Support Local Journalism
Thank you for all of your comments, ideas, photos and support!
by Sarah Bryner, Center for Responsive Politics
The group of candidates running for Congress in 2020 is the most diverse in recent memory.
However, early OpenSecrets analysis suggests that the 117th Congress will be only slightly more diverse than the 116th. Congress is already less diverse than the U.S. population, with 78 percent of members identifying as white, a much higher percentage than the 60 percent of white Americans. Women hold 24 percent of all congressional seats. Despite the relatively small number of women in Congress, especially women of color, and the underrepresentation of people of color as a whole, the 116th Congress is the most diverse in history.
If candidate recruitment, helped in no small part by organizations working to build a diverse field of candidates1, is producing a much more diverse field, why are we not seeing those gains translate into a much more diverse — and representative — Congress?
Three major issues explain the disconnect.
First, incumbency reelection rates are high and since the current body of representatives is largely white and largely male, we would expect the following Congress to be the same. Second, when incumbents do retire, creating open seats, voters tend to select candidates who “look like” the member they are replacing. Finally, candidates of color — particularly women of color — face hurdles in fundraising.
Simply put? Congressional elections are not that competitive
OpenSecrets has long documented the sky-high reelection rate of members of Congress — usually over 90 percent. Many factors contribute to these rates, including that incumbents generally have a much easier time raising the millions of dollars necessary to run a successful campaign and that districts have historically been drawn in such a way as to favor incumbents. Additionally, many incumbents run unopposed. But, increasingly, more and more challengers to incumbents are entering the political arena, furthering opportunities to diversify Congress. The 2018 cycle saw a record-setting crop of viable challengers, although few of those challengers were successful — in 2018, as in most election years, the major changes to the racial and gender makeup of Congress came from incumbents retiring, not losing.
The 2020 cycle looks to continue that pattern with a large, and diverse, set of viable challengers from both parties — but the number of challengers is only one part of a much larger story.
As of June 22, or midway through the 2020 primary season, just under half of all states have held primaries. Those states will send 280 lawmakers to the 117th Congress2, and although the general election has not yet taken place, we already know a great deal about what the 117th Congress will look like based on the racial and gender makeup of the nominees chosen in these primaries. After all, 38 incumbents are running unopposed and some of them faced no primary opposition, either. This is higher than the number of open seats — 25 — where the incumbent is retiring. Additionally, 27 senators represent these states and will be returning because they are not up for reelection.
Of the remaining contests, where an incumbent is facing a challenger, only 49 are even mildly competitive, defined here as races where both candidates have raised significant money or the district is marked as competitive by the Cook Political Report. This means that only 30 percent of general election races for those 253 seats will feature a viable challenger. This is only slightly higher than the number of races where incumbents face no challenger at all, guaranteeing their reelection in November.
When combined, the number of open and competitive seats leads to 74 opportunities to realistically increase diversity in Congress. There are certainly cases where a candidate might upset an incumbent in a district that isn’t traditionally competitive but this kind of outcome is rare. One recent example is Rep. Sharice Davids, a member of the Ho-Chunk nation, who unseated the white incumbent Rep. Kevin Yoder in 2018 in a traditionally Republican Kansas district.
If these 74 competitive elections all resulted in a new member of Congress taking office, which they almost certainly will not, that change could indeed be significant for representation. We already know that in 41 of those 74 races regardless of who wins, the member will be white. This won’t result in Congress getting more white, because every one of those 41 seats is currently held by a white person.
Of the 280 districts that have already had their primary, only 31 — 11 percent — could potentially send a member of Congress to the 117th who is a different race than the member serving in the 116th, and in 12 of those districts a member of color could lose to a white person. The main takeaway? In mid-June, around half of all states have held their primaries and in only 19 races (7 percent) is it possible that a BIPOC3 candidate will win a seat currently held by a white person. There is only one contest where it is highly likely that a Latino man (Rep. Ben Ray Lujan (D-NM)) will win a seat formerly held by a white person (Sen. Tom Udall (D-NM), who is retiring) should Democrats win the general election, as expected.
Increasing representation of women appears to be slightly more plausible. Voters from six of the districts will almost certainly elect a woman to a seat currently held by a man, either because the two candidates chosen in the primary are both women or because the race is not competitive and the candidate from the safe party is a woman. However, it is also possible in another seven races that a woman could lose to a man.
