Granddaughter of farmers and ferry-keepers, Tina Shannon grew up on the Ohio River in a rivermen’s hotel built a century before. When the hotel’s long porch beams were raised, this spot where West Virginia, Pennsylvania, and Ohio meet was at the start of an arc of growing prosperity: a rising power’s emerging industrial heartland. By the 1960s, when Tina arrived, the arc had curved down. Tina’s grandmother raised grandchildren whose parents were gone, supplementing her Social Security checks by taking in boarders and cooking them meals.

Tina got in a few years of college before she had to drop out to care for a younger sister. She worked as a mechanic’s apprentice and for a handyman before finding her destiny as a children’s daycare provider and dogged political organizer. Tina co-founded a Progressive Democrats of America chapter in Beaver County in the early 2000s, precisely when support for national progressives was collapsing among Democratic voters in this stretch of the Rust Belt: pummeled by loss of unions, loss of jobs, and anger among those left paying taxes about the undeserving others they had come to believe just took advantage.

Across two long decades in which local Christian conservatives spread the Prosperity Gospel and grew in political clout, Tina and allies swam upstream carefully. They nurtured their partnership with the county Labor Council, strengthening reform voices within it when possible. They helped the handful of local environmental activists organize so the environmentalists could pull alongside the unions when possible, yet wouldn’t force a rupture when not. Meanwhile all around them the good jobs stayed gone. Addiction and opioid deaths rose. And in 2016 some voters stayed home and others did not, and in the end Donald Trump was elected with 57 percent of Beaver County’s vote. Hilary Clinton got 38 percent.

Beaver County is one of America’s Middle Suburbs, as classed by the American Communities Project: counties that have medium population density and are below the the national average in education, income, and diversity. These aren’t the upscale suburbs ringing today’s prosperous major metros, full of cosmopolitan knowledge workers. Rather, they are the product of an earlier, industrial heyday: places whose economies stopped booming before the 1970s diversification of U.S. immigration, leaving a tapestry of aging white subdivisions and aging black urban cores.

Counties with this shared history, and hence this shared current profile, are overwhelmingly concentrated in Iowa, Ohio, Michigan, Wisconsin, and Pennsylvania. It’s not coincidental that these were all states Donald Trump flipped. His strength in the Middle Suburb counties, which are so widespread within these states, was systematic, and crucial. He would not have won Ohio, Michigan, Wisconsin, or Pennsylvania without the vote swings in these counties alone.

That consequential electoral result has brought overdue attention to the economic and social challenges within these regions, challenges also tracked by the opioid epidemic—in 2016 Beaver County had the third-highest overdose death rate in all of Pennsylvania’s 67 counties. But if the 2016 election inspired long overdue attention to economic straits and human consequences, it has tended to reinforce a false understanding of whose votes shifted and when. It is not the region’s most impoverished and vulnerable residents who have voted the Rust Belt to the right: They have more often dropped out of voting altogether. It is those among the region’s non-college and some-college voters who are older, wealthier, and whiter, and who for 30 to 40 years have been distancing themselves—literally and figuratively—from the region’s distressed factory-town cores. What shaped 2016 was not sudden flight but the continuation of decades-long trends.

Will those trends continue? The sections that follow begin by tracing the entwined economic, social, and political trajectories that led to 2016, trajectories with significant overlaps in former industrial regions and the rural areas linked to them. I then describe the course that new center-to-left grassroots organizing has taken since 2016 in three different kinds of locales: Romney-to-Clinton upscale suburbs, Romney-to-Trump rural areas, and the declining industrial areas often associated with Obama-to-Trump voters.

Rust Belt regions are not exempt from the surge of local organizing on the center-to-left observed nationwide since 2017. But it is playing out in systematically distinct ways. Passionate new volunteers—here as elsewhere, disproportionately college educated women—are helping rebuild local Democratic politics in former Democratic strongholds. However, the results of November 2019’s municipal elections, which I conclude with, underline the difficulty of that task. The old industrial-Democratic coalition continues to crumble, and the replacement emerging is much more diverse both ideologically and demographically, in ways that bring short-term challenge as well as long-term promise.

