The Forces Tearing Us Apart Aren’t Quite What They Seem
A toxic combination of racial resentment and the sharp regional disparity in economic growth between urban and rural America is driving the class upheaval in American partisanship, with the Republican Party dominant in working class House districts and the Democratic Party winning a decisive majority of upscale House seats.Studies from across the left-right spectrum reveal these and other patterns: a nation politically divided by levels of diversity; the emergence of an ideologically consistent liberal Democratic Party matching the consistent conservatism of the Republican Party, for the first time in recent history; and a striking discrepancy in the median household income of white majority House districts held by Democrats and Republicans.Four scholars and political analysts have produced these studies: Michael Podhorzer, former political director of the AFL-CIO, in “The Congressional Class Reversal,” “Socioeconomic Polarization” and “Education Polarization”; Oscar Pocasangre and Lee Drutman, of New America, in “Understanding the Partisan Divide: How Demographics and Policy Views Shape Party Coalitions”; and Alan Abramowitz, a political scientist at Emory, in “Both White and Nonwhite Democrats are Moving Left.”Podhorzer’s analyses produce provocative conclusions.“Throughout the first half of the 20th century,” he writes in his class reversal essay, “Democrats were solidly the party of the bottom of the income distribution and Republicans were solidly the party of the top half of the income distribution.” In 1958, Podhorzer points out, “more than half of the members of the Democratic caucus represented the two least affluent quintiles of districts. Today, that is nearly the case for members of the Republican caucus.”The result? “In terms of income,” Podhorzer writes. “the respective caucuses have become mirror images of each other and of who they were from Reconstruction into the 1960s.”The shift is especially glaring when looking at majority-white congressional districts:From 1994 through 2008, Democrats did about equally well with each income group. But, beginning with the 2010 election, Democrats began doing much better with the top two quintiles and much worse with the bottom two quintiles. In 2020, the gap between the top two and the bottom two quintiles was 50 points. Since 2016, Democrats have been doing worse than average with the middle quintile as well.The income shift coincided with a deepening of the urban-rural partisan schism.“As recently as 2008,” Podhorzer writes, “40 percent of the Democratic caucus represented either rural or sparse suburban districts, and about a fifth of the Republican caucus represented majority-minority, urban or dense suburban districts. Now, the caucuses are sorted nearly perfectly.”As if that were not enough, divergent economic trends are compounding the urban-rural split.In his socioeconomic polarization essay, Podhorzer shows how median household income in white majority districts has changed.From 1996 to 2008, in majority white districts, there was virtually no difference in household income between districts represented by Republicans and Democrats. Since then, the two have diverged sharply, with median household income rising to $80,725 in 2020 in majority white districts represented by Democrats, well above the $62,163 in districts represented by Republicans.Podhorzer ranks congressional districts on five measures:1) Districts in the lowest or second lowest quintile (the bottom 40 percent) of both income and education; 2) districts in the lowest or second lowest quintile of income but in the middle quintile or better for education; 3) districts that are not in the other four measures; 4) districts that are either in the fourth quintile on both dimensions or are in the fourth for one and the fifth for the other; and 5) districts that are in the fifth quintile for both dimensions.Using this classification system, how have majority white districts changed over the past three decades?“For the entire period from 1996 through 2008,” Podhorzer writes,none of the white socioeconomic groups was more than 10 points more or less than average, although we can see the highest socioeconomic group trending more Democratic through that period. But everything changed dramatically after 2008, as the two highest socioeconomic groups rapidly became more Democratic while the lowest socioeconomic group became much less Democratic.In 1996, Democrats represented 30 percent of the majority white districts in the most educated and most affluent category; by 2020, they represented 86 percent. At the other end, in 1996, Democrats represented 38 and 42 percent of the districts in the bottom two categories; by 2020, those percentages fell to 12 and 18 percent.In examining these trends, political analysts have cited a growing educational divide, with better educated — and thus more affluent — white voters moving in a liberal Democratic direction, while whites without college have moved toward the right.Podhorzer does not dispute the existence of this trend, but argues strenuously that limiting the analysis to education levels masks the true