Skip to main content
ABC News
What We Know About Crises And Domestic Violence — And What That Could Mean For COVID-19

The novel coronavirus pandemic, in addition to its direct impact on people’s health, has unleashed a perfect storm of conditions that may increase the rate of domestic violence. Social isolation, for example, is one of the most common tactics used by abusers to distance survivors from their support networks, and now physical isolation is government-sanctioned. Unemployment claims are hitting historic highs, as are levels of economic anxiety; both of these circumstances are linked to a higher incidence of domestic violence. Firearm ownership is tied to a greater chance of domestic homicide, and gun sales in the U.S. rocketed in March, with reports that many of those sales were to first-time gun buyers. National hotlines even have to devise strategies to combat abusers’ weaponizing COVID-19 itself to terrorize survivors — e.g., threatening to infect someone with COVID-19 or hiding cleaning supplies so the survivor cannot access them.

The relationship between these factors and domestic violence is well researched and well documented, as many other articles have discussed. But tracking the pandemic’s actual effects on domestic violence is nearly impossible. Many people who experience abuse don’t report it through official channels. Stigma and fear of retribution are just a few of the reasons someone wouldn’t contact the police. Because of that, we don’t have good data showing domestic violence incidence in normal times, let alone now. In fact, we’re getting conflicting signals at the moment — while the National Domestic Violence Hotline continued to see its average of 1,800 to 2,000 contacts a day through March and April, New York Gov. Andrew Cuomo announced on Friday that reports of domestic violence rose in the state during that same time.

So while we have solid, empirically derived reasons to believe that COVID-19 might increase the rate of domestic violence across the country, we don’t have the numbers to back that claim. That can make the issue feel less real to the public and to policymakers, even though, due to the impact of COVID-19, the increase in intimate partner violence (IPV) — i.e., physical, psychological or sexual harm in an intimate relationship — may be significant.

How significant? There’s no perfect case study to draw from, but we can make some reasonable conclusions by looking at research on IPV1 and past disasters with some similarities to the COVID-19 crisis.

Economic anxiety and joblessness

Economic hardship, employment instability, unemployment and perceptions of economic strain have all been linked to IPV, according to Jennifer Copp, a professor at Florida State University’s College of Criminology and Criminal Justice. Domestic violence is about power and control, and when your job, finances and livelihood are all up in the air, abuse becomes a place where people seek to regain that sense of control.

Indeed, a 2001 research report from the National Institute of Justice found that rates of violence in couples experiencing high levels of financial stress were 3 1/2 times the rates in couples with low levels of stress — rising to 9.5 percent from 2.7 percent. According to a working paper by the Center for Global Development, the effects of unemployment on IPV vary across populations due to cultural differences, but in the U.S., the NIJ found in 2009 that periods of unemployment by a man in a heterosexual relationship are correlated with significantly elevated rates of abuse. That paper didn’t show a causal link but found that couples in which the man was employed had a 4.7 percent prevalence of abuse, while that figure was 7.5 percent for couples in which the man experienced one period of unemployment and 12.3 percent for couples in which the man experienced two or more periods of unemployment.

Economically, the most recent circumstances analogous to the current pandemic may be those caused by the Great Recession, when the overall unemployment rate doubled from 5 percent in December 2007 to 10 percent in October 2009 and the net worth of households and nonprofits fell over 20 percent, a loss of some $14 trillion, in a year and a half. A 2016 study published in “Demography” on economic indicators and IPV during the Great Recession found that an increase in the unemployment rate was correlated with a small but significant change in IPV — a 50 percent increase in the unemployment rate over the prior 12-month period, for example, was associated with a 10 to 12 percent increase in the prevalence of physically violent or controlling behavior in intimate relationships.

Of course, a correlation between economic hardship and IPV may be due to other factors that are linked to greater hardship or job loss and IPV — such as mental illness, stress or substance abuse. Establishing a causal relationship is much more difficult since economic hardship typically goes hand in hand with many other variables. Yet coupled with what we know about the interplay of economic precarity and domestic violence, the economic fallout of COVID-19 presents perhaps one of the greatest risks in terms of increasing abuse.

Natural disasters

It’s more difficult to draw comparisons between behavior after natural disasters and the current COVID-19 pandemic. Generally, though, women have a much more precarious social and economic standing than men, and that vulnerability only increases during a disaster.2 Normal resources and social support that usually curb abuse can disappear entirely depending on the type of disaster — hotlines, access to legal help and shelter availability can all be disrupted. On top of that, IPV increases in the aftermath of disasters, but many factors are at play that don’t apply as much to COVID-19 — most notably displacement, which is not yet a major effect of the pandemic.3

While research has shown that economic downturns are accompanied by physical, mental and emotional violence, the intersection of gender-based violence and disasters is only a recent area of focus. It’s unclear whether increased violence is a simple matter of proximity or of stress or some other unknown variable.