From Stronghold to Skeleton: Democratic Decline in Rust Belt Pennsylvania, 1980-2016

America’s Rust Belt is home to the suburbs that unions built: not at once, but in multiple steps. In Beaver County and places like it, from northern Appalachia into the Midwest, the first dense population growth came in the late nineteenth and early twentieth centuries, as booming mining and manufacturing industries drew hundreds of thousands of Southern and Eastern European immigrants, alongside smaller numbers of African Americans from the South. They fought fractious organizing campaigns. The labor protections of the New Deal revolutionized the terms of engagement and cemented Democratic loyalties. By the middle of the twentieth century, these communities were home to a dense associational life: a civic world of lodges, churches, and shop floors in which precinct-level Democratic Party organizing and active union locals went hand in hand.

By the 1950s and ’60s, strong union salaries were making it possible for union men and their families to move up and out into new townships, sometimes miles out from the old mill town cores. In the 1970s, those workers closer to retirement and with more generous pensions stepped up their pace of departure, leaving behind the grime and burdened school districts of the factory-adjacent urban cores. Those remaining were disproportionately African Americans, who had been excluded from choice jobs during the boom years.

“They didn’t want their taxes to go to those schools,” a 50-something white-ethnic union painter who stuck it out in Washington City told me, describing the neighbors who left. There was lots of talk about crime, he added, but it looked just like racism to him.

Southwest Pennsylvania union strongholds like Beaver and Washington counties did not become filled with Reagan Democrats while Ronald Reagan was in office. Across the 1980s, as factories closed and the region hemorrhaged jobs, the steel valleys’ laboring people cleaved even tighter to the party they identified with the unions rather than the bosses. Jimmy Carter carried these counties by a 10-point margin, Walter Mondale by 20, Michael Dukakis by more than 25. But around 1990 things finally shifted. There was exhaustion, and exodus from place and party alike. As factory towns cratered, the portion of the county’s population living in rural areas rose, from 24 percent in 1980 to 27 percent in 2000. In 1980, per capita income placed Beaver County in the 54th percentile nationally. By 1990, it had fallen to the 22nd percentile.

Research has shown that increasingly, in modern America, poorer states vote more heavily Republican, but not because poor people vote Republican. Rather, it is the more comfortable voters in hard-hit states who are the firmest Republicans. The disadvantaged there are more likely to vote Democratic, if they vote at all. The same is found at the county level: Middle-income voters shift rightward as the rate of reliance on public assistance in the communities around them rises.

From 1992 onward, the Democratic vote share in every aging industrial or coalmining county in western Pennsylvania fell with each successive election. As the chart below shows, only Allegheny County—home to the city of Pittsburgh—was different. Here, Democrats gained voters in near-in suburbs at exactly the same pace at which they lost votes in former industrial areas, creating a two-decades-long flatline in election margins instead.

Rust Belt southwest Pennsylvania’s move away from electing Democrats at the top of the ticket spread slowly down the ballot. The last Democratic senator to carry Beaver County was incumbent Senator Bob Casey in 2012, by a margin of 2 percentage points. That was a far cry from the 24-point margin he earned there in 2006, itself a stark drop from the 52-point margin his father, Governor Bob Casey Sr., earned in 1990.

The generation whose loyalties were forged in the fire of active labor struggle was dying off. And those just behind them—the one-time party stalwarts who had moved up and out of the factory towns—were turning their backs, a Democratic state representative who held his seat for 20 years told me. By the 1990s, he said, “I’d lost them. I’d go out there [to the subdivisions] and I’d say, Bill, you know me. ‘Can’t do it.’”

Western Pennsylvania had 73 state house representatives in Harrisburg in the 1990s. As late as 1994, three-quarters of them were Democrats. By 2018 the region was down to 55 representatives, and more than half were Republican. With the exception of Allegheny County, only one quarter of the region’s representatives are Democrats.

Map by Ben Forstate

Deindustrialization and decline were not the end of the story. Of the once-industrial counties, those that have retained enough population to impact statewide elections—and the Electoral College—are the ones that have some new source of economic dynamism: an interstate highway that makes commuting to Philly or Pittsburgh possible, or enough “eds and meds” (that is, schools and hospitals) in the county seat to keep the local labor market alive. Working-age adults in these counties are still disproportionately likely to be unionized, but for the majority this is not through the Steelworkers and United Mine Workers. Rather, it is pink collar (female-dominated jobs) and public sector unions as well as the building trades that are currently helping pay the mortgages in much of Beaver, Washington, and Westmoreland counties in Pennsylvania’s southwest and in Luzerne and Lackawanna counties in the northeast. These are the inhabitants of today’s suburbs: nurses, teachers, and cops; as well as electricians and plumbers.

