The two parties that were dominant by the end of Reconstruction have organized American political competition ever since. But everything that had a beginning will eventually have an end. As each party’s incapacities and contradictions become more pronounced, our partisan duopoly is experiencing an acute crisis. Republicans and Democrats are now engaged in an unprecedented form of conflict: Mutual Assured Self-Destruction.

In 2016 the Republicans won, for the first time in a decade, simultaneous control of the White House and both houses of Congress. They are all too aware that the previous era of GOP control, from 2003 to 2007, left few trophies on the mantel. President George W. Bush did nominate, and the Republican Senate confirm, two Supreme Court Justices, John Roberts and Samuel Alito. Comparably impressive conservative jurists ended up in lower federal courts—in all, 340 Bush nominees became federal judges, compared to 387 for Bill Clinton and 334 for Barack Obama.

Otherwise, the political capital Bush vowed to use after his 2004 reelection yielded negligible returns. Much of it was squandered in shoring up support for the war in Iraq, wherein the prevention of terrorism morphed into the promotion of democracy—in a singularly inconducive venue. A good portion of the rest was devoted to the stillborn effort to establish private savings accounts within the Social Security system. Such privatization, a component of the “ownership society” Bush advocated, was the closest thing an administration preoccupied with terrorism and the Middle East had to a domestic policy framework. The ownership society’s chief legacy became government encouragement and facilitation of home mortgage loans to borrowers previously considered excessively risky. The resulting increase in demand from overextended home buyers contributed to a housing bubble, a major factor in the 2008 financial panic.

A plausible hope at the start of 2017 was that Republican leaders at each end of Pennsylvania Avenue would be disciplined and purposeful, desiring to surpass the previous decade’s legacy. If subsequent years are like the first one, however, they will do well to equal it. As was the case in 2003-07, Senate Republicans and the White House agree on placing able conservatives on the federal bench, and have added Neil Gorsuch to the Supreme Court.

The disagreements—between President Trump and congressional Republicans, among those Republicans, and within an exceptionally riven administration—are more notable. Against the conspicuous success of the Gorsuch nomination and confirmation is the conspicuous failure to repeal and replace the Affordable Care Act (“Obamacare”). Loudly proclaiming their opposition to it, Republicans won back the House of Representatives in 2010. They subsequently contended that repeal was impossible with a Democratic Senate. After voters elected a Republican one in 2014, the GOP argued that both houses were necessary but insufficient: only with a Republican president to sign a GOP health care bill could one be passed. In 2016, contrary to nearly every pundit and pollster’s forecast, the nation elected Donald Trump…and Republicans have since gone on to fail repeatedly at what they have described since 2010 as their first order of business.

That Obamacare’s demise is proving as difficult to arrange as Rasputin’s is only one sign of the GOP’s troubles, which extend beyond the Beltway. Roy Moore’s victory over Luther Strange in Alabama’s Republican Senate primary indicates that Trumpism—anti-establishment populism—is a wider, deeper phenomenon than Trump himself. The president endorsed and campaigned for Strange, the former Alabama attorney general appointed in 2017 to replace Jeff Sessions, who gave up his Senate seat to become United States attorney general. In this choice, Trump aligned with, rather than opposed, the GOP congressional leadership, who feared that the outspoken Moore would complicate their 2018 campaign strategies. Despite, or because of, these preferences, Moore won a clear victory by running against the Republican establishment in a way similar to Trump’s primary victories against more conventional opponents in 2016. If the wave that carried Trump to the White House did not crest in November 2016, then the transformation of the Republican Party has only just started. Whether that increasingly populist party is capable of governing, as opposed to posturing, is crucial but questionable.

