Over the last 2 years, the COVID-19 pandemic has created significant disruption to our personal and social lives. The lockdowns, designed to curb the spread of the virus, meant that people spent a lot more time at home than they normally would, as well as experiencing more stress and low mental health. Many have wondered about the impact this would have on the prevalence of violence and homicide, as we outlined in part I of this blog series.

In particular, organisations and scholars were concerned that the COVID-19 lockdown would be associated with increases in domestic violence. It is easy to see why; people spend much more time at home, and it was a time of high stress due to, for instance, illness, job loss, or school closures. It isn’t difficult to imagine that tensions might boil over into violence, or even homicide. At the same time, it has been suggested that the occurrence of violence in public spaces, such as ‘urban violence’ or violence associated with organised crime, would be reduced.

When it comes to non-lethal violence, these expectations seem to have been largely supported. Nivette and colleagues (2021) conducted a large-scale study of how the lockdown restrictions impacted crime rates in 27 cities across the world, using (primarily) police data. They found that the lockdown produced a 35% drop in assaults, and a 46% drop in robberies on average, although there was considerable variation between cities in different countries. Similarly, domestic violence service providers reported an increase in calls to helplines.

However, the data on lethal violence, such as homicide, seems to tell a different story. In the study by Nivette et al above, the only crime type that did not seem to decline during lockdown was homicide. This is confirmed by other scholars who show that, during lockdown in the United States, homicide was one of the few crime types that did not seem to decrease. This idea that rates of homicide have been relatively ‘immune’ to the restrictions associated with the COVID-19 lockdown seems to be true of the Netherlands as well – an analysis of the data from the Dutch Homicide Monitor suggests that there were no marked decreases in homicides during 2020.

Why is this? Domestic forms of homicide were relatively stable in 2020 relative to previous years. Between 2010-2019, there were on average 45 domestic homicide cases each year, and in 2020 there were 49 domestic homicide cases. Regarding homicides associated with organised crime, these also remained relatively stable in 2020 relative to the previous decade. There were 19 homicides associated with organised crime in 2020 compared 24 per year on average in the previous decade. However, there was one type of homicide that did see a marked drop in 2020 compared to previous years– namely homicides associated with robberies in private homes. Between 2010-2019 there were 6.3 homicides of this kind every year, but in 2020 there were only 2. This is likely the result of home robberies decreasing in general, given that people were at home for much of the year.

Although the numbers of homicides seemed to be relatively stable, we do observe some changes to other characteristics of homicides during the pandemic year. For instance, at the start of the lockdown in March 2020, 100% of the homicides occurred in private homes, compared to a ‘normal’ March, when about 57% of homicides take place in people’s homes. This pattern seems to suggest that during the lockdown the occurrence of homicides did not diminish, but their locations did shift. Specifically, it seems that homicides that would otherwise occur in public spaces now ‘moved inside’ - into people’s homes. In all this, it is important to keep in mind that data taken from a single year contain more random fluctuations than averages taken over a decade, and as such we must interpret these trends cautiously.

Overall, while the pandemic disrupted many aspects of our personal and social lives, rates of homicide seem to have remained relatively stable during the pandemic – a pattern observed in international as well as Dutch data. This highlights an surprising ‘robustness’ of homicide figures, and an interesting contrast between homicide and forms of non-lethal violence such as assault.