The Phantom Crime Wave
Why the Data Contradicts the Narrative
If you have turned on a television or scrolled through social media in the last two years, you have likely absorbed a specific, terrifying narrative about American safety. It goes something like this: The border is open, millions are pouring in, and a wave of violent crime, driven by these new arrivals, is crashing over our cities. It is a story of chaos, of a social contract breaking down, and of a direct causal link between immigration and danger.
There is just one problem with this story. According to the most rigorous data available from criminal justice bureaus, state correctional departments, and independent criminologists, it isn’t true.
In fact, the year 2025 is shaping up to be a historic anomaly—not for a surge in crime, but for a collapse in it. Preliminary data from the Council on Criminal Justice indicates that homicides in major U.S. cities fell by over 20% in 2025. If these trends hold, we are witnessing one of the sharpest single-year declines in lethal violence since record-keeping began. This decline is happening precisely in the places—New York, Chicago, Denver—that have absorbed the largest numbers of recent migrants.
So, how do we reconcile the headlines with the trend lines? To understand the real state of American public safety, we have to peel back the rhetoric and look at the “Great Decoupling”—the diverging paths of immigration numbers and crime rates—and examine the structural forces, like demography and geography, that are actually driving these trends.
The “Immigrant Paradox” and the Texas Rosetta Stone
The debate over immigrant criminality often stalls due to a lack of data. Most states do not systematically track the immigration status of everyone they arrest. However, Texas does. The state’s Department of Public Safety (DPS) checks the immigration status of arrestees against Department of Homeland Security (DHS) databases, making Texas the “Rosetta Stone” for understanding the criminality of the undocumented population.
The picture that emerges from the Texas data is unequivocal: undocumented immigrants are significantly less likely to commit crimes than native-born citizens.
According to an analysis of Texas DPS data covering the decade from 2013 to 2022, the homicide conviction rate for illegal immigrants was 2.2 per 100,000 residents. For native-born Americans, it was 3.0 per 100,000. That means an undocumented person in Texas was about 26 percent less likely to be convicted of homicide than a U.S.-born citizen living next door.
When you widen the lens to other crimes, the gap becomes a chasm. Peer-reviewed research published in the Proceedings of the National Academy of Sciences, which also utilized the Texas dataset, found that U.S.-born citizens are over two times more likely to be arrested for violent crimes and four times more likely to be arrested for property crimes than undocumented immigrants.
Criminologists call this the “Immigrant Paradox.” Conventional wisdom suggests that a population with lower education levels and higher poverty rates (like many recent migrants) should have higher crime rates. But immigrants, even those here illegally, display a hyper-aversion to risk. They are self-selected for ambition (the drive to leave home) and operate under the constant Sword of Damocles: the threat of deportation. A native-born citizen caught shoplifting faces a misdemeanor and a fine; an undocumented immigrant faces the total destruction of their life in the United States. This deterrent effect is powerful, visible, and statistically undeniable.
The 2025 Crime Decline: A Structural Correction
If migrants aren’t driving crime up, what is driving it down? The 2025 data paint a picture of a country that is stabilizing after the social tremors of the pandemic.
Homicides in 35 major U.S. cities dropped by 21% in 2025 compared to the previous year. Gun assaults fell by 22%, and carjackings—a crime that became symbolic of pandemic-era lawlessness—plummeted by 43%. This broad-based decline suggests that the spike in violence seen in 2020–2021 was not a “new normal” driven by migration, but a temporary shock driven by the disruption of social institutions (schools, courts, community centers) during COVID-19.
However, the most powerful force suppressing crime isn’t a policy; it’s a birthday.
Criminology’s most ironclad rule is the “age-crime curve.” Violent crime is overwhelmingly a young man’s game, with offending rates peaking in the late teens and early 20s and dropping off sharply as men age into their 30s. The United States is currently benefiting from a massive demographic dividend: the “crime-prone” cohort of males aged 15 to 29 is shrinking relative to the total population.
Furthermore, the behavior of this cohort has changed. Research from the Public Policy Institute of California indicates a “generational plummet” in arrest rates for young adults. The violent felony arrest rate for 18-to-22-year-olds dropped by more than 50% between 1994 and 2019. Today’s youth are spending more time at home, engaged in digital spaces. While this “digital incapacitation” has negative effects on mental health, it has a surprisingly positive effect on public safety: young men who are gaming in their bedrooms are not getting into bar fights or committing street robberies.
The Red State Murder Crisis
While the media focus remains fixated on “migrant crime” in Democrat-run sanctuary cities, the actual geography of violence in America tells a different story.
For over two decades, states that vote Republican (Red states) have consistently had higher murder rates than those that vote Democratic (Blue states). This “Red State Murder Gap” has only widened in recent years. In 2021 and 2022, the murder rate in Red states was 33% higher than in Blue states.