Class versus Identity

Amidst their troubles, Republicans can find solace in: a) the evidence that the Democrats are even more screwed up; and b) the knowledge that as long as our two-party, zero-sum system endures, power can be won and wielded by the party that is merely less screwed up. The Democrats’ astonishing, mortifying loss in 2016 brutally extinguished any remaining embers of hope-and-change optimism. The 2008 victories did not, everyone now knows, herald a new era of Democratic hegemony. Subsequent intra-party debates over why 2016 happened and how to prevent more defeats like it have been unusually acrimonious. One reason is that Republicans nominated the presidential candidate Democrats thought was least electable, making his victory especially galling. Another is that a close loss lends itself to a long list of “but for” explanations. Trump defeated Hillary Clinton in Michigan, Pennsylvania, and Wisconsin by a total of 77,744 votes, a sliver (.057%) of the nearly 137 million ballots cast nationwide. But for that sliver, Clinton would have won those blue states’ electoral votes, a total of 46, and the presidency.

Above all, Democrats have argued over whether Clinton lost because the party’s coalition is too narrow, or too shallow. Specifically, should Democrats do more, or less, to win votes from the white working class? According to CNN’s exit polls, such voters, typically defined as whites without four-year college degrees, accounted for 34% of the 2016 electorate, and favored Trump over Clinton in a landslide, 66% to 29%. Trump’s success extended a trend: John McCain won 58% of white working-class votes in 2008; Mitt Romney received 61% in 2012.

The case that Democrats should try harder to win white working-class votes is virtually self-evident. In a closely divided democracy, writing off one third of the electorate seems suicidal. Even if Democrats never again receive majorities from this constituency, as they did during the New Deal, smaller margins of defeat, such as those suffered by Barack Obama, are manifestly preferable to those voters’ overwhelming repudiation of Hillary Clinton. The difference may plausibly account for Obama’s two victories as opposed to Clinton’s defeat.

And yet, the “I’m With Her” theory of the case in 2016 was that Clinton’s disregard of the white working class was not only a plausible campaign strategy but her likeliest path to victory. Democratic pollster Stanley Greenberg laments that the Clinton campaign’s “malpractice” led to emphasizing the “rainbow base” at “the expense of the working class.” The campaign’s “fatal conclusion,” he argues, was that “she could not win white working-class voters, and that the ‘rising electorate’ would make up the difference.” It is all too fitting, in this view, that Clinton “finished her campaign with rallies in inner cities and university towns.”

By “rainbow base” and “rising electorate” Greenberg refers to a Democratic coalition of blacks, Hispanics, other minority groups, and whites with college degrees, especially with advanced or professional degrees. (According to CNN, voters of all races with a postgraduate education accounted for 18% of the 2016 electorate, and favored Clinton by a margin of 58% to 37%.) This rainbow coalition was constructed to secure Democratic electoral victories despite working-class whites’ growing estrangement from the party. Greenberg believes that the Clinton campaign bet, and lost, the presidency on rejecting class politics in favor of identity politics.

Why can’t Democrats appeal to the rainbow base and the white working class? It depends on whether the latter are primarily working-class, or primarily white. If economics is central, then Democrats should be able to do much better with this electoral bloc by responding to its aspirations, anxieties, and resentments with a newer, post-industrial New Deal. Both former Vice President Joe Biden and Senator Bernie Sanders, Clinton’s opponent for the 2016 nomination, have insisted that the white working-class voters who favored Donald Trump so decisively were not, by and large, bigots. Rather, Sanders said earlier this year, Trump understood “that there’s a lot of pain in this country.” Greenberg thinks Clinton could have defeated Trump if she had addressed that pain with the sort of economic populism that defined the Sanders campaign—by decrying, for example, “the special-interest and big-money influence that was keeping government from working for the middle class.” Instead, though a Clinton campaign ad showed her hugging the child of illegal immigrants, there was never an ad showing the nominee “with an autoworker’s daughter.”

What Clinton Wrought

If, on the other hand, sociological facts exert more power over votes, lives, hearts, and minds than economic ones, the Democratic Party must choose between its rainbow base and the white working class. And, all those who think a choice is necessary also think there’s only one choice: sticking with and shoring up the base. For Joan Walsh, author of What’s the Matter with White People? (2012), “political corner-cutting” designed to woo less affluent white voters is not only “politically risky, given the party’s majority-female-and-voters-of-color base,” but also “morally wrong.”