driving force: racial tolerance and racial resentment. “This factor, racial resentment,” Podhorzer writes in the education polarization essay, “does a much, much better job of explaining our current political divisions than education polarization.”In support of his argument, Podhorzer provides data showing that from 2000 to 2020, the Democratic margin among whites with and without college degrees who score high on racial resentment scales has fallen from minus 26 percent to minus 62 percent for racially resentful non-college whites and from minus 14 percent to minus 53 percent among racially resentful college- educated whites.At the same time, the Democratic margin rose from plus 12 to 70 percent over those twenty years among non-college whites low in racial resentment; and from 50 to 82 percent among college-educated whites low in racial resentment.In other words, in contradiction to the education divide thesis, non-college whites who are not racially resentful have become more Democratic, while college-educated whites who are racially resentful have become more Republican, in contradiction to the education divide thesis.Podhorzer makes the case that “the unequal distribution of recovery after the economy crashed in 2008 has been profoundly overlooked,” interacting with and compounding divisions based on racial attitudes:Educational attainment was among the important characteristics associated with those increasingly prosperous places. Add to that mix, first, the election of a Black president, which sparked a backlash movement of grievance in those places left behind in the recovery, and, second, the election of a racist president, Donald Trump — who stoked those grievances. We are suffering from a polarization which provides an even more comprehensive explanation than the urban-rural divide.Changing racial attitudes are also a crucial element in Abramowitz’s analysis, “Both White and Nonwhite Democrats are Moving Left,” in which he argues that “Democrats are now as ideologically cohesive as Republicans, which is a big change from a decade ago, when Republicans were significantly more cohesive than Democrats.”Damon Winter/The New York TimesIn 1972, on a 1 to 7 scale used by American National Election Studies, Abramowitz writes,Supporters of the two parties were separated by an average of one unit. The mean score for Democratic voters was 3.7, just slightly to the left of center, while the mean score for Republican voters was 4.7, to the right. By 2020, the distance between supporters of the two parties had increased to an average of 2.6 units. The mean score for Democratic voters was 2.8 while the mean score for Republican voters was 5.5.The ideological gulf between Democrats and Republicans reached its highest point in 2020, Abramowitz observes, “since the ANES started asking the ideological identification question.”While the movement to the right among Republican voters has been relatively constant over this period, the Democratic shift in an increasingly liberal direction has been more recent and more rapid.“The divide between supporters of the two parties has increased considerably since 2012 and most of this increase was due to a sharp leftward shift among Democratic voters,” Abramowitz writes. “Between 2012 and 2020, the mean score for Democratic voters went from 3.3 to 2.9 while the mean score for Republican voters went from 5.4 to 5.5.”By far the most important shift to the left among Democrats, according to Abramowitz, was on the question “Should federal spending on aid to Blacks be increased, decreased or kept about the same?” From 2012 to 2020, the percentage of Democrats saying “increased” more than doubled, from 31.3 to 72.2 percent. The surge was higher among white Democrats, at 47.5 points, (from 24.6 to 72.1 percent), than among nonwhite Democrats, at 31.2 points, from 41.1 to 72.3 percent.The growing ideological congruence among Democrats has significant consequences for the strength of the party on Election Day. Abramowitz notes that “For many years, white Democrats have lagged behind nonwhite Democrats in loyalty to Democratic presidential candidates. In 2020, however, this gap almost disappeared with white Democratic identifiers almost as loyal as nonwhite Democratic identifiers.”The increase in loyalty among white Democratic identifiers, he continues, “is due largely to their increased liberalism because defections” to the right “among white Democrats”have been heavily concentrated among those with relatively conservative ideological orientations. This increased loyalty has also been apparent in other types of elections, including those for U.S. Senate and House of Representatives. In 2022, according to data from the American National Election Studies Pilot Survey, 96 percent of Democratic identifiers, including leaning independents, voted for Democratic candidates for U.S. House and U.S. Senate.In their paper, “Understanding the Partisan Divide,” Pocasangre and Drutman of New America focus on race and ethnicity from the vantage point of an analysis of voting patterns based on the level of diversity in a district or community.