While 150 world leaders are gathered in Paris trying to slow down the human-caused increase in global temperatures, we’re hearing about many of the fearsome possible consequences of our warming planet. We’ve heard about low-lying nations that will disappear under the sea and about still more weather extremes like those the planet saw in 2014, including heat waves across a variety of regions, wildfires in the American west and tropical cyclones in Hawaii.

Here’s one more effect to worry about. In a forthcoming article in Climatic Change, we investigate whether rising temperatures are connected to rising rates of violence.

In short: yes.

Our research found that each degree increase in Celsius (or 1.8 degrees Fahrenheit) is linked to a nearly six percent average increase in homicide rates around the globe. Those rates vary by region. In Africa, each degree increase in global temperature is associated with an almost 18 percent average increase in homicides. In North America, Australia and New Zealand, by contrast, the increase is approximately 3 percent on average. While our study is not the first to examine this link, it is the most comprehensive –and the first to examine it at a global level.

Some technical information on how we studied this:

We collected data on a sample of 57 countries across the globe between 1995 and 2012. We selected these countries because they had enough data on homicide rates, temperatur, and related factors to be able to analyze. Collectively speaking, these countries comprise a sizable portion (over three billion) of the world’s people. The sample countries display temperature increases that are consistent with the expected global land surface temperatures during this period.

We studied homicides as our measure of interpersonal violence for a couple of reasons. First, homicide is one of the most serious crimes, and is largely defined in the same way across countries. Second, homicides are the most likely violent crime to be reported, detected, defined and taken note of locally. We took our homicide data from the United Nations.

We gathered our temperature data from NOAA’s Global Historical Climatology Network and selected average annual temperatures from one or several populous cities for each country, cities being the place where murders are most likely to be concentrated.

We controlled for other factors that have been linked to murder rates: consumer price index, male youth unemployment, infant mortality, the immigrant population (as a percentage of the total population) and war casualties (per 100,000 people). In the appendix to our article, we included other controls beyond these. Our results scarcely changed.

Why would higher temperatures be linked to more violence?

There are two main theories about this connection. First is what’s called the “Routine Activities Theory,” which suggests that when it’s warmer, people have more opportunities for violence. People are more likely to leave their homes and relate to other people, which increases the likelihood of interactions that turn ugly and physical. Climate change, in other words, simply gives people more opportunities to treat each other poorly.

Second is what’s known as the General Affective Aggression Model (GAAM). This argues that warmer conditions make people more irritable, aggressive and violent. GAAM fits in a wider group of heat-aggression theories that specifically focus on explaining why both legal and extralegal violence increase under warmer conditions. Drivers honk their horns more, baseball players commit more fouls, and police may use more force on warm days.

There are other explanations and influences, as well. For instance, some researchers suggest that climate change does or will drive up crime as extreme weather events displace people from their ordinary lives and creates greater competition over resources (jobs, education) in the new host region. Growing poverty and inequality develop as climate change starts chipping away at personal income levels by making food, housing and transportation costs more expensive.

In some countries, the state’s ability to govern will be undermined as governments are unable to protect their citizens from rising sea levels and a growing cost of essentials. In rare cases such processes may prompt civil strife. After decades of progress in limiting world hunger, malnutrition may become more commonplace as crops wither under hotter temperatures and – in some cases – changing precipitation levels. Increases in hunger and growing civil conflict will likely heighten migration.

It’s hard to predict how these indirect effects would work, because each might kick in only after the climate-induced social change exceeds a tipping point. For instance, immigration is typically linked to lower levels of crime. Most immigrants move from their homelands to seek out educational and economic opportunities. As the Syrian conflict shows, however, civil wars brought on – at least in part – by climate change may radically alter the dynamics of immigration.

To be blunt: Catastrophic climate change could boost violence much more than the dramatic results we have already found.

Like every piece of social science research, our findings have limitations. Let’s get technical for a moment to discuss those. We used aggregate annual data on temperature on the country level to measure the effects of climate change on cross-national homicide rates. This choice did not allow us to examine whether seasonal variations in crime, the effects of precipitation, or extreme weather events affect cross-national homicide rates. In addition, aggregating the data by country did not permit us to examine sub-national variations in homicide rates, like those that occur in cities.

Nonetheless, our results show that increases in global temperature, like those being discussed this week in Paris, have larger implications than our policymakers have generally been discussing so far: more murder in particular and more interpersonal violence in general is something to take seriously.

Notably, this increase hurts people in African countries more than people in wealthier ones. In other words, some of the poorest countries will suffer the most.

Kenneth W. Moffett is associate professor of political science at Southern Illinois University Edwardsville. 

Dennis Mares is associate professor of criminal justice at Southern Illinois University Edwardsville.