By: Emma Buchanan

Published: March 9, 2026

In an unprecedented manner of immigration enforcement, the Department of Homeland Security continues to use racial profiling and abusive tactics to indiscriminately detain noncitizens in cities and towns across the country.[1] Immigration and Customs Enforcement (ICE) agents have been authorized to patrol public places in Hispanic neighborhoods in order to detain people based on how they look and the language they speak.[2] A recent case, Perdomo v. Noem, challenged ICE’s unlawful and harmful actions.[3]

The Perdomo plaintiffs included U.S. citizens who were stopped on their way to work and questioned about their immigration status on a number of occasions.[4] These plaintiffs challenged warrantless arrests by “roving patrols” based on alleged reasonable suspicion informed by four factors: (1) apparent race or ethnicity, (2) speaking Spanish or speaking English with an accent, (3) presence at a particular location, and (4) type of work.[5] The plaintiffs’ temporary restraining order was granted and upheld in the lower courts, concluding that stops based on those four factors alone did not satisfy the Fourth Amendment’s reasonable suspicion requirement.[6] The government appealed to the Supreme Court via the emergency docket, better known as the shadow docket.[7]

On a writ of certiorari, the Supreme Court stayed the temporary restraining order, permitting the continued use of roving patrol enforcement.[8]  The majority offered no explanation for its decision; however, Justice Kavanaugh did provide a poignant concurrence.[9] It is unclear in Supreme Court jurisprudence whether Justice Kavanaugh’s opinion can be taken as that of the majority in this case.[10]

While the decision was not a ruling on the merits, it signals that the Supreme Court would likely uphold roving immigration patrols against a Fourth Amendment challenge if appealed on the traditional docket.[11] This indication is highly problematic because Justice Kavanaugh’s concurrence entirely fails to support a nexus between the four factors and a reasonable likelihood that those factors will identify an undocumented individual.[12]

Reasonable suspicion requires “specific, articulable facts” regarding a person suspected of committing a crime.[13] ICE officers may only stop an individual based on those specific, articulable facts that, along with rational inferences and the officer’s experience, reasonably warrant suspicion that the person is undocumented.[14] An “unparticularized suspicion or hunch’” is not enough.[15] A characteristic that is common to noncitizens and citizens alike does little to establish such suspicion because specific, articulable facts cannot “describe a very large category of presumably innocent people.”[16]

In his concurrence, Justice Kavanaugh failed to cite to any particularized evidence when stating, “[illegal immigrants] tend to gather in certain locations to seek daily work, . . . work in certain kinds of jobs, . . . and do not speak much English.”[17] Problematically, these unsupported statements are the basis for Justice Kavanaugh’s claim that the four factors constitute reasonable suspicion.[18] I seek to dispel the idea that a sufficient nexus could exist between the four Perdomo factors and a reasonable suspicion that a person is undocumented through precedent and statistics indicating that the factors are wholly unparticularized.[19]

First, as a factor for reasonable suspicion, apparent race or ethnicity has very little probative value.[20] While nearly half of the Central District of California identifies as Latino, permitting ICE to make assumptions about a person’s race and ethnicity means a specific demographic of Latinos will be disproportionately targeted—those who “look Latino.”[21] The Supreme Court considered this factor fifty years ago in United States v. Brignoni-Ponce, where it held that percieved Mexican ancestry alone was not enough for reasonable suspicion that someone was illegally present in the United States.[22] As the Court noted, “[a] large number of native-born and naturalized citizens have the physical characteristics identified with Mexican ancestry . . . .”[23] The Court’s rationale still holds today, particularly because ICE is now targeting “Latino appearance,” which is even less specific than Mexican appearance.[24] In actuality, Latinos are not a monolith, nor do they look like one.[25] Latinos have a wide range of skin tones and features.[26] Furthermore, a majority of Latinos have U.S. citizenship or other legal status, making this factor unlikely to help ICE more than it hurts their undeserving targets.[27]