Still, it’s worth looking at some examples that may hold lessons for dealing with the current pandemic, as long as we keep in mind that the circumstances of the COVID-19 crisis are quite different. In 2011, a 6.2 magnitude earthquake hit Christchurch, New Zealand, causing 185 deaths and thousands of injuries in addition to severe structural damage. Two weeks after the earthquake, area police said that reports of domestic violence had risen by a fifth. Afterward, police cited business closures as a reason why more people were drinking at home, aggravating the risk of domestic violence. Sexual violence rose after Hurricane Katrina, as many people were forced into temporary shelters. In the first six months after Hurricanes Katrina and Rita, nearly 40 percent of sexual assaults were committed by strangers (versus the 19.5 percent committed by strangers in all reported cases of sexual assault in the U.S.); 31 percent of these assaults took place in shelters or evacuation sites. For people whose lives were highly disrupted by the 1989 Exxon Valdez oil spill in Alaska, there was a nearly threefold increase in the likelihood of experiencing domestic violence afterward compared with people who experienced no disruption.

One of the most striking findings to come out of research on domestic violence in the wake of a natural disaster is that, separate from the rate of abuse changing, the severity of abuse can increase as well. In the aforementioned study of sexual assaults after Hurricanes Katrina and Rita, researchers found that in the early aftermath of the disaster, “rapes appeared to be more brutal, often involving multiple offenders,” and then returned to a more typical pattern a year later. After the 1997 Red River Valley flood, which displaced over 50,000 people in Grand Forks, North Dakota, and East Grand Forks, Minnesota, and damaged nearly all homes in the area, referrals to domestic violence shelters from emergency rooms increased, possibly pointing to a rise in severity of physical abuse. Even now, Katie Ray-Jones, CEO of the National Domestic Violence Hotline, said that survivors contacting them have seen an escalation in abuse tactics — including one person who said her partner pulled out a gun to prevent her from leaving the house, something that had never happened before.

In some ways at least, the COVID-19 pandemic is less problematic regarding IPV than other disasters: Most hotlines are still operational, and while shelter availability varies depending on locale, there hasn’t been a natural event that wipes out the structures. Many domestic violence services are still operating and are able to offer support, although survivors may be less willing to go to one for fear of catching COVID-19. Nathaniel Fields, CEO of Urban Resource Institute in New York, an organization that provides resources and shelter for survivors of domestic violence, said that their priority right now is communication — when everyone, including your friends, neighbors and the government, is telling you to stay home, it can be hard to leave an abusive situation. Survivors find themselves doing the mental calculus of whether it’s safer to stay home with the devil you know or risk the COVID-19 you don’t.

But those silver linings aside, the COVID-19 pandemic is still likely meaningfully increasing the rate of domestic violence. Past downturns and disasters doubled or even tripled those rates. Those were each unique circumstances, making it hard to say if something similar will happen now. Indeed, Claire Renzetti, a professor of sociology at the University of Kentucky, didn’t offer any speculation as to the scale of an increase. But she said that knowing what we know from previous experiences that haven’t been as bad as this one, our current economic climate will likely increase IPV.

Moreover, in normal times, domestic violence doesn’t affect all populations equally, and these inequities are likely to be exacerbated during the pandemic. Some studies have found that low-income, black and Hispanic populations are more likely to experience IPV than white populations or those with post-secondary education. Those are the same groups bearing a disproportionate share of the direct health and economic effects of the COVID-19 pandemic.

We probably won’t have the data to show much of this for years, if ever. Data collection during disasters suffers from logistical difficulties, and quantitative insights on domestic violence are chronically underreported. But that doesn’t mean it’s not happening.

Laura Bronner contributed research.

Help is available. If you or someone you know is experiencing domestic violence, reach out to the confidential National Domestic Violence Hotline. (Call 1−800−799−7233 or TTY 1−800−787−3224 or chat online at

Why does COVID-19 make some people sicker than others? l FiveThirtyEight


  1. IPV, domestic violence and gender-based violence are used interchangeably in this article. Domestic violence, however, can include abuse within any part of the family unit, though this would usually be further specified as “child abuse” or “elder abuse.”

  2. In general, this article is referring to violence in heterosexual relationships in which a man is abusive to a woman. Even considering underreporting from other genders, women experience the bulk of IPV. Underreporting and scarce focused research make it difficult to draw conclusions about changes in domestic violence in relationships in which heterosexual men are abused, in nonheterosexual relationships or relationships involving trans people.

  3. But it could be if people are forced to move to find work or evicted from their homes on a large scale.

Jasmine Mithani was a visual journalist for FiveThirtyEight.