Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
Fact check: Which US cities have seen the largest changes in murder rates since the 1990s?
Executive Summary
Since the 1990s the landscape of U.S. homicides has shifted dramatically: many large cities saw substantial declines through the 1990s into the 2000s, while the 2010s and early 2020s produced varied trajectories with spikes in some places and sustained drops in others. Recent analyses through 2024–2025 show a national rebound toward lower homicide totals compared with the peaks of the early 2020s, but city-level changes remain heterogeneous, with a handful of cities showing large percentage increases year-to-year even as national aggregates fall [1] [2] [3].
1. Big Picture: National declines mask a patchwork of city stories
National data through 2024–2025 depict a clear downward turn in murders after pandemic-era increases, with national measures showing consecutive year-over-year decreases and an estimated 19% decline through August 2025 following earlier falls in 2023–24 [1] [2]. That aggregate improvement, however, conceals substantial local variation: a Council on Criminal Justice snapshot found six cities with homicide increases from 2023 to 2024, while many other cities recorded significant drops, meaning the experience of homicide change depends heavily on which cities you examine [3]. Understanding “largest changes since the 1990s” therefore requires separating long-term declines from short-term spikes.
2. Who gained the most: cities with the largest long-term declines
Studies of the 1990s crime collapse identified major metropolitan declines in homicide rates that reshaped urban safety and demographics, though specific rankings vary by dataset and periodization [4]. Cities that experienced the steepest long-term drops tended to be those with very high 1990s homicide rates—large Northeastern and Rust Belt cities are often cited for the most dramatic percentage declines because they started from high baselines. The available materials note the historical 1990s drop’s big role in urban population and policy change, but do not list a definitive city-by-city percent change since the 1990s in the datasets provided here [4].
3. Who lost ground most: cities with notable recent increases
Recent, year-to-year work shows several cities with sharp recent upticks: the Council on Criminal Justice highlighted Colorado Springs with a 56% increase from 2023 to 2024, and similar single-year spikes occurred in a small subset of cities, even as the 29-city sample saw a net reduction in homicides [3]. These city-level surges often reflect volatile year-to-year counts rather than multi-decade trends, and can be driven by concentrated events, policing changes, and shifts in reporting or classification. Distinguishing multi-decade deterioration from short-term rebounds is essential.
4. Why numbers move: policy, pandemic, and local programs
Researchers attribute both the early-2020s spikes and subsequent declines to a mix of factors. Several analysts point to pandemic disruptions and subsequent reinvestments in community programs and policing tactics as contributors to the reversal toward lower homicides through 2024–25, and community violence intervention programs and policing changes have been singled out by local leaders as producing measurable reductions in some cities [5] [2]. That said, multiple causal stories exist—economic shifts, policing levels, social services, and chance can all matter—and the relationships are neither uniform nor fully resolved.
5. Methodology matters: measuring change since the 1990s is tricky
Comparing “since the 1990s” requires consistent denominators, reporting standards, and city boundaries across decades. Studies that focus on short windows (six months or single-year changes) cannot substitute for a comprehensive, long-term comparison, and the briefs provided underscore that many reporting efforts emphasize recent trends rather than a full 1990s-to-present timeseries [6] [7] [4]. Any authoritative ranking should therefore clarify whether it uses raw counts, per-capita rates, or percentage change from a 1990s baseline and account for population shifts and reporting changes.
6. Conflicting narratives and possible agendas in the data
Different organizations frame the same data with divergent emphases: national-level outlets highlight broad declines to suggest a return to pre-pandemic trends [1] [2], policy research groups emphasize local spikes or recovery patterns to argue for specific interventions [8] [3], and city briefs may stress short-term trends to press local policy agendas [6] [7]. Readers should note these framing choices—an organization focused on immediate policing policy will highlight recent spikes, while national trends-focused sources emphasize multi-year recoveries.
7. Bottom line for someone asking “which cities changed most since the 1990s?”
The evidence shows that the largest long-term percentage declines occurred in cities that started with the highest 1990s murder rates, while recent years produced notable single-year increases in a small number of cities (e.g., Colorado Springs in 2023–24) even as national totals fell in 2024–25. The datasets and summaries provided here point to heterogeneous, time-dependent patterns; a definitive ranked list requires a transparent choice of baseline, rate vs. count, and consistent geographic definitions—choices that materially affect which cities appear to have changed the most [3] [4] [1].