Union ties, however, no longer translate into Democratic votes. The imprint of Pennsylvania’s anthracite and bituminous coalfields are still visible in the geography of union membership, as the above map makes clear. But a different geography has come to shape political information and allegiance: the rural divide. Back in 1980, the portion of union members in a given county was tightly correlated with Democratic presidential vote share. Counties that were more rural were somewhat more likely than urban ones to vote Republican, but only somewhat. Steadily over the course of four decades, both of these things have changed. By 2016 union membership rates were nearly unrelated to which way a county swung, while how rural its population was had come to be tightly entwined with Republican allegiance. The chart below shows that, in both the loss of union impact and the growing rural divide, the 2016 election simply continued a generation-long trend.

The political arcs of rural and Rust Belt America overlap: unsurprisingly, since their histories do too. You don’t get a large rural population from a sparse agricultural past. You get it where capital has flowed in and then on to somewhere else, leaving the grandchildren of boomtime arrivals behind. Of the 60 million total rural residents in the United States today, one in four lives in a Rust Belt state (Pennsylvania, Ohio, Michigan, New York, Indiana, Illinois, Wisconsin).

Standing in Donald Trump’s America, it is easy to suppose that the election of Barack Obama and racialized white rural identification with the GOP arrived hand in hand. But as the chart above reveals, not only did the growth of the rural-urban divide predate Obama: That gap slowed rather than accelerated in 2008. Although Obama’s ill-chosen description of rural Pennsylvanians who “cling to guns or religion” got much press, his campaign’s strategy reflected his less-often-quoted preceding sentence, calling Democrats to show up and fight for votes in the “small towns” where “the jobs have been gone now for 25 years and nothing’s replaced them.” By election day, the Obama campaign had 81 field offices spread across 40 of the state’s 67 counties.

As this scatterplot shows, Obama in ’08 improved on John Kerry ’04 in counties urban and rural alike, including significant gains in some of Pennsylvania’s very most rural—and whitest—counties. In only six of Pennsylvania’s 67 counties did Obama underperform Kerry: all six of them in the southwestern ex-industrial crater.

Once more, the birthplace of American steel was a harbinger of things to come. For 2009 would see nationwide an explosion of local groups determined to keep government spending away from those they saw as undeserving. “Tea Party” activists were uniformly white, older, and middle- or upper-middle income, but not necessarily college-educated. That is, they were drawn from exactly the demographic that the union-driven upward mobility of the postwar boom years had helped to create in the Rust Belt, which is also the demographic least likely to believe anti-black racism exists, and most likely to believe it is whites and Christians who face the most discrimination today.

Obama’s vote margin fell from over 600,000 in Pennsylvania in 2008 to under 300,000 in 2012, and this time his losses were heaviest in rural terrain. Four years later, the longstanding trends of decreasing union/Democratic boost and increasing rural/GOP alignment passed a tipping point. Donald Trump’s gains over Mitt Romney in Beaver, Washington, Fayette, and Lawrence counties alone topped 45,000 votes. He won the state by 44,000.

After 2016: New Grassroots Explode in the Romney-to-Clinton Suburbs

Since 2016, it’s been a different set of suburbs making headlines. Upscale metropolitan suburbs have been the epicenter of political change in America under Trump. Many of these were “Romney-to-Clinton” areas that flipped Democratic at the presidential level in 2016. The migration down-ballot since 2016 of voters’ turn away from the GOP at the presidential level in these communities has been the mirror image of the movement down-ballot of the rejection of the national Democrats in rural and then Rust Belt regions over the previous generation. But it has happened at ten times the pace.

This was not automatic. Rather, it reflects the intentional and active labor scores of thousands of people channeled into thousands of new local groups nationwide. As Theda Skocpol and I reported in these pages in February 2018, national initiatives like the Women’s March, Indivisible, Sister District, and Swing Left were the visible pieces of a much broader and more diffuse trend. In community after community, following November 2016, likeminded acquaintances came together, debated how best to push back against Trump’s agenda and repair the system that allowed for his victory, and dived into action.

From January 2017 onward the new activists were on the march. Anywhere there were college-educated women with time and resources to spare, some number of them shifted that time and those resources into politics full on. They held town halls and recruited candidates. They knocked on doors for Democrats for Congress, Senate, governor, and state legislature—in their own districts if they had Republican incumbents, in adjacent districts if they did not. They wrote postcards to voters by the armful, so many that when the October 2018 deadline to mail them in time for midterm races loomed, grassroots Facebook pages buzzed with tips on which local post offices hadn’t yet run out of postcard stamps.

