U.S. Gun Violence vs. Peer Nations

Before examining whether gun control policies work, the starting point is the international comparison: how does the U.S. firearm death rate compare to other developed nations? The chart and table below show total firearm deaths per 100,000 across 16 OECD countries.

Sources: CDC WONDER (U.S.); WHO Global Health Observatory; IHME Global Burden of Disease; national statistical agencies. Data years vary by country.

The U.S. rate is about 6–8 times the OECD average, roughly 13 times the UK rate, and over 400 times Japan’s rate. Among G7 nations, no country comes close.

A crucial and frequently omitted fact: the majority of U.S. firearm deaths are suicides, not homicides.

In 2023, of about 46,728 total U.S. gun deaths: suicides accounted for ~27,300 (58%), homicides for ~17,900 (38%), and unintentional/law enforcement for the remainder (~4%). In 2024 (provisional CDC data), total gun deaths fell to about 44,000, driven by a 14% decline in homicides. But gun suicides reached a record high, and firearms were used in more than half of all suicides for the first time in at least 25 years.

Why this matters for policy: The interventions that reduce gun suicide (safe storage laws, waiting periods, means restriction) are substantially different from those that reduce gun homicide (background checks, anti-trafficking enforcement, focused deterrence). Treating “gun violence” as monolithic obscures this.

The U.S. firearm homicide rate (~6.2/100k in 2022) is where the international gap is most stark. This is roughly 23 times Germany’s rate, 100 times the UK’s, and over 200 times Japan’s. No other developed nation is in the same category.

The U.S. non-firearm homicide rate (~2.5/100k) is also 2–3 times higher than most Western European peers. America is more violent overall, with or without guns. But the firearms gap (20–100x) dwarfs the non-firearm gap (2–3x). This pattern is consistent with firearms amplifying the lethality of altercations that would otherwise result in non-fatal violence.

The 2020–2021 spike (about +35% single-year increase, the largest in decades) was likely driven by COVID-era social disruption, reduced policing, and court shutdowns. It has since partially reversed: by 2023, the rate had fallen to about 5.6/100k, a 16% decline from the 2021 peak, though still well above the pre-pandemic baseline of ~4.4. Provisional 2024 data shows a further decline in gun homicides.

Firearms are the leading cause of death for Americans aged 1–19, surpassing motor vehicles in 2020 and remaining so for four consecutive years. In 2024, gun deaths among those under 18 fell nearly 14%.

U.S. gun homicide is not evenly distributed. It is concentrated by race, age, sex, and geography:

• Race: Black Americans experience firearm homicide at about 26/100k (2021), roughly 13 times the rate for white Americans (~2/100k)
• Age and sex: Men account for ~86% of firearm homicide victims; peak victimization age is 20–34
• Geography: About 50% of all U.S. gun homicides occur in roughly 127 counties (4% of counties). Within cities, violence concentrates in small networks and neighborhoods

This concentration has direct implications: rural and suburban areas of the U.S. have gun homicide rates comparable to many peer nations. The national average is driven overwhelmingly by concentrated urban violence linked to poverty, inequality, and gang dynamics.

U.S. firearm suicide rate: about 7.5/100k, roughly 10 times the European average. In the U.S., about 54–56% of all suicides are by firearm, compared to <1% in the UK and Japan, ~5% in Germany, and ~16% in Canada. Firearms have a case fatality rate of about 90%, compared to ~5% for the most common alternative methods.

Gun suicides reached a record high in 2023 (27,300 deaths) and continued rising in 2024, constituting the highest share of gun deaths in at least 25 years. While homicides declined, suicides did not.

This is the central question for gun suicide policy: if firearms are less available, do suicidal individuals switch to other lethal methods, or does overall suicide decline?

The evidence strongly favors means restriction over substitution:

Israeli military study (Lubin et al., 2010): When the IDF restricted soldiers from taking guns home on weekends, weekend suicide rates dropped 57% with no compensatory increase in non-firearm suicides. Overall military suicide fell 40%.

Australian buyback (Chapman et al., 2006): After the 1996 National Firearms Agreement, firearm suicide rates declined significantly. Non-firearm suicide rates showed no significant compensatory increase. People did not switch methods at equivalent rates.

Multiple meta-analyses (Harvard Injury Control Research Center) find that firearm availability is associated with higher total suicide rates. The effect is not limited to firearm suicides. Reducing availability reduces overall suicide, with no offsetting rise in other methods.

The mechanism is the “crisis window”: suicidal crises are often brief (minutes to hours). If the most immediately lethal means is unavailable during that window, many individuals survive the crisis and do not later die by suicide. Firearms collapse the time between impulse and death to near-zero, eliminating the possibility of intervention or reconsideration.

Using a consistent definition (4+ killed, excluding shooter, in a public place, not gang/domestic/terrorism), the U.S. is a dramatic outlier. A study of 36 high-income countries (2000–2022) found the U.S. had 109 public mass shootings, 76% of the total among all these countries combined. The U.S. accounts for roughly 5% of world population.

Per capita, Americans are roughly 4 times more likely to experience a mass shooting than Canadians, 11 times more than Britons, and 20 times more than Australians.

Definitional caveat: mass shooting definitions vary widely. The Gun Violence Archive (4+ shot, any context) records ~600 per year. The FBI (4+ killed, public place) records far fewer. International comparisons require consistent definitions to be meaningful. All figures above use the stricter definition.

Australia’s response is the most studied: firearm homicide rates declined about 42–72% faster than pre-existing trends, and mass shootings effectively stopped for over two decades. Critical caveat: homicide was already declining pre-1996, and researchers debate how much of the accelerated decline is attributable to the NFA versus continuation of pre-existing trends.