The risk? Curtailing identity politics in order to emphasize greater economic security could result in a net loss for Democrats if the number of disaffected rainbow voters who stay home exceeds the number of white Trump voters attracted to a corner-cutting Democrat. Walsh has “come to despair” that economic populism has the capacity to attract significant numbers of white working-class voters, and over “the difficulty of luring [such] voters without turning away the Democratic Party’s loyal base.” And the moral wrong? According to Walsh, any Democratic pivot to the white working class requires denying “the primacy of racism”—denying that Trump did so well with such voters because they were willing to accept, and in many cases eager to embrace, “the very real racism and sexism that [he] deliberately channeled” in 2016.

In other words, retooling their message to make it more appealing to white working-class voters will probably leave Democrats worse off, due to more-than-offsetting losses from the rainbow base. But even if that analysis proves incorrect, it would still be better to lose elections than to win them by running morally bankrupt candidates and campaigns, ones who extenuate rather than deplore racism and sexism. Political parties must determine how to win elections, but also remember why. To win by betraying the fundamental reason for contesting an election in the first place validates the other party’s rejection of one’s own principles. Such victories are worse than defeats, more dishonorable but also more damaging in the long run to a party’s raison d’être.

The remarkable but inescapable conclusion of this line of argument is that Democrats would be better off if there had never been a Bill Clinton presidency. In the 1990s, after Republicans had held the White House for 12 long years, most Democrats were latitudinarian about Clinton’s rhetorical and substantive apostasies from liberal dogma on crime, abortion, immigration, welfare, and other charged issues that intersect identity politics. Over time, however, disdain for Clintonian triangulation grew. Barack Obama spoke to this attitude in January 2008 when, running against Hillary Clinton, he said that Bill Clinton had not changed America’s trajectory the way Ronald Reagan had, which implied that Clintonism amounted to the continuation and validation of Reaganism. By 2016, the party consensus was clear and vehement: Clinton’s compromises in the 1990s were deeply regrettable, if not contemptible. Both Clintons found that securing her nomination and uniting the party required apologizing for things they had said and done two decades earlier.

Because the Obama presidency affirmed identity politics in so many ways, beginning with his being the first black president, his legacy has been subject to relatively muted Democratic criticism. This does not mean that Democrats are happy about compromises made over the past decade. In the years between working in the Clinton and Obama administrations, Rahm Emanuel was a congressman from Chicago (where he is now mayor) and chairman of the Democratic Congressional Campaign Committee. In that capacity, he recruited moderate candidates to run, often successfully, in “purple” rural and suburban districts. As a result, Democrats won House majorities in 2006 and 2008, the only time since the 1994 Newt Gingrich revolution they have done so. Columnist Ryan Cooper now speaks for many Democrats, however, when he condemns the idea that his party might adopt the Emanuel template to fight back in the Age of Trump. The party needs “barnstorming populists,” he says, not the “milquetoast figures” preferred by “unpopular, uninspiring, morally compromised transactional politicians like Rahm Emanuel.”

Since the 2016 election, Columbia University’s Mark Lilla has become the most prominent Democratic critic of identity politics, most recently in his new book, The Once and Future Liberal. His motive is being fed up with “noble defeats.” By contrast, what Democrats like Walsh and Cooper cannot tolerate are tainted victories. Greenberg and Lilla believe that greater Democratic outreach to white working-class voters can pay electoral dividends without repudiating the party’s core beliefs. Walsh and Cooper speak for Democrats who consider such efforts to be futile, counterproductive, and shameful.

The competing arguments cannot both be right. It is possible, however, that both are wrong, or at least too flawed to solve the Democrats’ problems. Barack Obama’s infamous assessment in 2008 shows both positions’ defects:

You go into these small towns in Pennsylvania and, like a lot of small towns in the Midwest, the jobs have been gone now for 25 years and nothing’s replaced them. And they fell through the Clinton administration, and the Bush administration, and each successive administration has said that somehow these communities are gonna regenerate and they have not.

