If America is ever going to be just, white people and men are going to have to lose some things. For people who care about social justice, it’s often more rhetorically effective (and, depending on who you are, comfortable) to frame things in terms of gains for women and people of color. Often, these genuinely are rising-tide situations; in theory, at least, the appalling racial wealth gap could be closed through African-American prosperity, without any white people losing their houses. Everyone wins!
But some things are zero sum. The U.S. Senate is a particularly clear example: The only way to get more seats for women and people of color is to take them away from men and white people. If you care about basic justice, or have even a remote commitment to the principle of fair representation, your only option is to disempower white men. And this is true to an even greater extent than you might think—I’ve got some data about the Senate below, but you might find it interesting to try and guess before you look at it: How many white men would need to lose their seats for the Senate to reflect the racial and gender makeup of America?
First, here’s the gender balance of the senate (see the Appendix at the end of the post for information about my sources, the categories I’m using, and so on).
The bar on the left shows the real, present-day Senate; the bar on the right shows what the Senate would look like if it was truly representative of the United States population. We are 28 women Senators shy of a just Senate—more than we currently have. In fact, in the entire history of the Senate, there have been only 52 women Senators, meaning it would take almost every woman Senator ever just to reach the number we should have right now.
The numbers are even more striking when it comes to race/ethnicity:
With 91 Senators who are non-Hispanic white, this cohort is overrepresented by 29 seats. More important, they leave almost nothing for anyone else. We should have 13 African-American Senators; instead, in the entire history of the Senate there have been 10. Moreover, the current senate doesn’t have 13 seats for all people of color combined—and the situation is actually more dire than it appears here: Kamala Harris appears on this graph twice, since she was the only person listed twice in my source material, in her case as both Asian American and African American. (See the Appendix for more on multiracial identity in this data; here, my goal was to maximize the diversity I could show in the Senate—and this is still the result).
This decision to count Harris twice points to an interesting question about representation. In theory, there are more “efficient” ways to diversify the Senate than simply assigning seats by one race and gender at a time, as I’ve done here. If, for instance, ten seats held by white men went to ten new candidates like Harris, we’d have ten more women, ten more African Americans, and ten more Asian Americans all at once. These candidates really would reflect those identities, and the representativeness of the Senate would be improved for it. The problem is, we would still have too many white men. Overrepresentation is just as inaccurate a picture of the country as underrepresentation, even if we can correct them at different rates.
The actual Senate makes this clear. Here is the data where every Senator is categorized as either a Non-Hispanic White Man or not:
The 72 white men in the Senate are wildly out of proportion to the country. Only 62% of Americans are non-Hispanic white people in total, much less white men. That means we have 41 too many white men in the Senate. This is really a staggering number: There are more extra white men than there should be white men total—so, while African-Americans only have 23% of the seats they should, white men have 232% of the seats they should. In short, two out of every five people in the Senate are extraneous white men. And that means that a group that shouldn’t even come close to a majority has a supermajority with votes to spare. From the standpoint of diversity, adding 10 mixed-race women would be good in itself; but we would still be left with a grotesque imbalance of white male influence on legislation.
Incidentally, this problem persists for Democrats, Independents, and Republicans:
In both absolute and percentage terms, it’s much worse for Republicans. There are 42 white male Republican Senators, meaning that one party has 11 more than the whole Senate should have. This is another way to see the scale of the imbalance; remember that there are 41 extra white men, so from a race/gender perspective, replacing the entire Republican party with women, people of color, and one white guy would get us to… fair representation. Still, Democrats and Independents are also quite unbalanced; to fix the problem, the entire political system would have to change.
Of course, it’s not clear that this is possible. Senators are elected on a state level, and there are, for instance, no states with a majority black population. If white people, and especially white men, are reluctant to vote for women and people of color, the Senate is likely to stay profoundly unrepresentative of American diversity—almost as though it’s an inherently unjust institution.
But there’s something telling about laying the injustice bare. This is how far we are from equality. If everyone had a fair shot at life and voters were unbiased, the Senate would look much more like the country it’s supposed to serve. Instead, 41% of its members are living proof that the past is not even past. If women and people of color are to have what is rightfully theirs—if, in other words, the American promise of representative democracy is ever to be achieved—the only solution in the Senate is to take from white men.
Appendix: Notes on the data
I’m taking racial and ethnic information about the U.S. from this Wikipedia page, which aggregates census data. Racial and ethnic categories are constantly in flux (as the history of the census shows), and the data reflect that ambiguity. For the graphs, I’m using the categories African American and Asian American, which the census considers racial, the category Hispanic/Latino, which the census considers ethnic, and the categories Non-Hispanic White American and All Other, which are somewhere in between. “All Other” includes the categories Native American and Pacific Islander, among other things. Overall, these are imperfect divisions with a lot of overlap, but I think they give a rough idea of American race and ethnicity in the kind broad strokes used in this piece.
For the racial/ethnic breakdown of the Senate, I’m using this government site. (I asked the Senate Historian’s Office how they reach the conclusions represented on that site, and a respondent said that most of the information there is based on historical sources, some on public statements, and for contemporary Senators on self -report—it seems that a Senator who disagreed with her own information could have it changed.)
Using this site, I assume everyone not listed there is non-Hispanic white. Kamala Harris is listed as both African American and Asian American; she is the only person who is listed as multiracial. In part, this may be because I used specifically Non-Hispanic White American—e.g., perhaps Ted Cruz and Marco Rubio would claim both identities. My decision here reflects an attempt to maximize the picture of Senate diversity, so my rule of thumb is to assign each person any identity that would affiliate them with people structurally disadvantaged by American racial and gender politics.
For gender, I used this page and assumed that any Senators not on it were men. This is a good example of how very limited this post is, in that it confines gender to two cis possibilities. In this post, I’m not addressing representation of trans people—or queer people, people with disabilities, different national or language backgrounds, religious affiliation, or anything else, though these are all important areas where the Senate is similarly unrepresentative of the country.