Open seats have always been the best opportunity for representation to change because Congress is generally less diverse than the candidate pool. And, indeed, in nine of the 25 open seats, we know that either the race or the gender of the incoming member will be different than the current member (in all cases, either a woman winning a seat held by a man or a person of color winning a seat held by a white person). In six open seats, we are not certain what will happen due to the primaries headed to runoffs, or the challengers have different racial and gender identities from the retiring incumbent (e.g. a white man retiring and the candidates running to take the seat are a man of color and a white woman). In one case a white woman will take a seat held by a white woman, and in the eight remaining seats, a white man will take a seat currently held by a white man.
Given that so many congressional seats are safe and so many members face little to no opposition, all signs point to the 117th Congress being more diverse than the 116th, but only slightly, despite major gains in representation within the candidate pool. Congress is currently about 78 percent white, and 73 percent of the districts that have already essentially selected their member of Congress will be sending a white person — a percentage likely to increase as runoffs are settled.
Do the parties differ?
Currently, Democrats have a much more diverse group of congressional members than Republicans on both racial and gender measures. Will this change in the 117th Congress?
At first glance, it seems that it is likely that the makeup of the 117th Republican class will be more diverse than its 116th class. The party may even make more gains than Democrats. The 116th class of Republicans currently has 22 women, all of whom are white women apart from Rep. Jaime Herrera Beutler (R-Wash.) who identifies as Latina, as well as six men of color. As of June 18, Republicans have nominated more people of color (30), and women of color (15), than Democrats (25 and 14 respectively) as challengers and in open seats. This could dramatically increase representation within the Republican party in Congress.
Table 1: Number of nominees running as challengers or in open seats, by party
Export to CSV
|White||65 (166)||70 (127)|
|BIPOC||30 (33)||25 (83)|
|Numbers in parentheses are the total number of candidates, including incumbents|
However, despite these numbers, it is likely that Democrats actually will likely make more gains for representation than Republicans because their candidates are running in more competitive districts. Roughly 63 percent of these Republican nominees are running in safe Democratic seats, defined here as districts where the Cook Partisan Voting Index (PVI) favors Democrats by a 10-point margin or more4. Of the remaining nominees, only three (Wesley Hunt (Texas-07), Michelle Steele (CA-48), and Darrell Issa (CA-50)) are running in districts where the PVI favors Republicans by more than 1 point.
The Democrats, while fielding fewer people of color as challengers or in open seats so far, see those candidates running in districts they actually have a chance of winning. Only seven (28 percent) are running in districts with the PVI favoring Republicans by 10 points or more, and seven are running in seats where the Democrat is favored by at least a point. Some of this is explained, certainly, by the fact that Democrats tend to perform better than Republicans in districts with fewer white voters. Therefore, Republican BIPOC (Black, Indiginous, and people of color) candidates may come from districts that are unlikely to favor Republicans whereas the opposite is true for Democrats.
How does race affect campaign fundraising and primary outcomes?
What role might campaign finances play in the process of electing more candidates of color to Congress? This is complicated primarily by the fact that different elections cost different amounts of money (due to a variety of factors, including competitiveness of the district, the cost of media buys, whether the candidate is trying to defeat an incumbent, etc.).
In past research, OpenSecrets has shown that candidate race does play a role, although not necessarily a straightforward one, in candidate fundraising success. Lived experiences of candidates also document the difficulties in fundraising for candidates who are not white, who are not male, and who do not have access to wealthy funders and institutions. While we intend to replicate that research this year, an analysis taking into account district-level factors is not yet possible given the limited number of states that have held primaries. However, campaign fundraising totals do tell us something about how race and gender influence a candidate’s electability.
Generally, hundreds of candidates run for federal office, but many of these candidates never get off the ground. Of the 857 candidates running for House and Senate this cycle who raised any money at all, 521 raised more than $100,000, which is not generally enough to win; the average amount raised by primary winners prior to their victory was $1.4 million.
However, despite being a generally insufficient amount to win a primary, $100,000 does indicate that the candidate is legitimate, as opposed to one of the hundreds of people who run but raise little or no money. Of course, a candidate can raise hundreds of thousands of dollars and still have a miniscule chance of winning, but $100,000 is a reasonable threshold to show legitimacy.
To dive into this a little more, we look at woman candidates who lost their primary. A similar proportion of BIPOC and white women won (around 60 percent) their primary contests, although we have already established that many of those candidates are in districts where they are highly unlikely to win the general election. Fifteen BIPOC women and 22 white women raised more than $100,000 but lost in the primary. Of those who lost to non-incumbents, eight of the BIPOC women lost to white candidates, with three of those losing candidates raising more than the winner. This is a different breakdown from the white women who lost in the primary — of the 19 who lost to non-incumbents, only three lost to candidates of color. These numbers suggest that women of color are less likely to beat white candidates than vice versa, although this data is also very preliminary as many primaries have not yet taken place and most fundraising has not yet been completed.
