Skip to main content
ABC News
Latinos In Three Cities Are Reporting Fewer Crimes Since Trump Took Office

Latinos in several big cities across the country appear to be reporting fewer crimes this year than last, a pattern that some experts say could be evidence that President Trump’s tough stance on immigration is sowing mistrust of law enforcement officers among some immigrants and their families.

In recent months, police chiefs in Los Angeles and Houston have said that reports by Latinos of certain types of crime are down in their respective cities. Both chiefs blamed the declines on heightened fear of deportation among undocumented immigrants, a majority of whom are Latino. They say the trend is concerning because police departments rely on members of the community — regardless of their legal status — to report crimes when they occur.

“When you see this type of data, and what looks like the beginnings of people not reporting crime, we should all be concerned,” Houston Police Chief Art Acevedo said at a press conference last month, according to The Houston Chronicle. “A person that rapes or violently attacks or robs an undocumented immigrant is somebody that is going to harm a natural-born citizen or lawful resident.”

Most U.S. Latinos, of course, are not immigrants, and most immigrants are not undocumented. But a majority of undocumented immigrants are Latino, and many U.S.-born Latinos have friends or family members who are undocumented. As a result, Trump’s immigration policies are likely to disproportionately affect Latinos. And because the Latino immigrant population is large and spread across the country, the effects of immigration policies are likely to show up more clearly in data on Latinos than in data on smaller immigrant communities that might also be affected.

There are no comprehensive national statistics for crime reports broken down by race, ethnicity or immigration status. And many individual cities either don’t collect that data or refuse to release it. But a FiveThirtyEight analysis of data from three other major cities — Dallas, Denver and Philadelphia — supports the notion that immigrants, or Latinos more generally, could be reporting fewer crimes since Trump took office. Two of those cities — Denver and Philadelphia — have seen a marked decline in crime reports1 from Latinos relative to those from non-Latinos in the first three months of 2017. Dallas, meanwhile, has seen a decline in Latino crime reports in immigrant-heavy neighborhoods.

President Trump speaks at the National Peace Officers’ Memorial Service in Washington, DC, this week.

getty images

It’s impossible to say for sure at this point whether those patterns are driven by Trump’s policies — or even whether the patterns themselves will last as more data becomes available. Some experts are skeptical. Jessica Vaughan, director of policy studies at the Center for Immigration Studies, an advocacy organization that supports stricter limits on immigration, accused police chiefs of placing political objectives above the data. “It looks to me like they’re trying to enter the political debate,” Vaughan said in an interview. “It’s very difficult to ascertain whether crime reporting has changed nationally.”

There is little question, however, that Trump has stepped up federal enforcement of immigration laws. He is also trying to change the way local authorities deal with undocumented immigrants. Trump has greatly expanded the list of offenses that can lead to swift deportation and has tried (so far unsuccessfully) to withhold federal funding from so-called sanctuary cities that refuse to cooperate with immigration authorities under certain circumstances. These policies have increased immigration arrests to their highest level in three years.

Immigrant advocacy organizations say such policies are already deterring both undocumented immigrants and their families from reporting crimes. “The immigrant community feels like they are under attack,” said Amy Fischer, policy director at RAICES, a Texas-based immigrant advocacy organization. Fischer said her organization had seen immigrants becoming more cautious about engaging with the criminal justice system. “Victims of crime are scared to report crimes or go to court.”

Asked whether the new administration’s policies could be affecting Latinos’ willingness to report crimes, U.S. Immigration and Customs Enforcement spokeswoman Dani Bennett said in a written statement that ICE “take[s] into consideration if an individual is the immediate victim or witness to a crime in determining whether to take enforcement action.” She pointed to a special immigrant status called a U-visa, which allows non-citizens — including those in the country illegally — to report crimes without fear of deportation, although law enforcement agencies can award such visas inconsistently. ICE declined to comment further.

In an effort to see whether the trends reported in Houston and Los Angeles were part of a broader pattern, I contacted the police departments in more than a dozen cities with large undocumented immigrant populations2 to gather information on crime reports through the first three months of 2017. Only two cities — Denver and Philadelphia — provided relevant data broken down by race or ethnicity. I obtained similar data for Dallas from the city’s open data portal. (Neither Houston nor Los Angeles provided data, despite their police chiefs’ previous comments on this issue.)

Data from all three cities is consistent with the idea that immigrants have become less likely to report crimes. In both Denver and Philadelphia, crime reports among Latinos — who make up 45 percent of all immigrants nationally and a majority of undocumented immigrants, specifically — fell relative to non-Latinos in the first three months of the year. In Denver, crime reports among non-Latinos increased 3.6 percent in the first three months of 2017 compared with the same period last year; among Latinos, crime reports fell 12 percent. In Philadelphia, crime reports by non-Latinos declined by 1.0 percent, while they fell 4.3 percent among Latinos.3 Notably, the decline in crime reports from Latinos appeared to cut across several types of crimes, whereas the Houston and Los Angeles police chiefs highlighted declines in sexual assault and, in the case of Los Angeles, domestic violence. Neither Denver nor Philadelphia provided data sufficient to evaluate trends in domestic violence and sexual assault.