The United States has 120.5 civilian firearms per 100 residents, more guns than people, and by far the highest rate in the world. Yemen (a war zone) is second at 52.8. The total civilian gun stock: about 393 million as of 2017 (Small Arms Survey), likely 400–450 million by 2025 given pandemic-era sales surges.

Household ownership: about 42–45% of U.S. households report having a gun (Gallup/Pew, 2023–2025). This rate has been relatively stable for a decade, even as total gun numbers have risen, suggesting existing owners are acquiring more guns rather than new households joining.

Cross-nationally, gun ownership rates correlate strongly with firearm death rates (r ≈ 0.7–0.8 for homicide among developed nations). Within U.S. states, the correlation is similarly strong (r ≈ 0.62). This is correlation, not proven causation, and confounders (poverty, inequality) matter, but the consistency across studies and methodologies is notable.

In U.S. homicides where weapon type is known: handguns account for about 53–64%, rifles (including “assault-style”) for about 3–4%, shotguns for 2–3%, and the remainder is unspecified.

Policy implication: Banning “assault weapons” targets roughly 3–4% of firearm homicides. These weapons account for a disproportionate share of mass shooting fatalities, but not everyday gun violence. If the goal is reducing overall firearm homicide, the data points to handguns as the numerically dominant issue.

The presence of a firearm in a domestic violence situation increases the risk of homicide for the victim by about 500%. About 55% of women killed by intimate partners are killed with firearms. Women in the U.S. are 21 times more likely to be killed by firearms than women in other high-income countries.

Sources: Saltzman et al., CDC; Campbell JC, intimate partner homicide studies; Everytown for Gun Safety (advocacy organization, but citing CDC/FBI data)

After Port Arthur, Australia banned semi-automatic rifles and shotguns and conducted a mandatory buyback of about 660,000 firearms (~20% of the civilian stock). Results:

• Firearm homicide rate declined about 42–72% over the following decade
• Firearm suicide rate declined about 57–65%
• Zero public mass shootings for 22 years (1996–2018)
• Non-firearm suicide showed no significant compensatory increase

Critical caveat: Firearm homicide and suicide were already declining before 1996. Chapman et al. (2006) found the decline was significantly steeper post-NFA than the pre-existing trend predicted. Leigh & Neill (2010) found the evidence stronger for suicide and mass shootings than for homicide specifically.

The AWB (1994–2004) banned manufacture of semi-automatic firearms with certain features and capped magazine capacity at 10 rounds. Findings:

RAND Corporation and multiple studies found the AWB had little to no measurable effect on overall firearm homicide. This is largely because the banned weapons were used in a small fraction of crimes (~3–4% of homicides) and were easily substituted. However, DiMaggio et al. (2019) found mass shooting fatalities were lower during the ban period and increased roughly 183% after its expiration.

The most rigorous systematic review of U.S. gun policy is the RAND Corporation’s “Science of Gun Policy” (3rd edition, 2023). Their evidence ratings:

Source: RAND Corporation, “The Science of Gun Policy,” 3rd edition, 2023 (rand.org/research/gun-policy)

RAND’s is the most methodologically rigorous systematic review available. Their “supportive” rating means multiple well-designed studies pointing in the same direction. Their “inconclusive” rating means the evidence is mixed, not that the policy doesn’t work. It means we don’t know.

The international comparison is stark: the U.S. firearm homicide rate is 20–100x higher than comparable wealthy nations. This magnitude of difference is unlikely to be explained entirely by confounders.

Means restriction works for suicide, the Israeli military study and Australian buyback provide strong natural-experiment evidence. The domestic violence nexus (500% increased fatality risk from a gun in the home) represents a specific, addressable intersection. RAND’s systematic review finds supportive evidence for multiple specific policies. The 2020–21 spike following record gun sales is consistent with (though not definitive proof of) a causal relationship.

Defensive gun use is real and potentially large. The NCVS (government survey) finds about 60,000–80,000 DGUs per year. Kleck’s survey found 2.5 million. Harvard researchers estimate the truth may be 300,000–1 million. The wide range is methodological, not political, both sides cherry-pick the estimate that suits them, but the evidence that DGU occurs at meaningful scale is consistent.

Most gun owners will never use their weapon violently. About 100 million gun owners commit extremely few crimes with their firearms. Policy should weigh the burden on the law-abiding majority against the actions of a small minority.

Enforcement of existing laws is incomplete. Many convicted violent felons obtained firearms illegally, not through background-checkable channels. The “gun show loophole” is already a crime at the federal level for straw purchases; the question is enforcement.

Urban violence is concentrated and addressable without broader restrictions. Focused deterrence programs (Operation Ceasefire, violence interrupter models) target the specific networks responsible for most gun homicide and show effectiveness. These require no new gun laws.

International comparisons are confounded. The U.S. non-firearm homicide rate is also elevated versus peers (2–3x). This suggests inequality, poverty, drug policy, and incarceration drive violence independent of gun access. Switzerland has high gun ownership and low homicide. The correlation is moderated by other variables.

Constitutional framework: Heller (2008) and Bruen (2022) established an individual right to firearm possession. Policy must operate within these constraints, and Bruen significantly narrowed permissible regulations.

Genuine bipartisan research support exists for: child access prevention / safe storage laws, crisis intervention / red flag laws, background check enforcement and expansion, waiting periods, and focused deterrence programs in high-violence neighborhoods. These policies have “supportive” evidence in RAND’s schema and are supported by majorities across party lines in polling.