The virtuous circle of grassroots growth and down-ballot campaigns—campaigns that proliferated as groups sought to contest “every seat, every election”—spurred extraordinary levels of engagement. Some 155 million doors were knocked on behalf of Democratic candidates in 2018, fully 40 percent more than in the presidential year of 2016. And unlike what had become standard in Democratic races, with campaign-run “ground games” disappearing the day after the election, now formal ties among core volunteers both preceded the campaigns and outlasted them, carrying lessons forward to the next.

These patterns repeated nationwide, with similar intensity in demographically similar places. In the upscale suburban “collar counties” around Philadelphia (Montgomery, Bucks, Delaware, and Chester), the impact was transformative. What had been firm white-shoe Republican terrain, especially for state and county elections, saw multiple municipal offices not just contested but also flipped in 2017. A year later, Republicans lost 12 state house seats and four state senate seats in the collar counties, along with a fifth state senate seat in the demographically similar Romney-Clinton suburbs north of Pittsburgh. Each one reflected an 18-month backstory of local grassroots groups organizing, recruiting, door-knocking, and more.

There are Librarians Everywhere: The “Resistance” in Romney-to-Trump Rural Terrain

Anti-Trump grassroots groups emerged in Pennsylvania’s rural and Rust Belt counties, too—in smaller numbers than in the upscale suburbs and exurbs but present nonetheless.

To understand where groups are, and where they aren’t, we need to begin from the recognition that American political space is not flat red or blue but increasingly fractal. While that spatial clustering may not matter for statewide vote totals, it does matter for local organizing, for which a critical mass of like-minded people is crucial.

Zoom in on any “red” district and you’ll find some Democratic-majority centers: small cities or big towns with more density, diversity, and graduate degrees. Of course, rural county seats are not just metropoles in miniature. A significant proportion of the civic-minded middle-aged and retired women here are firm stalwarts of conservative churches and are passionately pro-life, among other things: They are not going to jump into joining an Indivisible group whatever their feelings may be about a President’s tweets. Yet old coal lands do have old environmentalists. College towns have some left-leaning professors even when the colleges are Christian and conservative. Even the smallest cities, population 5,000 or so, tend to have at least one church where faithful with same-sex spouses or gay kids feel welcome.

Circa 2016, such people did not necessarily add up to a visible progressive presence within local civic life. You had some people like Tina Shannon, the ferry-keepers’ granddaughter, taking lonely stands against the Iraq War, railing against Wall Street foreclosures, and (as in Tina’s case) campaigning hard for Bernie Sanders in 2016. And you had others choosing not to make waves. In particular, individuals whose educational and economic status placed them within the upscale and hence Republican layer of Rust Belt and rural communities, but whose personal views aligned better with national Democrats, often chose not to take a public stand on politics. Until Donald Trump’s election raised the stakes, that is.

Consider Armstrong County, about an hour northeast of Pittsburgh. Two thirds of the county’s population is rural; less than 1 percent is employed in agriculture. “When I was growing up there were still a lot of union jobs here,” recalls local librarian Connie Ruffner. “It was a Democratic county. The companies up and moved out and just left people: dropped people on their heads.” Michael Dukakis was the last national Democrat to reach 50 percent here. In November 2016, Hilary Clinton got less than 23 percent. Trump supporters were walking tall.

And librarians were thinking about fascism. “I work in a library and my friends are librarians and teachers,” Connie Ruffner told me, explaining that Trump’s campaign, with its anti-immigrant rhetoric and attacks on the press and intellectuals, had raised echoes in her mind of 1930s Europe. “So when he was elected I was talking among other librarians and teachers. It was suggested that one thing we could do is start book clubs, and learn more about what’s happening, and they could be like seed groups if we need a more actual resistance. That may sound extreme but it seemed warranted.”

Connie stumbled across the Indivisible Guide online, and it became their club’s first reading. The suggestions for action looked doable even for a small group. “My friend put out the word to people that she knew to meet at a pizza place,” she says. “Twenty-eight people showed up; at the meeting after that, 42.”

