Written by: @Corncobanalysis
The debate over whether the 2024 election was decided by turnout or persuasion will likely persist to the end of time. That being said, my contribution is that voter file and precinct data do show some serious turnout issues for Democrats among Asian-American voters. That’s not to say there weren’t persuasion issues either, but it’s impossible to ignore such a significant differential turnout for Republicans in the 2024 presidential election.
Three quick points:
- Although the umbrella term “Asian-American” contains a multitude of ethnicities and cultural backgrounds, the rightward swings and turnout drops were fairly universal. There was a marked turnout drop for every group and ethnicity.
- The turnout decline was not limited by location or state. Although there was a slight improvement for very few Asian-Americans in swing states, it was not beyond the (small) turnout boosts we saw in competitive states. There were some *really* bad swings/turnout for Indians in Texas and Chinese in New York, but overall they were in keeping with major declines everywhere (and those areas also have many other minorities contributing to the swing).
- Outside of one curious case among Vietnamese-Americans in Southern California, none of the declines in turnout benefited Democrats.
The Data:
I’ll start with demographic data scraped from the voter file helpfully made available online via TargetSmart and the California GOP returns site. California tracks racial categories during voter registration and has the largest number of Asian-Americans by far, which makes it the most reliable dataset for analysis.
Overall turnout in California dropped; TargetSmart uses turnout as percentage of registered voters which suggests a much larger drop of 9.6 percentage points, but the more reliable Voting Eligible Population stats from Michael McDonald suggest a more modest 6.8pp drop.*
From this chart (set to percentage of turnout), we can see some glaring dropoffs by ethnic group.
Turnout Dropoffs by Race, CA
White: 5.8%
Black: 7.2%
Hispanic: 12.9%
Asian: 13.8%
Other: 12.9%
As we can see, the largest turnout dropoff in California on the voter file record came from Asian voters. Because the numbers of Asians and Hispanics in California have grown since 2020, having fewer votes than 2020 is quite poor. While there were fewer White voters than in 2020, this masks a significant relative decline among nonwhite voters that hurt Democrats.
As we drill down into party registration figures, it becomes self-evident that these voters were disproportionately Democratic or at least had voted Democratic in the past.
CA-28 (D-Judy Chu), one of California’s two Asian-American VRA districts, saw dismal swings against Democrats on the Congressional and Presidential level (Harris did about 7 points worse than Biden and Chu did about 3 points worse than her 2022 Biden midterm performance).
Biden – 242,549
Trump – 117,111
Harris – 204,263
Trump – 115,237
Yes, you read that correctly, Trump got ~2,000 fewer votes than he did in 2020, but he improved his margin because Harris got ~38,000 fewer votes than Biden did.
So, what the hell happened? Well, further drilling into party registration data in California shows an astonishing 16.1pp dropoff from 2020 for Unaffiliated Asian voters (registered Democrats and Republicans saw a roughly equal decline). All age cohorts saw substantial dropoffs; interestingly, the largest came from 30-39 voters, not 18-29 as one might expect. Overall, there was a significant turnout advantage for older Unaffiliated Asian voters, which are more likely to be conservative, for obvious reasons.
Yet while poor Latino turnout has been “expected” over the years, poor Asian turnout is not. California Asian turnout had been steadily increasing over the previous three presidential cycles and was relatively better in both the 2018 and 2022 midterm cycles. But, for some reason, the Harris campaign did not motivate Asians to vote—and once again, the raw vote margins declined whether the district was more Indian, Chinese, or Korean. The only exceptions were Vietnamese areas in Southern California, which, for reasons that no one has fully explained, resisted the swings against Democrats despite turnout drops.
(Caption: Can someone explain this? It swung about 23 pts toward Trump from 2016 to 2020, but had less turnout than 2020.)
