Alex Nowrasteh and Krit Chanwong
Ezra Klein recently interviewed David Shor, a data scientist at the Democratic consulting firm Blue Rose Research. Shor made two important immigrant-related points. First, the foreign-born share of the population in a county was highly correlated with a shift toward Trump. Second, Trump likely won the immigrant vote. Naturalized immigrants went from favoring Biden in 2020 by 27 points to favoring Trump by one point.
This abrupt change destroys the common immigration-restrictionist argument that more open immigration policies will tilt the country leftward, an argument commonly made by Elon Musk that explains why he decided to support Trump in the 2024 election.
Shor’s interview contains much evidence for his claims that complements data elsewhere. There is some evidence that demographics with higher immigrant shares shifted rightwards in 2024. CNN’s 2024 Election exit polls, for example, found that 46 percent of Hispanic voters cast their ballots for Trump, a 14 percent rightward shift from 2020. Asian voters also shifted rightwards: 40 percent of Asians voted for Trump in 2024, a six percent increase from 2020, cementing a reversal of the trend where pollsters identified increasing Democratic vote shares and identification among Asians. Moreover, The Economist found that the “foreign-born share of a county’s population explains half of the geographic variation in the electoral swing from 2020 to 2024.”
Here, we follow The Economist’s methodology by analyzing the relationship between variation in the foreign-born population of a county and increased support for Trump. However, we make two changes:
The Economist article was published nine days after the 2024 election and used demographic information from the Census Bureau’s 2022 5‑Year American Community Survey (ACS) data. We use more complete election data (compiled by Tony McGovern) and the more recent 2023 5‑Year ACS.
We excluded Alaska due to data limitations.
We go into greater detail on the relationship between the change in Trump’s support and the county-level foreign-born population share than The Economist and David Shor’s publicly shared slide deck. The figures below are not causal, and they merely show the relationship between the share of a county’s foreign-born population, the change in support for Trump from the 2020 election to 2024, and report the R‑squared that shows how much of the change in vote share can be explained by the foreign-born share of a county’s population.
However, the progression of figures below gets more specific and, we hope, provides some evidence that changes in the immigrant voter share of a county’s population are likely driving much of the change in Trump’s vote share.
Figure 1 shows that the foreign-born share of a county’s population explains around 38.6 percent of Trump’s increased vote in 2024. These effects seemed to be nationwide, occurring in New York, New Jersey, Texas, Florida, Arizona, and elsewhere. However, Miami-Dade seems to be the best indication of this trend: Trump was the first Republican to win in Miami-Dade County since 1998. This was likely, in no small part, due to his increased support from immigrant voters.
This also shows that increased support for Trump in these counties was electorally somewhat inefficient because they didn’t flip these states except for probably Arizona, which wasn’t enough to change the outcome of the election anyway but merely ran up his popular vote totals in states he would win (Florida and Texas) or narrowed the margins in states that he had no hope of winning like New York. Trump’s victory in the swing states of Pennsylvania, Michigan, and Wisconsin was narrow.
Figure 2 focuses on how the naturalized immigrant share of a county’s population is related to changes in the Trump vote share. Naturalized immigrants are those who have become citizens and can thus vote in presidential elections, so Figure 2 gets closer to estimating how immigrant voters may have directly changed county-level vote shares. Figure 2 finds trends identical to Figure 1: The percent naturalized citizens in an area explained 38.8 percent of the nation’s 2024 pro-Trump vote share.
Which immigrant group most explains the country’s rightward shift in 2024? Figure 3 plots the Asian foreign-born population against increases in Trump’s 2024 vote share. Figure 4 reproduces Figure 3 but uses the Asian naturalized population instead of the Asian foreign-born population. There is a positive trend between the Asian foreign-born shares and increased 2024 Trump votes, but it’s slight and can only explain about 7 percent in both cases. In other words, the Asian immigrant population can’t explain much of the nationwide shift toward Trump. This is unsurprising as Asian immigrants are less than 4 percent of the American population.
Differences in the Hispanic foreign-born population across the counties can explain much more of the variation in changes in Trump support. Figures 5 and 6 plot the Hispanic foreign-born and Hispanic naturalized population with increases in Trump votes in 2024. The trend is quite positive in both cases. Interestingly, Miami-Dade is below the countrywide trendline as Trump increased his advantage in Miami-Dade County by only 9.29 percent. If Miami-Dade followed countrywide trends here, Trump would have increased his advantage in Miami-Dade by 12 to 12.5 percent. Quite significantly, the Hispanic foreign-born share and the Hispanic naturalized share of the population explained 42 to 43 percent of the variation in Trump’s increased vote share in 2024.