Their 2017 was busy, and their 2018 more so. Indivisible Armstrong collaborated to host a debate among three primary candidates seeking to defeat Republican Congressman Mike Kelly in what was then an R+26 district according to the Cook Partisan Voter Index. They joined a new multi-county network of emerging grassroots groups and Democratic organizations including Lawrence County Action, Slippery Rock Huddle, Marcellus Outreach Butler, and Butler County Democrats. They marched alongside the Armstrong County Democratic Committee float and local labor leaders in the Labor Day Parade. They joined Indivisible Mercer County in front of Kelly’s district office in Sharon, population 13,000, for a candlelight vigil to protest child separations at the border.

They also phone banked and canvassed weekly for Susan Boser, a college professor from neighboring Indiana County who gamely mounted a campaign against incumbent Glenn Thompson in what became, after court-ordered redistricting in January 2018, an R+43 congressional district. “We gained 16 points in deep red Armstrong County,” Connie told me with pride. Boser’s 31 percent of the vote was not only an improvement on Hillary Clinton’s showing, but an improvement over the last time a Republican congressman had faced a challenger in the region. You might see Boser’s two-to-one loss as a rout. For Connie Ruffner and friends, it was evidence that progress was possible.

“This is a very, very red county,” Ruffner said. “So much so that one of Susan’s campaign people said he had a hard time talking people into letting their neighbors know they were Democrats. They were afraid. Just seeing Democratic yard signs was a big deal.” Trump supporters are still in the majority. But, “we went out as Democrats to the Dayton fair, and people stopped to talk to us and they were willing to be seen talking to us: That wasn’t true in the past.”

In places like rural Armstrong County, the relationship between groups like Connie’s and existing local Democratic committees has been largely collaborative. Indeed, often women (and occasional men) whose trajectory into political life parallels Connie’s have ended up running the show. The Democratic committees of two counties adjacent to Armstrong are both now chaired by such women: one an English professor, the other a retired critical care nurse. Rural broadband, regional investment, water quality, and health-care access: There is broad overlap among former insiders and passionate newcomers about what a Democratic agenda should look like in these places.

Where the Past is Not Even Past: The New Grassroots in the Old Industrial Core

In former industrial Democratic strongholds, like Beaver and Washington counties in southwest Pennsylvania and Luzerne in the northeast, it’s been more complicated. Not only has a labor-linked, patronage-channeling Democratic Party machine existed here within living memory: Pieces of it are still there. And some of the folks running it have policy preferences that look more like Donald Trump’s than like Nancy Pelosi’s—to say nothing of the policy preferences of Indivisible’s national leaders, eager to prod Pelosi even further left on immigration and ICE.

Here, too, 2017 saw grassroots groups spontaneously emerge, and 2018 saw members elected onto county Democratic committees and into leadership positions within them. The committees now encompass folks who joined Democratic politics across very different generations. Union stalwarts in their eighties; conservative insiders in their sixties; a handful of activists far to their left (Tina Shannon and her husband are both committee people now); and the new-grassroots-linked librarians, retired teachers, and moms with post-college degrees. And, for the first time in a long time, some young people are also present.

The campaign of Darcelle Slappy highlights how grassroot growth is diversifying Democratic politics here. Slappy is a licensed cosmetologist and mother of five, Big Beaver school board member, wife of a union steelworker, concealed-carry permit holder and mother of a gun crime victim—and, as an African-American woman, represents the face of the Rust Belt working class least seen in media coverage. She also knows “everyone, everywhere” from her travels as basketball referee in the powerhouse Western Pennsylvania Interscholastic Athletic League, something that also carries special meaning here. Mike Ditka, Tony Dorsett, and Ty Law all hail from Beaver County factory towns and came up through the WPIAL; most famously of all, so did Joe Namath, who grew up in Beaver Falls just a generation before Darcelle did.

When a Republican incumbent state legislator captured both Republican and Democratic slots by persuading registered Democrats to write him in in the primary—the kind of thing only truly possible in a district with a large number of registered Democrats who years ago switched to voting Republican—groups new and old mobilized to recruit Slappy for a third party run. Those who got her on the ballot and knocked on doors on her behalf included Beaver County Young Democrats; local chapters of NOW, the NAACP, and the Progressive Democrats of America; and the post 2016 grassroots group Beaver County Voice for Change.

An African-American woman running as a member of the Green Party in a district that voted strongly for Donald Trump, Slappy wasn’t expected to come close, and she didn’t. Still, as one longtime SEIU organizer kept telling me, elections lay tracks.