Second, speaking Spanish or speaking English with an accent is a factor that applies to at least 41.9 million people in the U.S.[28] In fact, over one-third of people in the Central District of California speak Spanish at home.[29] In her dissent, Justice Sotomayor also noted that language and ethnicity are often intertwined, such that language can become a proxy for ethnicity that is inadequate to provide reasonable suspicion.[30] Furthermore, despite speaking Spanish, 72% of Latinos report that they speak English proficiently, whether with an accent or not.[31] Particularly in California, this factor does little to provide ICE with reasonable suspicion that a person is undocumented, as it describes so many people who are documented.[32]

Third, presence at a particular location is a volatile factor.[33] The Supreme Court considered presence at a particular location in Brignoni-Ponce, where the Court reasoned that both documented and undocumented people could be present near the border, so finding someone in that location was not enough for reasonable suspicion.[34] In Perdomo, there was also no particularized evidence offered to suggest that car washes or bus stops are predominantly or exclusively used by undocumented people.[35] In contrast to Brignoni-Ponce, the Central District of California is farther from the border and populated with both documented and undocumented people; therefore, this factor must fail.[36]

Lastly, in this case, one’s type of work also refers to an entire sector of low income workers.[37] In 2024, nearly half of all employed Americans were hourly-wage workers, like the day laborers and agriculture workers Justice Kavanaugh singled out.[38] Roughly 60% of Latinos in the U.S., documented and undocumented, work for hourly pay.[39] In United States v. Manzo-Jurado, the Ninth Circuit held that, despite officials who stated they encountered numerous work crews who were illegally present, a group’s appearance as a work crew was only “marginally relevant” to reasonable suspicion.[40] Latinos are already likely to make less money than white people in the U.S., so their hourly jobs are more likely connected to their lower-incomes than their immigration status.[41]

With this data, it is clear that ICE’s four factors fail to provide the specific, articulable facts needed to pass constitutional muster, both separately and jointly.[42] The factors amount to nothing more than unparticularized suspicion that someone is undocumented, as they describe a great number of law-abiding U.S. citizens.[43] In fact, more than 170 U.S. citizens have been arrested in immigration raids and protests in 2025, despite declaring their citizenship and even presenting valid documentation.[44] ICE’s immigration strategy contradicts our fundamental freedom from unwarranted search and seizure and sets a dangerous precedent.[45] I join with Justice Sotomayor’s view in stating: “[w]e should not have to live in a country where the Government can seize anyone who looks Latino, speaks Spanish, and appears to work a low wage job. Rather than stand idly by while our constitutional freedoms are lost, I dissent.”[46]

 

 

[1] See Johnathan Wolfe, L.A. County Declares State of Emergency Over Immigration Raids, N.Y. Times (Oct. 15, 2025, 12:21 PM ET), https://www.nytimes.com/2025/10/15/world/americas/los-angeles-emergency-immigration-raids.html; Heather Cherone, ‘Back Off’: Pritzker Blames ICE Strike Teams for Creating ‘Mayhem’ Across Chicago, WWTW (Oct. 15, 2025, 12:12 PM), https://news.wttw.com/2025/10/15/back-pritzker-blames-ice-strike-teams-creating-mayhem-across-chicago.

[2] See Maanvi Singh, At Home Depot, Ice Raids Terrorize the Workers Who Helped Build LA: ‘They Just Come and Grab You’, The Guardian (June 16, 2025, 7:44 AM ET), https://www.theguardian.com/us-news/2025/jun/16/home-depot-ice-raids-los-angeles (describing a raid at a Home Depot in Huntington Park, California, a city that is nearly 97% Latino).

[3] Perdomo v. Noem, 790 F. Supp. 3d 850, 865 (C.D. Cal. 2025), aff’d, 148 F.4th 656, 664 (9th Cir. 2025).