Overall fundraising totals paint an even messier picture. Although we do not like to classify people into only two racial categories (White and BIPOC) due to differences within and among different racial categories and other identity factors, we do so here to account for the advantages white candidates have in congressional contests versus other racial groups. These pre-primary fundraising totals, broken up by whether the candidate was an incumbent, challenger or running in an open seat as well as by gender and race, point to multifaceted trends.
Generally, winning male candidates of color — aside from incumbents — tend to raise more money than winning white candidates in the same type of contests. This does not hold for women, at least so far this cycle. This could be because a successful campaign in districts where a person of color can win might cost more, or because candidates of color need to raise more money than their white counterparts to be successful. In other words, women, especially women of color, may have a harder time raising the money necessary to win a primary.
Table 2: Average total receipts as of pre-primary report for all active candidates, by gender
Export to CSV
|Type of Race Result||White Women||BIPOC Women||White Men||BIPOC Men|
|Open Seat Winner||$1,089,318||$1,197,380||$654,551||$2,022,155|
|Open Seat Loser||$582,279||$329,846||$406,058||$282,413|
When it comes to fundraising from political action committees, a more consistent pattern develops. When excluding challengers and open seat candidates, who historically raise little from PACs, incumbent white women raise more from PACs than incumbents from any other racial group. This could be explained by the fact that nearly all incumbent women running as Republicans are white (and many Democrats this cycle have avowed not to take PAC money, although not as many incumbents). However, when looking at Democratic incumbents only, white female candidates still raise more from PACs than do their fellow incumbents of color. Although this is preliminary data, this may be a place where even established politicians of color have a harder time raising money than white incumbents.
Table 3: Average amount raised from PACs as of pre-primary report, by gender and race
Export to CSV
What comes next?
Primary season is only half over, and at the time this report was published it appears likely that at least one incumbent, Rep. Eliot Engel (D-N.Y.), is likely to lose his primary to Jamaal Bowman, a Black man. While change in congressional demographics is usually a slow process, surprises do still happen. The challenges to diversifying Congress can seem intractable but change does happen — albeit slowly, as we show here.
Those surprises have some things in common. First, successful challengers are often successful fundraisers. Two of the three successful primary challengers outspent their opponent. Should Bowman defeat Engel, Bowman was outspent, but did raise more money from small donors (donors giving less than $200). The candidate raising more money from small donors does not always win, by any means, but small donations can serve as a proxy for enthusiasm from individual voters — as long as those donors are in the candidate’s district.
Bowman also received a higher proportion of his campaign funds from New York than did Engel, as did the two other candidates who defeated incumbents in the primary so far this cycle (Marie Newman (IL-03), and Randy Feenstra (IA-04)). As former Sen. Heidi Heitkamp’s ultimately unsuccessful campaign showed in 2018, a candidate can be flooded with money from out of state, but those dollars do not ultimately show that the candidate has support from their own district.
But, in order to win a congressional seat, a candidate must run for that seat. While the number, and diversity, of the challenger classes of 2018 and 2020 is higher, many of these challengers of color are running for seats they are unlikely to win no matter how much money they raise. To ultimately change the demographic makeup of Congress, challengers of color need to be well-funded when they run in close contests — because as it stands now, those are the contests where we may ultimately see change.
Grace Haley contributed to this report.
2 See the full list
3 In this report we chose to use the term BIPOC, meaning Black, Indiginous & People of Color as opposed to POC, meaning People of Color. We do so to highlight the different relationship Black and Indiginious people have to white Americans as informed by current discussions in these communities. In our 2018 paper on race and campaign fundraising, we did find significant differences in fundraising between racial groups and so we believe that BIPOC is a more accurate term for the purposes of this work. We recognize that the terminology on this is evolving.
4 Pennsylvania, which recently redrew its districts, is excluded from all analysis referencing PVI because we cannot find updated PVI assessments for the new districts.
All data in this report are current through the pre-primary filing deadline for the candidate’s race. Racial identification data is collected by Center staff and relies on self-reported race information. More information about this process can be found here. This report reviews data for primaries that occurred before June 22. Races that resulted in runoffs are excluded from most of the analysis unless the runoff will make no difference to the ultimate gender and race of the member of Congress. A full list of states and their primary dates can be found here.
This report originally appeared on OpenSecrets.org, The Center for Responsive Politics on July 1, 2020, and is reprinted with permission under Creative Commons Attribution-Noncommercial-Share Alike 3.0 United States License