By the end of the Obama Administration, manufacturing employment regained almost all the ground lost during the Great Recession—it stood at 12.56 million jobs in January 2009 and 12.36 million in January 2017. More Americans than that worked in manufacturing in the 1940s, however, when the country’s population was less than half what it is today. Treading water isn’t going to cause the small towns of Pennsylvania and the Midwest to regenerate.

As a result, it’s hard to believe that aggressive economic populism will restore Democrats’ Rust Belt prospects. A New Republic autopsy on John Kerry’s 2004 defeat is truer today than it was 12 years ago. “Democrats have run up against the limits of what they—or anyone else—can do to create and protect good jobs,” Noam Scheiber wrote. As a result, “working-class whites seem more and more aware of the fact that Democrats have lost the ability to deliver stable, well-paying jobs.” If, as Obama discerned in 2008, working-class voters have concluded that no party or politician can deliver security and progress, then they will naturally gravitate to Republicans on the basis of national security, social issues, and cultural affinity.

To Have and Have Not

Two recent books—White Working Class (reviewed on page 27) by Joan C. Williams, a professor at U.C. Hastings College of the Law in San Francisco, and Dream Hoarders by Richard V. Reeves of the Brookings Institution—gauge the chasm between this group of voters and the Democratic Party in ways that give little reason to expect economic populism will repair the relationship. Throughout 2016, for example, Democrats made clear their belief that working-class voters wouldn’t and shouldn’t put their faith in a billionaire real estate mogul. It didn’t work out that way, Williams argues, because the white working class “resents professionals” of the sort over-represented in the rainbow base coalition, “but admires the rich.” Members of the working class are not rich, of course, but find the desire to be rich entirely comprehensible. By contrast, why someone would want to be, say, a community organizer is baffling and more than a little disquieting. Worse, members of the working class have little direct contact with the rich, but a good deal with professionals—much of which consists of being bossed around, second-guessed, and condescended to.

Reeves elaborates why Democratic efforts to win working-class votes by running against the rich have yielded meager electoral returns. According to Occupy Wall Street and the Sanders presidential campaign, the most crucial division in American life is between the 1% and the 99%. (In 2015, a household needed an income of just over $400,000 to be in the top percentile of the income distribution.) Reeves argues, to the contrary, that “the most important fracture in American society” is actually between the top 20% (households with an income above $116,890 in 2015) and the other 80%: the upper middle class, and everyone else. Echoing Williams, Reeves says that Trump supporters without a lot of money “have no problem with the rich,” but detest “upper middle-class professionals: journalists, scholars, technocrats, managers, bureaucrats, the people with letters after their names.”

What’s more, Reeves argues, this working-class attitude may reflect spite or resentment, but is ultimately based on an accurate assessment of how modern America works. Through interlocking policies, especially ones affecting taxes, education, and real estate, the upper quintile has both fortified its advantages and steadily increased its ability to transmit them intergenerationally. Mortgage interest and property taxes, for example, are deductible from the federal income tax. According to a 2014 Urban Institute study, about 70% of the benefits from those two deductions goes to households in the top quintile of the income distribution, compared to 8% for the middle quintile and less than 2% for the bottom two quintiles.

Additionally, good schools and nice houses usually go together. A high property tax base improves a school district’s finances, even as being located in a desired school district enhances a home’s market value. More resources can be devoted to education, and fewer to dealing with disciplinary problems and learning disabilities, in schools that educate advantaged rather than disadvantaged students. Finally, the journalists, scholars, technocrats, managers, and bureaucrats overrepresented in upper-quintile school districts are especially adept and confident when it comes to prodding educators to do better or fix a problem. With all these advantages flowing to children who are already advantaged, it is easier for them to get into the selective colleges that play a large role in determining career paths and marriage prospects, the variables most likely to affect socioeconomic mobility.