The results we found are consistent with other sources, which find that Trump’s greater vote share in 2024 was significantly aided by increased support from Hispanic voters, many of whom are immigrants. Florida, New York, New Jersey, Texas, and Arizona best exemplify this trend. Of those states, only Arizona likely flipped from Biden to Trump due to the shift in Hispanic votes.
Figure 7 shows that Trump improved his 2020 margin in every county in Florida. Yet there were only three counties (Osceola, Hendry, and Miami-Dade) where he improved his margins by more than seven percent. Hispanic immigrants were at least 20 percent of the population in these counties. This surge in Hispanic support in Florida allowed Trump to improve his margins of victory by almost nine percent statewide.
Figure 8 plots Trump’s improved margins by county in New York. He improved his margins everywhere in New York, but by more than 10 percent in only Queens County and Bronx County, where Hispanic immigrants made up more than 16 percent of the population.
Figure 9 plots Donald Trump’s improved margins by county in New Jersey. Here, Donald Trump was able to improve his 2020 margins by increasing his Hispanic support in counties near New York City. This was especially true in Passaic and Hudson counties, where Hispanic immigrants made up approximately a quarter of the population.
Figure 10 plots Donald Trump’s improved margins by county in Texas. Here, Trump was able to improve his 2020 margins in border counties with a high share of Hispanic immigrants. In Maverick and Laredo Counties, where approximately 30 percent of the population are Hispanic immigrants, Trump was able to improve his vote share by more than 10 percent.
Figure 11 plots Trump’s improved margins by county in Arizona. Unlike the other states examined here, Arizona was a swing state in 2024. Trump’s performance in Arizona significantly improved in heavily Hispanic counties like Yuma (7.5 percent) and Santa Cruz (8.6 percent). This surge in Hispanic support for Trump is likely why he was able to decisively defeat Kamala Harris by more than five percentage points in Arizona.
Figure 12 plots the percentage of (all) foreign-born naturalized citizens by the year they gained their citizenship with increased Trump vote share in 2024. There is some evidence that the longer the naturalized immigrant share of the population spent in the United States, the higher their county’s vote share increase for Trump in 2024. The percent share of citizens naturalized from 1995 to 2014 explains slightly more variation in Trump’s 2024 increased vote share than the percent share of citizens naturalized after 2015, but not by much.
The major missing variable in our above analysis was education. Political sorting into different political parties by education levels is happening across the country and it would have been best to see if this trend also holds for immigrants. It likely does because Hispanics typically have less education than Asians and became much more Republican, but we can’t say for certain using the ACS datasets we accessed. Census microdata would have allowed us to look at education levels by foreign-born, ethnicity, and naturalization on the county level, but we leave that analysis to others.
An alternative hypothesis is that the presence of many immigrants in a county prompted native-born Americans to become more nativist and vote for Trump in the 2024 election in greater numbers. This would be consistent with a localized immigrant backlash theory that has some support but doesn’t hold up under scrutiny in many cases.
Some political partisans argue that the bipartisan Immigration Act of 1965 was a long-term Democratic plot to help their political party, even though politicians and political parties rarely think beyond the next election. The Republican Party in California likely sabotaged itself for an entire generation in California by supporting a harsh immigration enforcement proposition in 1994, but that seemed to be unique to California in that year. Meanwhile, then-governor George W. Bush doubled his party’s vote share among Hispanic voters beginning in 1994.
However, policy is downstream from politics and the actual important metric we care about. It matters not which political party liberalizes the economy, cuts taxes, and protects our constitutional rights but that those actions be taken. The general effect of immigration on economic and political institutions is either positive or neutral, with few academics saying otherwise and none getting their work through peer review—as flawed of a process as that is.
Regardless, much of the growth in support for the political party that wants to reduce legal immigration comes from the voting habits of immigrants and their descendants. Incidentally, the GOP is also perceived as more supportive of free markets, even if that conclusion is justifiably much weaker than in years past. “Demographics is political destiny,” their line of thinking goes. As the 2024 election shows, politics and policy are less likely to be downstream of demographics than we’ve been told.