“Republican districts,” they write,are some of the least ethnically diverse districts. But voters within these districts have diverse policy views, particularly on economic issues. Democratic districts are some of the most ethnically diverse districts. But voters within these districts are mostly in agreement over their views of both social and economic issues.Pocasangre and Drutman’s study reinforces the widespread finding “That Republican districts are predominantly white and, for the most part, less affluent than the national average. In contrast, Democratic districts are less white than the average but tend to be more affluent than average.”Pocasangre and Drutman find that the household income differences between Democratic and Republican-held seats continues to widen. From 2020 to 2022, the income in Democratic districts rose from $95,000 to $100,000 while in Republican districts it grew from $77,000 to $80,000, so that the Democratic advantage rose from $18,000 to $20,000 in just two years.Republican districts, the two authors continue, are “conservative on both social and economic issues, with very few districts below the national average on either dimension.” Democratic districts, in contrast, areprogressive on both policy domains, but have quite a few districts that fall above the average on either the social or economic dimension. In particular, of the 229 Democratic districts in 2020, 14 percent were more conservative than the national average on social issues and 19 percent were more conservative than the national average on economic issues.On average, competitive districts tilt Republican, according to the authors:Very few competitive districts in 2020 were found on the progressive quadrants of social and economic issues. Instead, of the 27 competitive districts in 2020, 70 percent were more conservative than the national average on economic issues and 59 percent were more conservative than the national average on social issues.These battleground districtslean toward the progressive side when it comes to gun control, but they lean toward the conservative side on all the other social issues. Their views on structural discrimination — an index that captures responses to questions of whether Black people just need to try harder to get ahead and whether discrimination keeps them back — are the most conservative, followed by views toward abortion.In addition, a majority of competitive districts, 57 percent, are in Republican-leaning rural-suburban communities, along with another 13 percent in purely rural areas. Democratic districts, in contrast, are 17 percent in purely urban areas and 52 percent in urban-suburban communities, with 31 percent in rural-suburban or purely rural areas.I asked Pocasangre about this tilt, and he emailed back:For now, most swing districts go for Republicans. The challenge for Democrats right now is that most of these swing districts are in suburbs which demographically and ideologically look more like rural areas where Republicans have their strongholds. So, Democrats do face an uphill battle when trying to make inroads in these districts.But, Pocasangre continued, “majorities in Congress are so slim that control of the House could switch based on idiosyncratic factors, like exceptionally bad candidates on the other side, scandals, changes in turnout, etc. Democrats need to get lucky in the suburbs, but for Republicans, they are theirs to lose.”Pocasangre and Drutman classified districts as Democratic, Republican, or competitive, based on the ratings of the Cook Political Report in the 2020 and 2022 elections: “Competitive districts are those classified as toss ups for each cycle while the partisan districts are those rated as solid, likely, or lean Democratic or Republican.”The Cook Report analysis of 2024 House races lists 20 tossup seats, 11 held by Democrats, 9 by Republicans, one of which is held by the serial fabulist George Santos, whose threatened New York seat is classified as “lean Democratic.” Eight of the 11 Democratic toss-ups are in three states, four in North Carolina and two each in Pennsylvania and Ohio. Four of the nine Republican tossups are in New York, along with two in Arizona.The changing composition of both Democratic and Republican electorates and the demographics of the districts they represent is one of the reasons that governing has become so difficult. One result of the changing composition of the parties has been a shift in focus to social and cultural issues. These are issues that government is often not well equipped to address, but that propel political competition and escalate partisan hostility.Perhaps most important, however, is that there now is no economic cohesion holding either party together. Instead, both have conflicting wings. For the Republicans it’s a pro-business elite combined with a working class, largely white, often racially resentful base; for the Democrats, it’s a party dependent on the support of disproportionately low-income minorities, combined with a largely white, college-educated elite.One might question why all these cultural and social issues have come so much to the fore and what it might take for the dam to give.The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram. More