[4] Perdomo, 790 F. Supp. 3d at 868-69.

[5] Id. at 871, 892-94.

[6] Id. at 897; Perdomo, 148 F.4th at 690.

[7] Garrett Epps, Chamber of Secrets, Wash. Monthly (June 19, 2023), https://washingtonmonthly.com/2023/06/19/chamber-of-secrets/.

[8] Noem v. Perdomo, No. 25A169, 2025 WL 2585637, at *1 (U.S. Sep. 8, 2025).

[9] Id. at *1-5 (Kavanaugh, J., concurring).

[10] Epps, supra, note 7.

[11] See Perdomo, 2025 WL 2585637, at *1-5 (Kavanaugh, J., concurring) (considering the likelihood of success on the merits as a factor in whether to approve the stay).

[12] See id. at *3 (stating that there is a high percentage of undocumented people in Los Angeles, that undocumented people tend to gather in certain locations to seek work, that they tend to work certain kinds of jobs, and that many come illegally from Mexico or Central America and do not speak much English).

[13] Id. at *9 (Sotomayor, J., dissenting); e.g., Perdomo, 790 F. Supp. 3d at 889-90.

[14] Id.

[15] Perdomo, 2025 WL 2585637, at *9 (citing Illinois v. Wardlow, 528 U.S. 119, 123-24 (2000)); e.g., Terry v. Ohio, 392 U.S. 1, 21 n.18 (1968) (emphasizing specificity as critical to the Court’s Fourth Amendment jurisprudence); see Perdomo, 148 F.4th at 680.

[16] Perdomo, 2025 WL 2585637, at *9 (Sotomayor, J., dissenting).

[17] Id. at *3 (Kavanaugh, J., concurring).

[18] Id.

[19] See id. (failing to support conclusions about Latinos with statistics).

[20] Perdomo, 148 F.4th at 683.

[21] Id.; see Perdomo, 2025 WL 2585637, at *7 (Sotomayor, J., dissenting) (describing U.S. citizens who look Latino who fear being detained by ICE and had encounters with ICE).

[22] 422 U.S. 873, 885-86 (1975).

[23] Id. at 886.

[24] See Perdomo, 2025 WL 2585637, at *7 (Sotomayor, J., dissenting).

[25] Luis Noe-Bustamante, Ana Gonzalez-Barrera, Khadijah Edwards, et al., Majority of Latinos Say Skin Color Impacts Opportunity in America and Shapes Daily Life, Pew Rsch. Ctr. (Nov. 4, 2021), https://www.pewresearch.org/race-and-ethnicity/2021/11/04/majority-of-latinos-say-skin-color-impacts-opportunity-in-america-and-shapes-daily-life/; Isabel M. Skilton, Brignoni-Ponce and the Establishment of Race-Based Immigration Enforcement, 308 Wash. Int’l L.J. 292, 307-308 (2022) (stating that Latinos present visually as indigenous, white, Black, Asian, and as other races).

[26] Noe-Bustamante et al., supra note 25 (showing in a self-assessed study that 28% of Latinos ranked their skin color as lighter and 15% as darker, with others falling in between).

[27] Cary Funk & Mark Hugo Lopez, A Brief Statistical Portrait of U.S. Hispanics, Pew Rsch. Ctr. (June 14, 2022), https://www.pewresearch.org/science/2022/06/14/a-brief-statistical-portrait-of-u-s-hispanics/ (reporting that U.S. born Latinos are the majority and foreign-born Latino immigrants are declining since its peak at 40% in 2000); see Perdomo, 2025 WL 2585637, at *6 (Sotomayor, J., dissenting) (describing ICE asking Jason Gavidia, a Latino U.S. citizen who looks Latino, whether he was a citizen at least three times, despite Jason affirming that he was repeatedly).

