How many femicides? International comparisons are impossible

The second part of EDJNet's investigation on femicides once again highlights the lack of data that would enable useful comparisons at the European level. The case of France provides a good example of such shortcomings.

Published On: April 22nd, 2024


The art of comparison is not an easy one. For the second year running, EDJ Net has brought together fifteen European newsrooms to work on femicide in Europe, under the aegis of the Mediterranean Institute for Investigative Reporting (MIIR), a Greek NGO set up in 2019.

As was the case last year, the survey highlights the impossibility of comparing figures from one European country to another.

The data between EU Member States remains very heterogeneous,” says Cristina Fabre Rosell, head of the “gender violence” unit at the European Institute for Gender Equality (EIGE).

“The data is based on national statistics and other administrative sources (health systems, courts) for homicides, as well as analysis of press articles by associations. The data is therefore not comparable, and comparisons between Member States are currently impossible”. However, Roselle adds that it is “certain that such violence is under-reported”.

Sources that are difficult to cross-reference

Reports by the European Institute for Gender Equality (EIGE), an agency of the European Union, are a primary source on the subject. Unfortunately, the Institute’s most recent report dates back to 2021, and is based on data ending in 2018.

The other source is Eurostat, the official statistical body of the European Union. It does not count femicides – i.e. the murder of a woman because of her gender – as such, but gives figures on intentional homicides, with the possibility of specifying the relationship between the victim and her attacker. This makes it possible to determine whether the assailant was the victim’s partner or a family member.

Cross-referencing these sources does not produce consistent results. If we look at the EIGE figures, between 2012 and 2022 (for figures that can be updated) there were 4,221 femicides in Europe, bearing in mind that this data excludes Belgium, Bulgaria, Denmark, Ireland, Poland and Romania.

According to Eurostat, there were 14,143 intentional homicides of women in the EU-27 over the same period. The EDJNet survey, conducted in nineteen countries, recorded 6,754 femicides, of which 4,334 were committed by a partner and 2,472 by another family member.

This inconsistency is mirrored at the national level. In France, the gap between the reality of violence against women (as measured by victimisation surveys) and the figures recorded by the police is colossal. In 2022, 321,000 women reported having been victims of domestic violence, but only 15 percent of these women filed a complaint.

In terms of sexual assaults, attempted rapes and rapes (whether committed within or outside a marital context), there were 217,000 female victims in 2022, of whom only 6 percent filed a complaint, according to the Observatoire national des violences faites aux femmes. Apart from forced suicides, femicide is the exception, since the police are aware of the facts by default.

This does not mean that femicide is analysed for what it is. In France, in the absence of a feminist perspective in the construction of figures and in judicial categories (see “Femicide: what can public policy do?”), femicide is thought of solely in terms of marital relationships, thereby neglecting many murders of women that are committed outside of this context and could certainly be classified in this category. Hence the importance of the work of feminist associations on this issue.

What’s more, there are several official sources: in the public debate, the authoritative figures are those of the Délégation aux victimes (DAV), which has published a national survey of violent deaths within couples every year since 2006.

Poorly segmented data

The ministerial statistical service for internal security (SSMSI) also publishes data on homicides in the “ Insecurity and delinquency survey. But the way they are presented is surprising: the homicide figures do not break down by gender, including for domestic or sexual violence.

To obtain this breakdown by sex, we have to refer to Eurostat, which is based on data supplied by… the SSMSI itself! Given the processing time involved, this data is not available as quickly as national data.

Another surprise is that the SSMSI figures are not consistent with those of the DAV. Three differences explain this discrepancy. Firstly, the geographical scope: in addition to metropolitan France and the Drom (overseas departments and regions), the DAV includes the COM (overseas collectivities), which creates a slight difference.

There is also a significant difference in the reference periods. The DAV takes into account homicides committed during the year (which may be recorded later), whereas the SSMSI takes into account homicides recorded during the year (regardless of the year in which they were committed). A final difference is that, in the SSMSI statistics, the course of the proceedings may lead to a change in the classification of the acts committed.

While the figures sent to Eurostat are those of the SSMSI, it is the DAV figures that are referenced in French public debate. To top it all off, the SSMSI figures are diminished by a category that is not taken into account in Eurostat: “intentional assault and battery resulting in death without intent to kill”. Eurostat only considers intentional homicide, which covers the French statistical categories of murder and assassination (premeditated murder).

As a result, it’s hard to make sense of it all. Not to mention the fact that many femicides are not counted (murders of sex workers or women placed in institutions, etc.).

Stay up to date with our newsletter!