of member states is based on a number of critical official statistics. In particular, the deficit and debt-to-GDP ratios are central to Europe’s Stability and Growth Pact, the fiscal surveillance framework. The sources of these statistics are largely national statistical offices, even though the European statistical office Eurostat plays a role in validating reported statistics.

This system has shown weaknesses. It is well known that prior to the introduction of the euro, statistical tricks were used to comply with the Maastricht criteria. Research has shown that during the first years of the euro, member states’ deficit numbers were modified when compliance with the Stability and Growth Pact’s three percent deficit number was endangered.

But the Greek experience of 2009-10 still came as a shock and turned into a wake-up call on the shortcomings of the European statistical system, which sadly remain substantial even after the post-2009 reforms. The size of the correction of misreported Greek deficit numbers came as a surprise to policymakers, markets and citizens. The size of the Greek fiscal problem was greater than officially reported for many years.

Clearly, the European statistical system needs to become better to avoid future surprises. The author of this essay makes a proposal on how the European statistical system could be reformed to reduce the likelihood that statistics will be misused by national governments against European interests.

I am happy that Andreas Georgiou has agreed to reflect on how Europe’s statistical system could evolve. He is, of course, uniquely qualified to write about the topic. His successful International Monetary Fund career and his tenure during 2010-2015 at the helm of the Greek national statistical institute ELSTAT, provide him with unique insights into the system’s problems.

This is not an essay about his time at ELSTAT, nor is it an essay about the still ongoing legal travails to which he is subjected in the context of his reporting of statistics in compliance with Eurostat rules. Instead, this essay is a facts-based empirical and theoretical exposition of the problems confronting the European statistical system. It then provides a blueprint of how it could evolve. Even those who won’t agree with his vision for Europe’s statistical system will concur that Europe badly needs to reflect on how to improve the quality of its official statistics. I hope this essay can make a contribution to this debate.

Guntram Wolff, Director of Bruegel

Brussels, November 2018