Sensitivity analysis in the Council of the European Union

Abstract

The Council of the European Union is an important decision-making body in the European Union (EU) that negotiates, agrees, and adopts legislative acts, mostly together with the European Parliament. It is an intergovernmental institution, where each country is represented by a national minister whose voting weight is proportionate to the population of its country.

If the Council votes on a proposal by the European Commission or the High Representative of the European Union for Foreign Affairs and Security Policy, the proposal is adopted if it is supported:

(i) by member states representing at least 65% of the total EU population (population criterion); and (ii)by member states representing at least 55% of the EU member states (states criterion). Furthermore, any blocking minority must include at least four countries. Qualified majority is the most widely used voting method in the Council of the European Union; about 80% of all EU legislation is adopted with this procedure. Therefore, the above conditions of qualified majority, stipulated by Article 16 of the Treaty on European Union, have a powerful impact on the voting power of the member states.

The previous literature has extensively studied the effects of Brexit, as well as other possible entries and exits. On the other hand, it has never been examined thoroughly how the power distribution changes if the qualified majority rules are modified. Our paper aims to address this research gap.

Besides the academic community, the issue can be interesting for policy-makers and political scientists. For example, the French and German governments have recently convened a ``working group on EU institutional reforms'', which has submitted a report entitled ``Sailing on High Seas: Reforming and Enlarging the EU for the 21st Century'' in September 2023. This includes an adjustment of the two qualified majority thresholds to 60-60%, however, the impact of the reform with respect to the voting power of the member states has not been examined.

Our sensitivity analysis is based on the Shapley--Shubik index, one of the most widely used measures of voting power in weighted voting games. The population criterion is analysed between 51% and 80% since 51% is a natural lower bound and the population weight of the largest member state, Germany, is 18.59%. The number of supporting member states is assumed to be between 11 (40.7% of the 27 EU countries) and 20 (74.1%). A lower threshold would be harmful for the small countries, and a higher threshold would make it challenging to adopt any proposal.

The populations of the countries are updated every year; the weights in our analysis come from the Council Decision (EU, Euratom) 2022/2518 of 13 December 2022 amending the Council's Rules of Procedure. The Shapley--Shubik indices are computed with the python package powerindices of Frank Huettner. They are exact values, not approximations by simulations, but the blocking minority rule is ignored as it affects only 21 coalitions.

The decision ability of the European Union is measured by the proportion of winning coalitions. The inequality of voting powers is quantified by the normalised Herfindahl--Hirschman index.

The main results can be summarised as follows:

(i) Currently, 13.2% of all possible coalitions is able to accept a proposal. The decision ability is guaranteed to decrease if the population criterion is at least 69% or the states criterion is at least 17.

(ii) The proportion of winning coalitions can be increased only by lowering the states threshold if the relative power of large countries is not allowed to decrease.

(iii) The recently suggested reform with 60-60% thresholds strongly favours small countries as the normalised Herfindahl--Hirschman index would drop below one-third of its current value, and it also reduces the decision-making probability to 8.4%.

(iv) The voting power of Hungary cannot be substantially lower than its population share of 2.17% under any pair of thresholds. However, it cannot be higher than 3% if the states criterion remains at 15, and will be lower than 3.5% even if the states criterion becomes higher.

Our findings are essential to assess any proposal for reforming the qualified majority voting system in the Council of the European Union.