The upper-quintile facility at shaping policy in its interests, and at portraying these exertions as altruistic rather than self-serving, extends far beyond the educational system. In 2015, a developer proposed plans to build 224 affordable housing units in Marin County—California’s most affluent—where the median home price is $1.25 million. Newly formed civic groups argued that their opposition to “high density development” was necessary to “protect and preserve the character of the area.” In 2017, these efforts succeeded when a Democratic state legislator from Marin County secured passage of a bill that, as described by the Los Angeles Times, “lets Marin’s largest cities and incorporated areas maintain extra restrictions on how many homes developers can build.” Hillary Clinton won 79% of the vote in Marin County, compared to 16% for Donald Trump.

Reeves’s nightmare is the emergence of an American caste system. To explain it, he draws our attention to the social science distinction between two kinds of economic mobility: absolute and relative. The former is about purchasing power, “a measure of whether you are economically better off than your parents were at the same age,” in his words. If you’re 45 years old, with a household income of $75,000, and your parents had an income of $50,000 (adjusting for inflation and family size) when they were in their mid-40s, then you’ve experienced absolute upward mobility. By contrast, if your household income was at the 50th percentile of the income distribution when you were a kid, and is at the 75th percentile now that you’re an adult, you’ve experienced relative upward mobility. Reeves describes it as “a measure of which rung of the ladder you stand on in your generation, compared to the rung your parents stood on in their own generation.”

Absolute mobility is most easily attained during prosperous times, such as the long economic boom from 1945 to 1973. There’s no simple or easy way to guarantee strong, durable economic growth, of course, but one of its clear benefits is that it can be widely shared. There’s no reason one family’s affluence has to come at any other’s expense.

Relative mobility, on the other hand, is inherently a zero-sum game. No matter what we do, Reeves notes, there will always be exactly one fifth of the population in each income distribution quintile. This means there can only be as much upward relative mobility as there is downward mobility. Everyone loves the idea of a Horatio Alger society where it’s a common occurrence and realistic aspiration for people who spent their childhoods in the poorest quintile of the income distribution to spend their adulthoods in the richest one. That ideal cannot be realized, however, unless it’s also common for people who grew up affluent to end up significantly less so. But the whole point of the lifestyles of the self-satisfied and comfortable—which includes helicopter parenting, hired coaches for college entrance exams and application essays, and regulatory and legislative moats around gentry suburbs—is to eliminate the risk of downward intergenerational mobility. For all the talk about glass ceilings, Reeves argues, the more severe problem is that the upper middle class is building an ever thicker and less permeable glass floor.

New Alignments?

The Democrats’ two main options, class politics and identity politics, become clearer in light of Reeves’s analysis. Both give ample opportunity for the well-credentialed upper middle class to sound noble while acting selfishly, as in Marin County. The jihad against the 1% serves the purposes of the less prosperous 19% of the upper quintile. (Reeves reports that more than a third of the demonstrators at one Occupy march in 2011 had incomes in excess of $100,000.) By claiming they have the same cause and same enemy as the less affluent 80% of the income distribution, upper middle-class Democrats can draw attention away from the policies and political tactics that do so much to strengthen the top quintile.

The “size and strength of the upper middle class means that it can reshape cities, dominate the education system, and transform the labor market,” Reeves writes. Furthermore, it “has a huge influence on public discourse, counting among its members most journalists, think-tank scholars, TV editors, professors, and pundits in the land.” And, most dismaying of all, the transmission of these advantages means they will expand and solidify. “As inequality between the upper middle class and the rest grows, parents will become more determined to ensure their children stay near the top.” This “dream hoarding,” he warns, is how “[i]nequality and immobility…become self-reinforcing.”

Furthermore, the merely affluent are acutely aware of how much better their lives would be if they had the truly rich’s wealth and power. Those rungs of the ladder are within their view. For the working class, by contrast, “The dream is not to become upper-middle-class, with its different food, family, and friendship patterns,” Williams writes. It is, instead, “to live in your own class milieu, where you feel comfortable—just with more money.” Opposition to the rich also addresses distinctively upper middle-class concerns with “positional goods,” such as living in the most prestigious zip codes. Like occupying the top income percentile or quintile, positional goods are inherently scarce. Preventing the Silicon Valley or hedge-fund billionaire from using donations and connections to grease his child’s way into Harvard won’t help a plumber get his kid a scholarship at the state university. But it might open up an admission slot that can be filled by the child of upper middle-class strivers.

