"Race and Ethnicity in the 2024 Elections" by Craig M. Burnett

 The 2024 presidential election proved to be most interesting, and perhaps somewhat confusing.  While voters’ perception of the economy — especially inflation and its lingering effects — likely had the largest single impact on voters, racial and ethnic identities also mattered.  In general, as it relates to race and ethnicity, the following conclusions are reasonable:  1) Latino/Hispanic voters have never voted routinely as a bloc, with 2024 suggesting that even more diverse patterns in voting are on the horizon; 2) Black voters, which have typically voted as a bloc, would seem to have voted less as a bloc for Harris in 2024 (turnout may also have played a role here, too, but I don’t dive into that).

To understand why requires a (small) amount of knowledge about how statistics work, specifically as it relates to distributions.  Distributions are a fancy way of saying “this is how a trait is spread among the population.”  While many distributions can describe many political phenomena (e.g., incidence of war are rare, others may be bounded by what we can physically observe, etc.), the most common one is a normal distribution (see Figure 1).  

Lots of things follow a normal distribution:  height and weight are typical ones, but ability (generally conceived) also follows this distribution.  Take for instance basketball ability.  I, unfortunately, fall somewhere around the -1.5 mark — that is, I’m definitely worse than average, but I’m likely not so terrible you would walk off the court (or so I tell myself).  Lebron James, however, is clearly beyond the +3 standard deviation.  He’s also super tall, whereas I’m just reasonably so.  Accordingly, his salary is much larger than mine due to this rather rare combination of traits.  The fact is, most of us reside somewhere in the middle on most things, including political preferences. 

Parties also occupy this space:  political preferences are also normally distributed.  Most of us are somewhere in the middle, likely with mixed up ideas about what should happen.  We don’t often find ourselves so neatly organized when it comes to politics.  But… we have a two-party system thanks to single-member districts and plurality voting, so two parties are all we get.  That means that many people may feel somewhat poorly represented.  That is, many people are relatively more liberal or conservative than their preferred party.  Or, likewise, a party can move too far to the right or left and alienate people in the middle.

Groups can become “captured” (see Paul Frymer’s book Uneasy Alliances for a deeper explanation) when they are to the left or right of their preferred party and vote as a bloc.  Groups become captured because parties rely on their votes but do not have to deliver on policy — these voters have nowhere else to go.  It allows parties to chase the median voter without having to directly service their captured groups.

Scholars have identified Black voters as one of these “captured” groups.  Dawson’s work in particular argues that, unlike other groups, Blacks overwhelmingly subscribe to the concept of “linked fate.”  This requires one to view individual actions as being at least partially shaped by the group’s interests.  In Dawson’s estimation, subscribing to “linked fate” allows for people to vote against their personal interests in favor of the groups’ interest.  Specifically, Dawson highlights the observation that middle-class Black voters tend to support economically liberal policies.

Black voters are unique in this observation.  Take, for example, the following Hispanic/Latino subgroup tendencies:

• Puerto Ricans: ~ 71% Democratic or Democratic Leaning
• Mexican Americans: ~67% Democratic or Democratic Leaning
• Other Latino Groups: 53% Democratic or Democratic Leaning
• Cuban Americans: 69% Republican

Clearly, there is no bloc voting pattern here. Asian American voters are equally diverse in their approach, too.

Is “linked fate” still dominant?

Before the 2024 election, The New York Times released survey data on representative samples of Black and Latino voters:

The data here tell a consistent story:  Black support for Democrats may be somewhat waning (or oscillating, perhaps) and Hispanic/Latino voters continue to have diverse preferences and do not vote as a bloc.

Other data, too, suggests patterns may have shifted in 2024.  The Associate Press’s analysis of their exit poll data suggest Black voters’ support for Trump may have double (https://www.pbs.org/newshour/politics/voter-anxiety-over-the-economy-and-desires-for-total-upheaval-brought-trump-back-to-office-ap-votecast-shows).  While I would prefer to wait for academic sources to confirm such a shift — for example, the Congressional Election Study or the American National Election Study — but those are likely a few months off.  Additionally, I’d like to see an additional presidential election before rendering a full opinion, but views from today suggest there is some movement.  Given that Dawson argues that a shared experience (slavery into the Civil Rights Era) drives “linked fate,” it stands to reason we may be on the precipice of witnessing its decline:  as the years pass, the connection to the Civil Rights Era declines as eyewitnesses pass away.  Whether this is a lasting effect — one that likely reflects a longer-term realignment — remains to be seen.

Craig Burnett is an Associate Professor of Political Science at Hofstra University. His research focuses on voting behavior, direct democracy, electoral institutions, race and ethnic politics, and research methods.