[28] Countries with the Largest number of Native Spanish Speakers Worldwide in 2024, Statista, https://www.statista.com/statistics/991020/number-native-spanish-speakers-country-worldwide/ (ranking the U.S. as the fifth most populous country for Spanish speakers).

[29] Perdomo, 148 F.4th at 683.

[30] Perdomo, 2025 WL 2585637, at *10 n.6 (Sotomayor, J., dissenting) (citing Kansas v. Glover, 589 U.S. 376, 390 (Kagan, J., concurring)) (arguing that relying on apparent ethnicity and language as two factors is no different than relying on ethnicity alone because they are often so interrelated).

[31] Funk & Hugo Lopez, supra note 27 (finding that a growing number of Latinos in the U.S. are proficient in English and the number of families speaking Spanish at home has declined).

[32] Spanish, a Growing Global Language, Observatory, https://cervantesobservatorio.fas.harvard.edu/en/about/spanish-in-united-states (describing that 500 million people speak Spanish as their native language and by 2060, the U.S. will be the second-largest Spanish-speaking country in the world after Mexico).

[33] Kyle Ross, The Minimum Wage Is a Poverty Wage, Ctr. for Am. Progress (July 24, 2024), https://www.americanprogress.org/article/the-minimum-wage-is-a-poverty-wage/.

[34] Perdomo, 148 F.4th at 683; United States v. Brignoni-Ponc, 422 U.S. 886 at 873.

[35] Perdomo, 148 F.4th at 683-84 (adding that presence at a business does little to arouse reasonable suspicion even if the officer has knowledge that undocumented people frequented that location).

[36] See id. (stating that when U.S. citizens as well as noncitizens are likely to frequent a particular location, presence there does not support reasonable suspicion).

[37] Ed Gresser, 80% of American Hourly-Wage Workers are in ‘Services’, Progressive Pol’y Inst. (June 25, 2025), https://www.progressivepolicy.org/80-of-american-hourly-wage-workers-are-in-services/.

[38] Id. (demonstrating there were 80.35 million hourly-wage workers out of 161.35 million U.S. workers in 2024); see Perdomo, 2025 WL 2585637, at *3 (Kavanaugh, J., concurring).

[39] Shannon Schumacher, Liz Hamel, Samantha Artiga, et al., Most Hispanic Immigrants Say Their Lives Are Better In The U.S. But Face Financial And Health Care Challenges: The 2023 KFF/LA Times Survey of Immigrants, KFF (Jan. 18, 2024), https://www.kff.org/racial-equity-and-health-policy/most-hispanic-immigrants-say-lives-are-better-in-the-us/.

[40] 457 F.3d 928, 938 (9th Cir. 2006).

[41] Dedrick Asante-Muhammad, Alexandra Perez, & Jamie Buell, Racial Wealth Snapshot: Latino Americans, Nat’l Cmty. Reinvestment Coal. (Sep. 17, 2021), https://ncrc.org/racial-wealth-snapshot-latino-americans/ (stating that Latino households’ median wealth was $14,000, or 9% of white households’ median wealth, $160,200).

[42] See Perdomo, 2025 WL 2585637 at *9 (Sotomayor, J., dissenting); see also Melendres v. Arpaio, 989 F. Supp. 2d 822, 896-97 (D. Ariz. 2013).

[43] Perdomo, 148 F.4th at 683 (calling the profile the four factors describe “broad”).

[44] Jesus Jiménez, Congressional Democrats Investigate Arrests of Americans During Raids, N.Y. Times (Oct. 20, 2025), https://www.nytimes.com/2025/10/20/us/investigation-arrests-of-americans-ice-raids.html.

[45] Letter from Juan Vargas, Henry C. Johnson, Jr., & Betty McCollum, et al., Members of Congress, to Kristi Noem, Secretary of Homeland Security 3 (Sep. 9, 2025) (on file with Representative Juan Vargas).

[46] Perdomo, 2025 WL 2585637, at *6 (Sotomayor, J., dissenting).