> ## Content Index > Fetch the complete content index at: https://www.vitalcitynyc.org/llms.txt > Use this file to discover other available public pages before exploring further. # Did Federal Local Fiscal Recovery Funding Drive America’s Postpandemic Homicide Decline? - URL: https://www.vitalcitynyc.org/postpandemic-homicide-decline-federal-funding/ - Published: 2026-10-28T17:28:50.000Z - Updated: 2026-10-28T17:28:50.000Z - Description: A close look at the evidence says it doesn’t look likely. - Author: Ethan Doshi - Tags: Crime, Federal Government, Economics Between 2019 and 2021, the United States saw the largest increase in homicide since national record-keeping began. Then it reversed. Homicide dipped in 2022, fell sharply in 2023 and 2024, and [kept falling in 2025](https://www.pew.org/en/research-and-analysis/articles/2026/08/06/why-are-homicides-declining-across-the-united-states?ref=vitalcitynyc.org). In the 770 cities I studied in a recent [working paper](https://www.crimrxiv.com/pub/bmqcnp64/release/5?ref=vitalcitynyc.org), the homicide rate climbed from 7.7 per 100,000 residents in 2019 to 11.3 in 2021, then dropped back to 8.3 by 2024\. A rise that took two years was largely undone in three. What’s more, the dramatic fall in homicide has been fairly universal. Between 2021 and 2024, homicide fell in 84% of the 56 U.S. cities with at least 250,000 residents. The remarkable rise and fall has left many people in the field reaching for plausible causal explanations. One theory has become especially popular, and it relates to a federal policy change. In March 2021, Congress passed the American Rescue Plan, which sent $350 billion in flexible recovery money to state and local governments. The first payments reached cities that May, right around the homicide peak. The White House later [counted more than $15 billion in public safety commitments](https://bidenwhitehouse.archives.gov/briefing-room/statements-releases/2024/03/11/the-american-rescue-plan-arp-top-highlights-from-3-years-of-recovery/?ref=vitalcitynyc.org) from the program, and [advocacy groups](https://www.americanprogress.org/article/nationwide-2024-crime-data-demonstrate-the-value-of-violence-prevention-and-local-law-enforcement/?ref=vitalcitynyc.org) would go on to point to the recovery funds as an apparent reason for the decline in homicides. When the Council on Criminal Justice [surveyed 48 crime researchers](https://counciloncj.org/whats-driving-the-homicide-decline/?ref=vitalcitynyc.org) in September 2026, nearly three-quarters said that rebuilding local government — work financed, in large part, with these funds — contributed at least somewhat to the homicide decline, and 4 in 10 called it a top contributor. Because of the broad-based decrease in homicides across so many large U.S. cities, this is a reasonable educated guess. But timing alone can't tell us whether the money did the work, and so I went looking for the footprint this story should leave behind if, in fact, the recovery funding was making its mark. Metropolitan cities (classified by the Department of Housing and Urban Development as cities with populations generally greater than 50,000) received $45.6 billion directly from the Treasury, [allocated according to a formula](https://home.treasury.gov/system/files/136/Allocation-Methodology-for-MetropolitanCities-508A.pdf?ref=vitalcitynyc.org). The rest of the $350 billion went to states, counties and smaller towns, and some states passed part of their share on to cities. Because I compare each city only with others in the same state, that state-level money is largely accounted for, including in states that steered extra funds toward their hardest-hit cities. I look, primarily, at the entirety of the allocation, not just the share spent on policing or violence prevention, because the claim is mainly that the funding as a whole, by shoring up city budgets and services, helped bring homicide down.The median city got about $211 per resident, roughly one-seventh of a typical year's municipal budget, or about three-quarters of what the average city spends on police in a given year. But these amounts varied from city to city — so if the recovery funding was a major driver of the decline in homicides, then it stands to reason that cities that got more recovery funding should have seen larger drops in homicide. To test this proposition, I linked the U.S. Treasury's [spending reports](https://home.treasury.gov/policy-issues/coronavirus/assistance-for-state-local-and-tribal-governments/state-and-local-fiscal-recovery-funds/public-data?ref=vitalcitynyc.org) (compiled local government expenditure accounts which were required to be reported back to the Treasury) for every metropolitan city with more than 50,000 residents to [crime data](https://www.openicpsr.org/openicpsr/project/100707?ref=vitalcitynyc.org) and city budget records, and asked whether more recovery funds predicted a larger decline in homicides among U.S. cities. At first glance, such a relationship exists. Better-funded cities saw much larger drops in homicide. But that pattern turns out to be an artifact of the formula itself. The formula used by the federal government to allocate funding sent more funding to cities with greater poverty, more crowded housing and older housing stock, which are also some of the strongest predictors of urban violence. As a result, the cities that got the most money tended to be the cities hit hardest by the pandemic-era homicide surge, and those same cities went on to see the largest declines. For instance, Philadelphia got $884 per resident, while Dallas got just $267\. Between 2019 and 2021, Philadelphia’s homicide rate jumped by about 60 percent, while Dallas’ barely moved. By 2024, Philadelphia’s rate had fallen below where it was before the pandemic, while Dallas’s was about where it started. That isn’t a guaranteed outcome, but it is common. When a city’s homicide rate jumps far above its normal level, part of the jump often reflects temporary circumstances, and as those circumstances fade, the rate will tend to gravitate back to where it sat before the shock. In fact, the size of a city's homicide increase from 2019 to 2021 explains about half of the differences across cities in the extent to which homicide fell afterward. Statisticians call this [regression to the mean](https://www.vitalcitynyc.org/nypd-zone-strategy-crime-drop-analysis/), a phenomenon [first documented by Sir Francis Galton](https://doi.org/10.2307/2841583?ref=vitalcitynyc.org), who noticed that especially tall or short parents had children whose heights were comparatively closer to the average in the population. In the context of postpandemic homicide reduction, this doesn't