The story is a bit different in Dallas, where data shows no overall decline in crime reports among Latinos relative to non-Latinos.4 (One possible reason for the difference: Crime data released by the Dallas Police Department excludes “sexually oriented offenses” and incidents of domestic violence,5 the crimes identified by Houston and Los Angeles as experiencing the biggest declines in Latino reporting.) But crime reports among Latinos — or, more precisely, reports of crimes in which the alleged victims are Latinos — do seem to be falling disproportionately in immigrant-heavy neighborhoods. Specifically, since Trump’s election, there has been a statistically significant correlation between the share of a neighborhood’s residents who are non-citizens and the decline in crime reports among Latinos in early 2017. (Roughly half of non-citizens are in the country illegally, according to estimates from the Pew Research Center.) In other words, the more non-citizens in a ZIP code, the greater the dropoff in crime reports among Latinos.6 The five ZIP codes with the highest shares of non-citizens have seen an average decline in crime reports among Latinos of more than 16 percent in the first three months of Trump’s presidency. That pattern suggests that non-citizen immigrants are the ones driving the declines in crime reporting in those neighborhoods.7

It’s hard to know for sure whether fear of deportation is behind the decline in crime reporting. But the timing of the decrease suggests that it could be related to Trump’s more aggressive immigration policies. Before this year, crime reporting among Latinos in Philadelphia was, if anything, rising relative to reports from non-Latinos.8 Similarly, in Dallas, the correlation between Latino crime reports and the non-citizen share of each ZIP code emerged only after Trump’s election and has grown stronger since he took office.9 (Earlier data wasn’t available for Denver.)

There is historical precedent for immigration policies’ affecting reporting among undocumented immigrants. Salt Lake City encountered similar issues when the state legislature debated a bill that would require local law enforcement to detain unauthorized immigrants, recalled Chris Burbank, the police chief at the time. “What we found was, not surprisingly, undocumented individuals would be less likely to report crimes,” he said in an interview. “We had children go missing … but their parents wouldn’t call the police. We heard about them from neighbors.”

Salt Lake City’s experience could carry a lesson for the current immigration debate: The mere threat of tougher enforcement could affect immigrants’ willingness to report crimes, even before authorities change their behavior. Burbank, who drew national attention for his opposition to using local police to enforce immigration laws, said that while the bill was being debated, he told his officers not to ask people about their immigration status and then announced the policy on the news. Yet “people were still afraid,” he said. That suggests immigrants might become less willing to work with police even in “sanctuary cities” whose law enforcement agencies refuse to cooperate with federal authorities in enforcing immigration laws under certain circumstances. At least three of the cities that have experienced declines in crime reports among Latinos could be considered sanctuary cities.10

It’s possible that crime reports are an early sign of a broader withdrawal from the use of government and nonprofit services by immigrants. Randy Capps, director of research at the Migration Policy Institute, a think tank that generally supports liberal immigration policies, said the effects of fear in the immigrant community likely spill beyond law enforcement. “There are big concerns around interaction with the schools, public health clinics, food assistance programs,” Capps said in an interview. “They’re afraid information will be shared from one organization to another.”


  1. Denver provided data on “calls for service” and reported the ethnicity (“Hispanic” or “non-Hispanic”) of the people who called the police. For Dallas and Philadelphia, we acquired data on the race or ethnicity of the victim of the alleged crime. In most cases, however, the person calling the police and the alleged victim are the same person, and we will use the phrase “crime reports” to refer to both situations in this story.

  2. According to a Pew Research Center report on where unauthorized immigrants live.

  3. We excluded reports from juveniles in the Philadelphia data. Philadelphia provided data on reported crimes that fell into the FBI’s eight categories of major crimes, excluding arson. The data includes incidents of domestic violence if those incidents fall into one of the eight categories. Denver provided separate information on domestic violence, but Philadelphia did not. Denver provided data on calls for police services in nine categories, including assault, aggravated assault and harassment. Denver did not identify whether calls for service were made by juveniles or adults.

  4. This analysis excludes the nearly 40 percent of reported crimes for which the complainant’s race is either marked unknown or left blank.

  5. The Dallas data also excludes crimes involving juveniles.

  6. Non-citizen population is based on 2011-15 data from the American Community Survey. For “neighborhood,” I matched the ZIP code where a crime was reported to the ZIP Code Tabulation Area reported in the ACS. ZCTAs aren’t exactly the same as ZIP codes, but they align closely for most purposes; I ran the same analysis using census tracts and found the same overall results. For the ZIP code analysis, the correlation was r = -0.38, p = 0.006. Results varied slightly depending on whether we used the ZIP code in which the crime occurred or the ZIP code in which the victim lived, but both produced significant correlations.

  7. Dallas doesn’t report citizenship status for crime victims, so there’s no way to know for sure whether non-citizens are the ones driving the decline in crime reports among Latinos. But the data is strongly suggestive: According to 2011-15 American Community Survey data, 89 percent of non-citizens in these ZIP codes are Latino, and 43 percent of Latinos there are non-citizens. (Data on the ethnicity of non-citizens isn’t available for one of the five ZIP codes.) So it’s very likely that any decline in crime reports among Latinos in these neighborhoods would be driven, at least in part, by non-citizens.

  8. For the first three months of 2016, there was a 3.2 percent increase in crime reports by Hispanics compared with the same months in 2015. Among non-Hispanics, there was a 1.8 percent increase.

  9. I calculated the year-over-year change in Latino crime reports for 90-day periods beginning in May 2016 (the earliest period for which sufficient data is available). For each period, I then calculated the correlation between the change in crime reports and the non-citizen share of the population in each ZIP code. For most of 2016, the relationship was generally positive (more immigrants tended to mean a bigger increase in Latino crime reports) and not statistically significant. But starting in November, the correlation turns negative (more immigrants tended to mean a bigger decrease in Latino crime reports) and statistically significant. The relationship has been the strongest (in both magnitude and statistical significance) in the period since Inauguration Day.

  10. There is no definitive list of sanctuary cities, or even a universally agreed-upon definition. But Philadelphia, Los Angeles and Denver all have policies that limit the circumstances under which they will cooperate with ICE “detainer” requests.

Rob Arthur is a former baseball columnist for FiveThirtyEight. He also wrote about crime.