Eight of the ten states with the highest murder rates in the country—including Mississippi, Louisiana, and Alabama—are solid Republican strongholds. Critics often try to explain this away by blaming “blue cities” within those red states, arguing that places like Jackson, MS, or New Orleans, LA, skew the numbers. However, data analysis that removes the largest city from each Red state reveals that the remaining “rural” and “suburban” Red America still has a higher murder rate than Blue states.
Why? The answer likely lies in the “structural” variables of crime. Red states generally have higher rates of poverty, lower educational attainment, and significantly higher rates of gun ownership—all factors that correlate strongly with lethal violence.
Furthermore, the narrative that Blue states have “defunded the police” collapses under scrutiny. Analysis of 2021 spending data shows that Blue states spend considerably more per capita on policing ($453) than Red states ($341). Even when adjusted for the higher cost of living in Blue states, the investment in public safety infrastructure remains robust. The states struggling the most with violence are often those that have chronically underinvested in the social and police institutions required to keep the peace.
The Deportation Placebo
In 2025 and 2026, the political response to crime has heavily favored mass deportation. The logic seems intuitive: if you remove “criminal aliens,” crime should go down. But this strategy is failing to move the needle on national crime rates for a simple reason: the people being deported aren’t the ones committing the serious crimes.
When we analyze the federal data on non-citizens in the criminal justice system, we find that the vast majority are incarcerated for immigration offenses, not violent crimes. In FY 2024, 72.3% of federally sentenced non-citizens were sentenced for immigration violations (like illegal reentry), while only 1.7% were sentenced for firearms offenses.
Moreover, recent data on ICE operations reveals a shift away from targeting serious threats. In late 2025, approximately 73% of individuals booked into ICE custody had no criminal convictions whatsoever. These are “administrative” arrests—people picked up for being in the country without papers—rather than targeted operations against violent offenders.
Removing a dishwasher with a clean record does nothing to lower the murder rate. Even the targeted removal of specific gang members, such as those from the Venezuelan gang Tren de Aragua (TdA), has a limited statistical impact. While TdA is a genuine transnational threat involved in sex trafficking and robbery, its membership numbers are minuscule compared to the millions of peaceful migrants. Law enforcement can and should target TdA, but conflating this specific gang with the broader migrant population is a statistical error that leads to ineffective policy.
Conclusion: The Safety Dividend
The United States is getting safer. This is the reality in 2025 and 2026, despite the anxiety-inducing loops on cable news. This safety is not being delivered by a border wall or a deportation force. It is arriving via structural changes: an aging population, a cultural shift away from alcohol and physical confrontation among youth, and the stabilization of our communities post-pandemic.
The fixation on immigration as a driver of crime is a distraction. It draws resources and attention away from the real drivers of violence—poverty, the proliferation of firearms, and the lack of opportunity in neglected communities. The data tells us that our immigrant neighbors are, on average, among the most law-abiding people in the country. Recognizing this doesn’t require a political stance; it just requires looking at the spreadsheet instead of the screen.
In the end, the “Great Decoupling” is good news. It means that America can continue to be a nation of immigrants without sacrificing its safety. It means we can solve our crime problems without scapegoating our most vulnerable residents. It means that reality, thankfully, is much better than the rhetoric.
Next Week
Napa Valley’s Sea of Sameness:
Why Its Wine Brands are Drowning
* * *
Ted Hall is a Senior Partner Emeritus of McKinsey & Co., and a founder and former chairman of the McKinsey Global Institute. He has spoken and written on economics and the global economy in many venues, including the Wall Street Journal, McKinsey Quarterly, and the World Economic Forum at Davos.
Applying microeconomic principles and employing powerful new research tools, his disciplined, rigorous analysis often results in insights beyond the conventional narrative. In addition to his years of global client service, Ted has decades of experience in agriculture and food as a corporate director, entrepreneur, rancher, farmer, vintner, and restauranteur.
Thanks, Ted. Important analysis and good introduction to your new Substack work.
This was an excellent deconstruction of the "crime is everywhere" narrative. It can fuel highlight reels for days.
Those highlight reels will have a common problem--"so what?"
Age demography and media addiction have put would-be thugs on the sofa? So what?
Are late-stage crime stats are over-interpreted here in the effort to falsify the crime-wave meme?
"That means an undocumented person in Texas was about 26 percent less likely to be convicted of homicide than a U.S.-born citizen living next door."
More accurately stated, an undocumented person *who was apprehended and entered into the justice system* is 26% less likely to be convicted than *a US-born citizen who reached the same point of inclusion into the criminal justice system.*
We will never have an accurate denominator of immigrants from an open border that eventually drew unlawful entrants all the way from the Middle East? Nope. Thank Biden for that.