The US Murder Rate Debate: Red States, Blue Cities, and the Data They Both Ignore

U.S. homicide rates surged during the pandemic, reaching an estimated 6.4–7.4 per 100,000 in 2021 (commonly cited as 6.8) — the highest level since 1998, though still well below the 9.8 peak recorded in 1991 ^[12]. The spike has been linked to pandemic-era economic disruption, social isolation, and strained police-community relations.

The decline since has been sharp. A Council on Criminal Justice analysis of 35 U.S. cities found homicides were 21% lower for full-year 2025 than 2024, with the decline accelerating from 18% in the first half of the year to 24% in the second half ^[7]. The post-pandemic homicide spike has substantially reversed, even as the political debate over its causes continues.

Louisiana and Mississippi have repeatedly ranked among the states with the highest murder rates in recent years, underscoring the geographic concentration of American violence that fuels the red-blue debate below ^[3][4].

The "Red State Murder Problem" was popularized by the centrist think tank Third Way, which compared homicide rates in states won by Donald Trump in 2020 against those won by Joe Biden using CDC data. The gap is substantial: red states averaged 9.0 homicides per 100,000 to blue states' 6.8 in 2021, and 8.5 to 6.4 in 2022 — about 33% higher both years. Third Way says red-state rates have exceeded blue-state rates every year this century ^[3].

Some of the gap tracks with resources and demographics rather than ideology alone. Third Way found that, in 2020, the 25 largest Democrat-run cities spent 38% more per capita on policing than the 25 largest Republican-run cities ^[4]. Separately, a peer-reviewed study in PLOS Medicine found that county-level "institutional social capital" — trust in governing institutions — along with social mobility, income inequality, and public-welfare spending, showed robust statistical relationships with local gun-homicide rates ^[9]. Red states tend to show lower institutional trust and higher poverty — the same factors research links most strongly to violence, independent of partisan control.

The most common rebuttal to the state-level data is that red states' high rates are driven solely by their large, Democratic-led cities. To test this, Third Way removed homicides occurring in the county containing the largest city from 22 of 25 red states (three — North Dakota, Wyoming, and Idaho — were excluded because their largest cities were too small), while leaving blue-state data untouched ^[3].

Even with this advantage given exclusively to red-state data, the red-state rate remained 20% higher than the blue-state rate in 2021 and 16% higher in 2022. After Jackson was removed from Mississippi's count, the state still had the second-highest murder rate in the country in both years. Alabama rose from third to first place in the 2022 rankings once its largest city was excluded ^[3]. Red-state violence extends well beyond the big cities critics blame it on.

Third Way's red/blue classification is also a simplification: it is based solely on each state's 2020 presidential vote, ignoring swing states, split state governments, and the fact that crime policy is set at multiple levels of government.

The Heritage Foundation argues the city, not the state, is the more honest unit of analysis: as of mid-2022, 27 of the 30 U.S. cities with the highest murder rates had Democratic mayors. Heritage ties this to "progressive" or "rogue" prosecutors it says have weakened enforcement in many of these cities ^[5][6].

But 2024 data complicate that narrative: of 24 major cities tracked by Rochester Institute of Technology researchers, 70.8% saw homicide decreases from 2023 to 2024. New Orleans fell 32.4%, Washington D.C. 30.7%, Oakland 30.9%, Detroit 18.3%, Hartford 38.9%, and Greensboro 43.1% — all under Democratic city leadership ^[8].

Heritage's point about Democratic mayors is also partly a base-rate artifact: even among the 100 largest U.S. cities, Democratic mayors outnumbered Republican mayors 63 to 26 as of 2024, and that imbalance grows sharper among the very largest cities ^[11]. Attributing urban crime to Democratic governance alone comes close to attributing it to being a large American city.

The Manhattan Institute offers one of the more rigorous resolutions of the debate. Comparing counties and states with similar demographics, its researchers found that once they controlled for race, age, gender, urbanization, and per-capita income, "there is no relationship between the Trump share of the vote and homicides at the county or at the state level" ^[10]. Partisan affiliation looks like it predicts murder mainly because it correlates with deeper demographic and economic differences.

This squares with the PLOS Medicine finding above: structural conditions — poverty, inequality, social isolation, and institutional trust — predict violence far better than which party governs a state or city ^[9].

Both narratives also risk the "ecological fallacy" — drawing conclusions about individual behavior or policy effects from aggregate, state- or city-wide data. A state's murder rate alone cannot tell you which policy caused which killing, and a city's Democratic mayor alone cannot tell you whether its prosecutors are "soft on crime."

A Brookings Institution analysis of the 2024 election found that false and misleading claims about crime — including exaggerated statistics and selectively framed statistics about immigrants and crime — shaped how voters perceived candidates on public safety ^[13].

Gallup polling shows a long-running gap between perception and data: as of its most recent annual crime survey, 49% of Americans believed crime was increasing nationally, down sharply from 64% a year earlier and 77% two years earlier — even as FBI data showed violent crime falling roughly 4.5% and murder falling nearly 15% in 2024 ^[13]. Americans have consistently overestimated how dangerous the country is becoming, even during years of measurable decline.

The Global Disparity: The U.S. homicide rate is more than three times Canada's and roughly 28 times Japan's — far above its wealthy-nation peers, driven by firearms proliferation and structural inequality, not partisan governance ^[1][2][12].

The Political Scale Trick: Red states have higher aggregate murder rates. Blue cities have higher per-capita urban rates. Both are true. Neither is causal on its own — the scale of analysis determines the conclusion ^[3][5].

The Actual Predictors: Poverty, inequality, institutional trust, and social mobility predict homicide far better than party affiliation. When these are controlled for, the partisan correlation vanishes ^[9][10].

The Decline Is Real: The post-pandemic homicide spike — which peaked around 2021 — had substantially reversed by 2025, independent of which party held power in any given state or city ^[7][12].

The Disinformation Threat: Americans persistently overestimate rising crime even during periods of measurable decline, a perception gap that distorts both policy debates and election narratives ^[13].

The 2025 return toward pre-pandemic homicide levels is a significant milestone. But closing the gap with the safety levels seen in Europe and Asia requires addressing the structural conditions — poverty, inequality, firearms access, and institutional distrust — that produce American violence in the first place. Simply blaming "red states" or "blue cities" is not analysis. It is campaign advertising.