But enough about California, what about the rest of the country? Unfortunately, due to some data quality issues, the TargetSmart national stats aren’t quite as helpful for a small minority group, though they do show a significantly larger dropoff for the Asians they have in the file**. Across the 145M records in TargetSmart, Asians saw the largest turnout drop from 2020 to 2024 relative to registered voters of any demographic. More concerningly, though Latino and Black turnout “bounced” from 2022, Asian turnout saw the smallest relative gains from the midterm as well.
The results do not show much better turnout from Asian voters. When ElectionCord surveyed smaller districts of 100,000 people across the nation, districts with Asians saw the largest turnout decline of all.
There was a significant swing against Harris from Biden 2020 across majority Asian districts as well, which points to a real persuasion effect. Yet the significant turnout dropoff among a group that is growing in population and has more educational attainment is still rather baffling. In the surveyed areas, the median income of majority-Asian districts was $140K, and 47% had a college degree. Needless to say, Harris being the first Asian-American on the top of a presidential ticket did absolutely nothing.
It’s particularly baffling when you consider that Asian districts saw one of the largest turnout *increases* between 2016 and 2020, more than White voters and second to only Native American voters.
In swing states, the picture was a little different, but there are so few Asian-Americans in the swing states that they didn’t make much of a difference. There was a raw vote increase from 2020 to 2024 for Asian-American voters in the Georgia voter file, but their percentage of the vote and turnout was very slightly down from 2020 and even 2022 as White turnout skyrocketed. Still, the turnout among Asian voters wasn’t quite as bad, though there were still some margin declines for Democrats in Asian and Latino areas in swing states, most notably in Pennsylvania.
So, why did this happen? Well, the easiest explanation is that Asian-Americans, particularly lower-propensity ones that are historically favorable to Democrats, did not feel like voting for Kamala Harris. But in a relative sense, it’s more accurate to say that they simply did not keep up with the huge turnout numbers from low-propensity Republicans who carried over from 2020.
Was there a ballot access issue? There is still a major academic debate on whether increased ballot access due to vote-by-mail increased turnout (some studies say no, others say yes). The effects are slight, plus California and Washington are all-mail states and there were still disproportionate turnout declines among Asian voters, so I wouldn’t say that was a huge factor. Reduced mail-in voting may have dropped overall turnout slightly, but the enormous turnout declines we saw were likelier due to what Charlamagne Tha God referred to as “The Couch” problem—many Asian-American voters from Dem-leaning areas chose the couch.
Still, the theory that having Donald Trump on the ballot would yield consistently greater engagement among Asian voters has been upended. There were large relative increases in Asian turnout in 2016, 2018, and 2020, which Democrats trumpeted as signs of their diverse coalition. Issues with Asian-American voters and turnout did not start appearing until 2021-2024, so the enormous dropoff from 2020 to 2024 comes as a shock. Indeed, Republican turnout advantages among Asian-American voters were more reminiscent of the 2022 midterms or midterms when Obama was president, not a presidential election.
Across the country, lower turnout was generally correlated with larger swings against Democrats. This tracks with Democrats doing better with White voters (who turn out the most). However, when looking at the difference between 2020 turnout (a D+4 year) and 2024 (R+2), one can find a curved relationship (if there is one at all). This suggests that Democrats did very well in “super-voter” areas where everyone votes more, but also did very poorly in areas that dropped off from 2020 to 2024.
I do not have the media or policy prescriptions on hand to fix this issue, but it’s certainly something that Democrats need to study (yet again) before the 2026 midterm elections.
* For some reason, Wikipedia has the total number of votes in California about 300k votes lower than it is on the official certification page, and TargetSmart is closer to the “real” number.
**Some of them are likely “assumed” Asians due to modeling who mark themselves as “Other”, but have an Asian last name, but this is a marginal effect.
Picture via: Julienne Schaer/NYC & Company
Corncobanalysis:
@corncobananalysis on X (Twitter)
@corncobanalysis.bsky.social on Bluesky