mean that the violence fixed itself. But it does tell us that the best predictor of the fall in homicide is the preceding rise, a pattern that is not very consistent with a large role for federal recovery funds. To address this complication, I compared cities located within the same state that were demographically similar in 2019 but which received different amounts of recovery funding because of quirks in how the formula worked. For example, the formula used to allocate funding rewarded older cities that have grown slowly since 1960, which is why places like Royal Oak, Michigan, received more than otherwise similar cities. Before the money arrived, these better- and worse-funded cities had been on similar homicide paths. Once an apples-to-apples comparison is made, the funding advantage disappears. Cities that got more money did not see larger homicide declines, whether funding is measured per resident or as a share of the city's prepandemic budget. Each dot on the chart above (“More money, no bigger decline”) is one of the 453 cities with complete homicide reporting in 2019, 2021 and 2024\. The left panel compares each city's funding with what demographically similar cities in the same state received, and the right panel shows how much the city's homicide rate rose between 2019 and 2021\. The black lines show the average trend, and the gray shading shows the range where that trend could plausibly fall. The companion chart (“Bigger surge, bigger decline”) above shows the pattern. Each dot is a city, and dots lower on the chart are cities where homicide fell more between 2021 and 2024\. In the first chart, cities farther to the right received more money than similar cities in their state. If the extra money had paid off, the dots would slope downward from left to right. Instead, the line is flat. Cities that got a quarter more than their peers saw about the same declines, on average, as cities that got a quarter less. The second chart plots the same declines against how much homicide rose in each city during the pandemic, and here the pattern is unmistakable. The cities with the biggest surges saw the biggest declines, and on average about two-thirds of each city's increase had been undone by 2024. Interestingly, cities that spent their allocated recovery funding quickly did not see larger homicide declines than cities that spent it slowly. Cities that put more funding into public safety likewise did no better. One of the reasons for this is that comparatively little funding was allocated toward public safety initiatives. On average, about two-thirds of reported spending was labeled as replacing lost revenue, and only about 4% carried a public safety label. Among the 148 larger cities that reported any violence intervention spending, the average was $11 per resident. Nor did it matter how much of a city's pandemic revenue loss its federal award replaced. Roughly 3 in 10 cities with a Treasury estimate of their losses received less than they lost, while others received considerably more. If the funding worked by restoring city budgets, and if that restoration yielded crime declines, the cities whose awards fell short of their revenue losses should have seen smaller declines in homicide. They did not. Every large U.S. city received federal recovery funding, and so there is no group of cities that went without. As a result, we cannot know what would have happened if the program had never existed at all. This is not a clean experiment, but instead is a search for a signal in the available data that would be consistent with there having been an important role for federal recovery funding. With this much money landing all at once, more money should have lined up with bigger declines if this funding were a major driver. That signal isn't there. What’s more, the pandemic-era surge in homicides — and subsequent decline — was not seen with other types of criminal activity. In the 429 cities with complete monthly data in the [Real-Time Crime Index](https://crimeindex.org/?ref=vitalcitynyc.org), homicide fell 29% between 2021 and 2024\. Other violent crimes have dropped far less steeply, with violent crime as a whole falling just 4%. Property crimes have not fallen, ending 2024 at the same rate as in 2021, and certain property crimes like auto thefts remain significantly elevated, having risen a further 10% to stand 42% above their 2019 level. Yet [history suggests](https://doi.org/10.1111/j.1745-9133.2010.00661.x?ref=vitalcitynyc.org) that property crimes — and not violence — tend to be linked to business cycles, which suggests another mystery. If federal recovery funding were driving the great homicide decline since 2021, why would it not have had any impact on other types of crimes that are far more common and which have always been far more sensitive to macroeconomic conditions? A look back in time provides further context. In the 1990s, the federal Community Oriented Policing Services program, known as COPS spent about $9 billion hiring local police officers and was [widely credited](https://doi.org/10.1093/acprof:oso/9780195181159.001.0001?ref=vitalcitynyc.org) with that decade's great crime decline. [Later research](https://doi.org/10.1257/aeri.20200792?ref=vitalcitynyc.org) found that COPS grants [did reduce crime](https://doi.org/10.1016/j.jpubeco.2018.12.003?ref=vitalcitynyc.org), but that they [accounted for only a few percentage points](https://doi.org/10.1016/j.jpubeco.2006.05.014?ref=vitalcitynyc.org) of a very large drop in offending. Declines in crime of such a large magnitude [rarely have a single cause](https://doi.org/10.1257/089533004773563485?ref=vitalcitynyc.org). The recent decline in homicides has been remarkably broad. Homicides fell between 2022 and 2024 in 63% of the cities that had any homicides in 2022, and in 81% of those with at least 10\. The pandemic surge was also unusually concentrated in homicide, which rose 46% while overall violent crime rose 11%. Whatever explains the great decline has to fit that pattern. It has to be national, focused on lethal violence and strongest where the surge in homicides was the worst. Given that mean reversion explains a large share of the fall in homicides among large cities, the best available candidates include the recovery of policing and the courts, the reopening of schools and social services, and the burning out of the [retaliatory cycles](https://doi.org/10.1001/jamainternmed.2016.8245?ref=vitalcitynyc.org) through which gun violence spreads. None of this means the Rescue Plan funds were wasted. It simply means that they didn’t seem to drive the postpandemic homicide decline. We’ll have to keep hunting, and rigorously testing theories, to determine what did.