The dream hoarders’ relation to identity politics is also neither simple nor pure. Candidate Obama, for the sake of building the largest, most harmonious electoral coalition possible in 2008, attempted to explain red-state attitudes in blue-state terms. It’s “not surprising” that residents of towns that have declined for decades “get bitter,” he said, and “cling to guns or religion or antipathy toward people who aren’t like them or anti-immigrant sentiment or anti-trade sentiment as a way to explain their frustrations.” So, yes, there was fanaticism about guns and religion in the working class, as well as bigotry, but the socioeconomic context made it understandable and even sort of forgivable.

The remarks, delivered at a San Francisco campaign fundraising event, did not endear Obama to the working-class voters who were the subjects of his anthropological riff. Nor did they dispose affluent Democrats to give the white working class the benefit of the doubt. Walsh says she tried in her 2012 book to argue that the “white ethnic” backlash of the 1960s was about that era’s whole skein of social dislocations, not just race. She now believes, however, that the Trump victory, like Richard Nixon’s “silent majority,” was really about whites’ “racial resentment, belief in white superiority, and fear of their coming ‘minority’ status in the United States.” Ta-Nehisi Coates, the “most influential writer in America today,” according to the New Yorker’s George Packer, is even more categorical. In the Atlantic, he contends that “racism remains, as it has since 1776, at the heart of this country’s political life,” and draws a direct line from John Calhoun and Jefferson Davis to Donald Trump.

Slate’s L.V. Anderson spoke for many upper-quintile Democrats when she stated that Trump’s victory caused white liberals to “see our unjust, racist, sexist country for what it is.” This “othering” of Trump voters allows white liberals to define themselves against unjust, racist, and sexist attitudes—thereby congratulating themselves for lacking and opposing them. Williams calls this out as elite whites “displacing the blame for racism onto other-class whites.” And, given how frequently and fervently they extol it, affluent liberals do seem strangely reluctant to incorporate the wondrous benefits of diversity into their own lives. In the 2010 census, Marin County’s population was 80% white and 2.8% black. That those 224 affordable new homes would have left Marin looking a bit more like America must not have occurred to the affluent, highly educated Clinton voters living there.

Coates derides Packer for characterizing the Democratic Party as a coalition of “rising professionals and diversity.” Packer’s “rubric of ‘diversity,’” Coates complains, dismisses the highest aspirations of identity politics, including “resistance to a policing whose sole legitimacy is rooted in brute force” (emphasis added). Williams, however, cautions that such “demonization of the police underestimates the difficulty of ending police violence against communities of color.” More basically, it ignores that police must frequently “make split-second decisions in life-threatening situations,” a challenge that doesn’t characterize the workdays of people with letters after their names.

Joan Walsh believes that because “the ‘rising American electorate’ is still rising,” the Democratic Party should concentrate on its “most loyal constituencies, starting with black women,” rather than emphasize appeals to the white working class. The problem is that one originator of the rising electorate thesis, John Judis, keeps throwing larger and larger buckets of cold water on it. He recently insisted in the New Republic that a “majority-minority” America will not arrive before mid-century, and that if and when it does happen it will not guarantee the Democrats unassailable electoral advantages. The crude, linear extrapolation of recent demographic trends and partisan loyalties is “straight-out wrong and profoundly misleading,” he now thinks.

If Judis is correct, the Democrats’ need for the white working-class vote will be as substantial as their difficulty in securing it. This does not mean that Democrats must resign themselves to a long exile as the country’s minority party. Republican dominance from 2003 to 2007, after all, gave way to Democratic control of the White House and Congress after 2008. But as each party becomes more dependent on the other’s mistakes or unpopularity, the probability increases that new political alignments will eventually supplant the two-party system America has known for 150 years.