Red Flag Laws

A 'Red Flag' Study Raises the Question of How Often People Who Talk About Mass Shootings Actually Commit Them

The researchers identified 662 cases involving threats to multiple victims, but they concede that it's likely "there are many more threats than completed events."


"Red flag" laws, which authorize "extreme risk protection orders" (ERPOs) prohibiting gun possession by people who are deemed a threat to themselves or others, aim to prevent homicides and suicides. That hope has persuaded legislators in 19 states and the District of Columbia to enact red flag laws, and the Bipartisan Safer Communities Act, which Congress passed last year, included grants that encourage other jurisdictions to follow suit.

It nevertheless remains unclear whether these laws work as advertised, especially when it comes to stopping mass shootings, the main impetus driving such legislation. A recent RAND Corporation analysis found "inconclusive evidence" that red flag laws decrease suicides or violent crime and "no qualifying studies" showing an impact on mass shootings.

A study published last month in Preventive Medicine aims to start filling that evidence gap by analyzing a sample of 6,787 red flag cases from six states. University of Michigan public health researcher April M. Zeoli and her colleagues identified 662 cases involving alleged threats to multiple victims. "While we cannot know how many of the 662 ERPO cases precipitated by a threat would have resulted in a multiple victim/mass shooting event had ERPO laws not been used to prohibit the purchase and possession of firearms," they say, "the study provides evidence at least that ERPOs are being used in six states in a substantial number of these kinds of cases that could have ended in tragedy."

Although Zeoli et al. describe the threats in these cases as "credible," there is reason to doubt that assessment. The research is nevertheless useful to the extent that it illuminates the question of how to weigh the purported benefits of red flag laws against the risk that people will lose their Second Amendment rights because they were mistakenly or maliciously portrayed as dangerous.

The most commonly used definition of mass shootings requires four or more fatalities. Because of data limitations, this study uses a broader definition, encompassing threats involving at least three other people. If a man threatened to kill his wife and her relatives, for example, that would qualify as a potential "multiple victim/mass shooting event." So would a threat to attack a particular person in a location where bystanders would be present or a general threat to "shoot up" a public place.

ERPOs usually are presented as a way to prevent mass shootings. But data from Connecticut and Indiana, the first two states to enact red flag laws, indicate that ERPOs typically are deployed against respondents who are viewed as suicidal. "Earlier studies of the use of these kinds of laws," Zeoli et al. note, "reported that 32% of ERPOs in Connecticut and 21% in Indiana were issued to mitigate a threat of harm to others."

Curiously, the researchers do not say what the percentage was for their sample. Based on their analysis, we know that at least 10 percent of the respondents in ERPO cases were seen as a danger to others. That percentage surely would be higher if threats involving one or two potential victims were included, but how much higher is not clear from the numbers that Zeoli et al. report. It seems likely, given the experience in Connecticut and Indiana, that most of these interventions were aimed at preventing suicides.

Zeoli et al. concede that "we cannot know how many of the 662 ERPO cases precipitated by a multiple victim/mass shooting threat would have resulted in a multiple victim/mass shooting event had ERPO laws not been used to prohibit the purchase and possession of firearms." That is a crucial point if you are trying to assess the benefits of red flag laws and weigh them against the danger they pose to civil liberties. "We are unaware of any literature documenting how many individuals threaten mass shootings but do not go on to attempt or commit them," the researchers say, "although we suspect that there are many more threats than completed events."

That is certainly true when it comes to suicides. In 2020, according to estimates from the Centers for Disease Control and Prevention (CDC), "12.2 million adults seriously thought about suicide," 3.2 million "made a plan," and 1.2 million made an attempt. But the CDC counted a total of 45,979 completed suicides that year. In other words, about 1.4 percent of people who made a plan to commit suicide actually killed themselves.

That breakdown has direct implications for ERPOs justified by threats of self-harm. It suggests that such orders, even if limited to people who show clear signs of suicidal intent, can reasonably be expected to prevent suicides in only a tiny percentage of cases. If the gap between threat and action is similar for people who talk about shooting others, almost none of the 662 potential mass shooters in Zeoli et al.'s sample would have followed through even without judicial intervention.

Judges granted temporary ERPOs—which last from seven days to a month, depending on the state—in 93 percent of these cases. They issued final orders, which generally last up to a year and can be extended, 84 percent of the time. Zeoli et al. see those numbers as evidence that the threats were "credible." But that assessment ignores all the ways in which this process is rigged against respondents.

Initial orders are granted without an adversarial hearing, meaning that respondents have no opportunity to rebut the allegations against them. The standards in the six states that were included in this study (California, Colorado, Connecticut, Florida, Maryland, and Washington) vary. In California, for example, ex parte orders are supposed to be based on a "substantial likelihood" that the respondent "poses a significant danger" to himself or others "in the near future." In Colorado, "the petitioner must establish by a preponderance of the evidence that a person poses a significant risk."

These details matter less than the circumstances in which the petition is considered. The judge hears only from the petitioner, who argues that something terrible will happen soon if the order is not granted. Since no countervailing evidence is presented, it is not surprising that judges almost always grant these orders.

At the next stage, the respondent actually gets a chance to respond, although he generally has no right to court-appointed counsel if he cannot afford a lawyer. Now the petitioner's burden generally is heavier. In most states with red flag laws, he has to present "clear and convincing evidence" of a "significant" risk, although some states say proof by "a preponderance of the evidence" is good enough.

Both requirements are ambiguous, since it is not clear what counts as a "significant" risk. And the judge is deciding whether to maintain the presumptively protective status quo or take the chance that the respondent will in fact kill himself or someone else once his gun rights are restored. Again, it is not surprising that judges overwhelmingly decide not to take that chance. Even if the risk is very small, the prospect of a preventable suicide or homicide is apt to loom much larger than the risk of unjustifiably suspending someone's Second Amendment rights.

Keep in mind that the cases highlighted by Zeoli et al., where the respondents allegedly said something indicating an intent to shoot multiple victims, represent just 10 percent of the total sample. In 90 percent of those 6,787 cases, there was no threat of a mass shooting, credible or not, and many, probably most, did not involve any allegation of a threat to others.

Every time a gunman murders people in a state that already authorizes ERPOs, supporters of red flag laws argue that it only shows this option is underutilized. They say these laws would be more effective if they were used more frequently and aggressively. But that assumes this blunt instrument does a good job of distinguishing between harmless oddballs and people bent on mass murder. Given how bad even trained "mental health professionals" are at predicting violence, that is not a reasonable assumption.

Casting a wider net inevitably magnifies the danger that innocent people will be deprived of the constitutional right to armed self-defense. Instead of defending that tradeoff, supporters of red flag laws pretend it does not exist.