The Regional Authority Index

The Regional Authority Index measures multilevel governance across the world. The data is collected by Liesbet Hooghe, Gary Marks, Arjan H. Schakel, Sara Niedzwiecki, Sandra Chapman-Osterkatz, and Sarah Shair-Rosenfield and extensively described in two books, “Measuring Regional Authority: A Postfunctionalist Theory of Governance: Volume I” and “Community, Scale, and Regional Governance: A Postfunctionalist Theory of Governance: Volume II.”. The most recent version of the dataset (v3) can be accessed here. We will present the dataset below. While reading, please keep in mind the questions below and answer them once you reached the end. In the final panel, we will provide a link to a platform with an interactive version of the dataset and additional tasks.

Table 1: General Tasks for the Dataset

Tasks
Within Europe, which countries have particularly high levels of self-rule and which have high levels of shared rule?
Which countries have a lot of differentiations in EU law?

Dataset Description

Overall Information

The dataset covers 98 countries over the period from 1950 to 2018. Data is collected on the regional level, which are the authority levels between local government and national government. In each country, this means that there might be several distinct levels of authority in the dataset. For example, in Germany, there are data on both the Länder and also on “Kreise” (cities or collections of municipalities). In France, there are data both on the level of Départments and on Régions. In the UK, data exists for counties and regions. The depth of data varies with the degree of delegation of power to the subnational level.

Data for Europe

As outlined in Chapter 3 of the book, the RAI provides extensive coverage of Europe. The RAI is particularly high in Germany, Bosnia and Herzegovina, and Spain (in fact these three countries have the highest RAI in the whole dataset in 2018), and considerably lower in Estonia and Iceland

The two main components of the RAI are self-rule (n_selfrule in the dataset) and shared rule (n_sharedrule). In fact, the RAI measure is simply a sum of those two components. Figure 3 shows the measure of self-rule across Europe in 2018. The index is made up of the variables n_instdepth (autonomy of the regional government), n_policy (range of policies decided at the regional level), n_fiscauto (regional control over taxation), n_borrowauto (how extensively regional governments can borrow money) and n_rep (whether there are regional assemblies and executives). European countries with high levels of self-rule are Germany, Italy, Belgium and Bosnia Herzegovina.

The second component of the RAI is called n_sharedrule and refers to the extent to which the regional government can influence policy making at the national level. Its components are n_lawmaking (influence of regional presentatives on national legislation), n_execcon (influence of regional governments on national intergovernmental meetings), n_fisccon (influence of regional representations on distribution of national taxes), n_borrowcon (influence of regional representations on borrowing decisions) and n_constit (influence of regional representatives on constitutional changes). This indicator is particularly high in Spain, Germany and Belgium, but very low in most of Europe.

Data for Latin America and Asia

RAI datasets also provide information beyond Europe for Asia and the Americas. Figure 5 shows the RAI for Asia. The dataset provides information on most of South Asia (e.g. India, Pakistan, Bangladesh), South-East Asia (e.g. Indonesia, the Philippines, Malaysia) as well as Russia and China. Additionally, Australia and New Zealand are covered in the dataset. Finally, Figure 6 shows the data for Latin America.

Regional Data

Besides the country-level data, the RAI provides regional level data. This is interesting for countries in which some regions have more autonomy than others, such as the UK and Spain. Table 2 shows the RAi for the UK and Spain. Besides the name of the country and region, the table shows the following variables form the dataset: “type” describes the type of region, as coded by the researchers (S: standard, Y: asymmetric, distinctive authority on one or several dimensions of the RAI, A: autonomous, exempt from the country-wide constitutional framework, individual jurisdiction, D: dependency, region without or with very little autonomy)

In Spain, Ceuta and Melilla are categorized as autonomous regions, while Araba, Bizkaia, Gipuzkoa, Galiza, Navarre, Catalunya and Euskadi are classified as asymmetric regions with special statuses. Compared to the provincias in the rest of Spain, these regions have considerably higher levels of authority, both when it comes to self-rule and shared rule.

Meanwhile, in the UK, some power has been transferred to Northern Ireland, Scotland and Wales. However, the RAI shows that this devolution has not been uniform: Scotland enjoys the largest degree of self-rule among these regions, while Wales has considerably less. This speaks to the power of the regional parliaments and is rooted in the history of devolution as outlined in Chapter 3.

Table 2: RAI in the UK and Spain

country_name	region_name	selfrule	sharedrule	RAI
Spain	Catalunya/Cataluña	2.2487609	1.5259449	3.7747058
Spain	Ceuta	0.0249752	0.0178394	0.0428147
Spain	Comunidad Foral de Navarra/Nafarroa	0.2050998	0.1435699	0.3486697
Spain	Euskadi/Pais Vasco	0.5134861	0.4901458	1.0036320
Spain	Galiza/Galicia	0.8292264	0.5626893	1.3919157
Spain	Melilla	0.0243191	0.0173708	0.0416899
Spain	Araba/Álava	0.0753709	0.0411114	0.1164823
Spain	Bizkaia/Vizcaya	0.2716617	0.1481791	0.4198408
Spain	Gipuzkoa/Guipúzcoa	0.1664535	0.0907928	0.2572464
United Kingdom	Northern Ireland	0.3439317	0.1862964	0.5302281
United Kingdom	Scotland	1.1733632	0.5447758	1.7181390
United Kingdom	Wales	0.6303190	0.3151595	0.9454785

Differentiation

We now turn to another dataset important for studying the multilevel system of the European Union: The Datasets on legal integration and differentiation in the EU (EUDIFF), provided by the InDivEU project by Frank Schimmelfennig and Thomas Winzen. It can be accessed here.. The data is split across three datasets recording legal integration and differentiation in the EU since 1958. The first dataset collects the main consolidated treaties of the EU (n=11), the second the main legislative acts (n=4967) and the third EU-related international law (n=40). In the following we will focus on the second dataset.

Figure 7 shows the percentage of legislative acts that have at least one differentiation in the first year they were recorded. This number fluctuates over time, in some years over 30% of new legislative acts have differentiations.

Next, we look at differentiations by policy area as coded in the dataset. By far the most differentiations are present in the policy area of justice and interior, followed by monetary policy.

Finally, we can look at the member state level. Figure 9 shows that the countries that have the most differentiations are Denmark, the UK and Ireland which negotiate differentations in about 4% of legal acts.