Regional Crime Figures

Regional Crime Figures

/ EUROPEAN COMMISSION
DIRECTORATE-GENERAL
REGIONAL AND URBAN POLICY
Policy
Economic Analysis

Doc. CR/WG2013/6.1

Regional Crime Figures

1.Introduction

At the request of the DG for Regional and Urban Policy, Eurostat launched a voluntary collection of four types of registered crime per NUTS 3 and NUTS 2 region: homicide, car theft, burglaries and robberies. This note describes the results and outlines issues for consideration.

2.Participation

The participation in this voluntary data collection was high. Out of the 25 concerned MS[1], 22 provided regional data. In addition, six more countries provided data at the NUTS 3 equivalent level for all four types of crime: Croatia, Turkey, Norway, Switzerland, Iceland and Serbia. Liechtenstein and Montenegro are not concerned as they only have one level 3 region. For more see table 1 below.

The OECD regional database also publishes homicide rates, but only at NUTS 1 or 2. The OECD database includes NUTS 1 for the UK, the Netherlands and Greece and NUTS 2 for Sweden and Ireland at NUTS, which is a more detailed breakdown than we received.

3.Results

All four crime figures have been mapped using a three year average per capita. (See annex) and show a high degree of regional variation.

Robberies are clearly more frequent in regions with large cities. For example, in Belgium the NUTS regions that contain Brussels, Antwerp, Liege and Charleroi have a much higher robbery rate than the other regions in the country.

Burglaries also are more prevalent in the more urban NUTS 3 regions see for example the regions containing Vienna and Sofia. Regions with a high share of tourists also tend to see higher burglary rates, for example along the Mediterranean coast of France and Spain and the Algarve in Portugal.

Motor vehicle thefts exhibit both the urban and the tourism destination trend with rates considerably higher in urban regions and in regions attracting many tourists. In addition, some of the border regions also have a higher rate, for example, along the border between Belgium and France or between Germany, Poland and the Czech Republic.

Some urban regions have a higher homicide rate, but the trend is less strong than for the other types of crimes. In some cases, the region next to the capital has a higher rate than the capital region. This is the case for Sophia, Berlin and Bucharest. The regions bordering Moldova have higher homicide rate.

4.Issues

4.1.Under-registration of crime

Of these four types of crime, the number of registered homicidesis probably the most accurate and comparable measure of actual homicides as the under-registration rate is likely to be very low with little variation between countries.

For the other three types of crime under-registration is likely to be higher and more variable between countries. It will depend on things like: the ease of registering acrime, the trust in the police, if the victim is insured and the requirements of that insurance.

Although some of these issues may vary within a country, many of them will be uniform across the country. As a result under-registration is likely to vary less within a country than between countries. Therefore crime rates are likely to be more comparable between regions within the same country.

Without a proper victimisation survey, it is not possible to estimate the under-registration rate in different countries or regions. As a result, differences between countries as well as regions need to be interpreted with caution.

4.2.Data transmission issues:

–Some data was delivered for older NUTS region (IT and robberies in NL).

–Some regional data was provided without region codes (DE, SE)

5.Conclusion and Next Steps

The regional crime data show plausible and interesting patterns, particularly at the NUTS 3 level. Without a Europe-wide data collection, these patterns would be difficult to analyse.

In most countries, these data are collected (and disseminated) at smaller geographies. As a result, providing these data does not require an additional data collection but only an aggregation of existing data.

Delegates are encouraged to comment on the data collection and the results, including whether they would be willing to continue to provide this data.

Table 1: Overview of regional breakdown by country

Type of Crime / Homicide / Car Theft / Burglaries / Robberies / Comments
Country / Level of regional breakdown
Belgium / NUTS 3
Bulgaria / NUTS 3
Czech Republic / NUTS 3
Denmark / NUTS 3
Germany / NUTS 3
Estonia / Nat / Nat= 1 NUTS 2 & 5 NUTS 3
Ireland / Nat / Nat = 2 NUTS 2 & 8 NUTS 3
Greece / Nat / Nat = 13 NUTS 2 & 51 NUTS 3
Spain / NUTS 3
France / NUTS 3
Italy / NUTS 2* / Nat = 110 NUTS 3 (to be reduced) * In NUTS 2006
Cyprus / NUTS 3 / Nat= 1 NUTS 3
Latvia / NUTS 3
Lithuania / NUTS 3
Luxembourg / NUTS 3 / Nat= 1 NUTS 3
Hungary / NUTS 3
Malta / NUTS 3
Netherlands / Nat / NUTS 3 / NUTS 3 / NUTS 3* / * In NUTS 2006
Austria / NUTS 2 / Nat = 35 NUTS 3
Poland / NUTS 2 / NUTS 3 / NUTS 2 / NUTS 3
Portugal / NUTS 3
Romania / NUTS 3
Slovenia / NUTS 3
Slovakia / NUTS 3
Finland / NUTS 3
Sweden / Nat / NUTS 3 / NUTS 3 / NUTS 3
United Kingdom / NUTS 1* / Nat = 12 NUTS 1, 37 NUTS 2 & 139 NUTS 3
* England and Wales Combined
Norway / Level 3
Switzerland / Level 3
Liechtenstein / Level 3 / Nat= 1 level 3
Croatia / Level 3
Montenegro / Level 3 / Level 3 / Level 3 / Level 3 / Nat= 1 level 3
Iceland / Level 3 / Level 3 / Level 3
Former Yugoslav Republic of Macedonia / Nat / Nat / Nat / Nat / Nat= 8 level 3
Serbia / Level 3
Turkey / Level 3

1


1

[1]As Luxembourg or Cyprus consist of a single NUTS 3 region, national data is sufficient.