18. Religion and the European Union

See also Nelsen and Guth (2015).

We include all countries that were members of the Euro zone and EU member states in 2010 if the relevant data are available. We retrieved country-specific population shares of religious denominations from the World Religion Database. These shares are defined such that they assume values between 0 and 1. The 16 countries included in our analysis are Austria, Belgium, Cyprus, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Malta, the Netherlands, Portugal, Slovakia, Slovenia, and Spain. Monaco, San Marino, and the Vatican also used the Euro in 2010, but were not EU member states. Where available, we used shares based on wave 4 of the European Values Survey, which took place in 2008/2009. The only exception is Italy, for which we used shares based on the Pew Global Attitudes Project 2007. Long-term bond yields in percent were retrieved from the OECD. We use data for December 2011, the last month of the year examined in Chadi and Krapf (2017), but the results are robust to choice of the month for which we use data. Data on public debt were also retrieved from the OECD, but are missing for Cyprus, Estonia, Lithuania, and Malta. For these four countries, we filled in numbers published by the World Bank. Our measures of institutional quality, “rule of law” and “control over corruption,” were retrieved from the World Bank Governance Indicators. We use debt and institutional quality data for 2015, the latest year for which they were available. Debt is measured relative to gross domestic product (GDP) in percent. All indicators of macroeconomic performance and institutional quality are available online from https://​data.​oecd.​org/​interest/​long-term-interest-rates.​htm, http://​data.​worldbank.​org/​data-catalog/​world-development-indicators, and http://​info.​worldbank.​org/​governance/​wgi. Bond yields were not available for Cyprus and Malta.

For details on Christian Orthodox confessional culture, see Ware (1993), pp. 288–90.

The observed link between religion and economic outcomes may be related to Bordo and Istrefi’s (2018) finding that, controlling for other characteristics, Protestant members of the Federal Reserve’s Open Market Committee tend to be more hawkish, Jews more dovish, and Catholics centrist.

The issue of integrated versus multi-speed Europe goes back at least to 1994, when Protestant Wolfgang Schäuble, at the time leader of the CDU/CSU parliamentary group in the German Bundestag, proposed that there should be a “Two-Speed Europe.” This concept would have divided the EU into a more advanced and homogeneous “Core Europe” that progresses more quickly and into periphery countries that might join the core later based on certain criteria. The idea was instantaneously rejected by Catholic German Chancellor and CDU party leader Helmut Kohl. But the creation of the Euro was a first step toward such a multi-speed Europe with Protestant countries such as the United Kingdom, Sweden, and Denmark not immediately joining the common currency. A concept for EU reform outlined by the President of the European Commission Jean-Claude Juncker, a Catholic, in a speech to the EU parliament on September 13, 2017, reignited this discussion. Among Juncker’s proposals was the extension of the Euro area to all members of the EU. The opt-outs of the past may indeed have been a brake on European integration, and bringing all member states closer together may help implement the reforms that are necessary to solve the Euro crisis. Juncker’s speech received much applause but also immediately faced opposition, notably from Protestant Dutch and Danish Prime Ministers Mark Rutte and Lars Løkke Rasmussen.

Referring to a vast literature on the causes and consequences of the Reformation (Becker et al. 2016), Stark (2014), for example, argues that, nearly without exception, autocrats opted in favor of Lutheranism in places in which the church had the greatest local power. Local princes such as Frederick the Wise of Saxony may have been inclined to support Martin Luther because, according to his teachings, the church should not wield any secular power. Similarly, the Reformation may have failed in countries like Spain because the Spanish kings had already won over Rome and reformed their churches. In the fifteenth century, the Spanish grandparents of Emperor Charles V (Charles I in Spain), King Ferdinand II and Queen Isabella, had been able to deprive Rome of its secular powers in Spain. But he was not able to achieve the same in the Holy Roman Empire.

A possible explanation for the Bavarian paradox is that the influx of asylum seekers via the Balkan route in Lower Bavaria, one of Germany’s most Catholic regions, caused a disproportionate surge in AfD votes there.