Slappy carried four out of six precincts in her hometown of Beaver Falls. The town is 80 percent white and went for Hillary Clinton by 16 points—again, the notion that Trump’s stronghold is the hardest-hit white people in the hard-hit Rust Belt is wrong. What is true, though, is that the trends that feed cynicism and disengagement in such places have not reversed. In the 2015 municipal election, Beaver Falls’s turnout was 9 points below the county average. In 2019, a year after Slappy’s run, countywide turnout was up 1 percent, and Beaver Falls was up 2. That’s both impressive and tiny. For campaigns like Slappy’s to have an impact at scale would require major new efforts in labor organizing and more. Donald Trump’s election has not transformed politics in Beaver County’s least well-off communities.

Slappy’s recruitment, though, reflected a broad pattern. In places where the new grassroots are swimming against a decades-long tide of Democratic erosion, they have rarely sought to challenge the Democratic incumbents who remain, who are holding on by dint of personal ties and largely conservative positions. But in primaries for open seats, the new groups have supported a far wider range of recruits. Progressive women veterans and veterans of color; young white guys with local college degrees and good union jobs who support the Second Amendment and think Donald Trump’s an idiot; non-profit leaders, nurses, and more nurses: All have been among the newcomers contesting school board, state house, and county-wide seats with grassroots support in rust belt Pennsylvania.

The results are emerging cohorts of active and visible Democrats in the former industrial heartlands who look much more like the breadth of today’s Democratic party both statewide and nationwide—including in their ideological spread—than do the old-school white ethnic male insiders who defined the party in these places for so long.

The 2019 Elections: Tectonic Shifts Hit County Offices

The old industrial regions of Pennsylvania, Michigan, and Wisconsin saw significant recovery for Democrats in the 2018 midterms. In Pennsylvania, Democratic Governor Tom Wolf, Senator Bob Casey, and sundry congressional challengers all outperformed Hillary Clinton dramatically in one-time industrial strongholds. It was plausible to expect that 2019’s municipal elections would show the kind of scrappy, emergent coalitions profiled above winning back former Democratic voters at the county level as well.

That is not what happened.

Instead 2019 saw the further collapse of the aging white ex-industrial Democratic coalition, as middle-income Middle Suburb voters’ disengagement from the Democratic Party, which began at the top of the ticket a quarter century ago and has been travelling down-ballot ever since, finally hit the last of the countywide offices. New faces who the new committee members and grassroots had gotten onto the ballot didn’t win: And this time, conservative Democratic incumbents didn’t either.

County commissions in Washington, Westmoreland, Armstrong, Greene, and Luzerne all flipped to GOP control, as did positions like district attorney, sheriff, and recorder of deeds in counties where Democratic “row office” incumbents had hung on even after commissions had gone red. Pennsylvania’s industrial-and-building-trade-linked, socially conservative, patronage-wielding Democratic county machines seem to have finished their long collapse. The corruption scandals endemic to them were never enough to end them, but declining union sway and the disenchantment of their older white core with the national party’s positioning finally did. There just aren’t enough of their aging voting base left, and too many of those who remain aren’t voting Democratic anywhere on the ballot anymore.

The consequences of this pattern are not a simple rightward swing. The same crumbling of Democratic machines brought victories by progressive challengers in key cities: like Paige Cognetti, who challenged the Scranton Democratic machine as an Independent—for a seat left vacant by a bribery scandal—and won election as the city’s first female mayor. Meanwhile, in Pittsburgh, the “tough on crime” incumbent Democratic district attorney, whose 20 years in office have seen a black male incarceration rate that is double the already high national average, came close to being unseated by Lisa Middleman, a longtime public defender promising radical reform likewise running as an Independent. Despite the fact that the incumbent DA appeared on both Democratic and Republican tickets, he only outperformed Donald Trump countywide by 16 points. (I worked the polls for the challenger on Election Day, at a Romney-Trump precinct way out in the suburbs. She won it.)

Meanwhile, in a reverse mirror of developments in former Democratic strongholds, 2019 was the year in which Romney-to-Clinton upscale suburbs’ rejection of the Trump-era Republican Party hit the county level. Democrats gained control of county commissions or councils in Bucks, Chester, Delaware, and Lehigh for the first time in memory. The speed of Democrats’ electoral takeover did not reflect some mass change of heart on marginal tax rates or zoning. It was the combination of upscale suburban voters’ refusal to identify with Donald Trump’s GOP along with intense local grassroots labor being used to build persuasive alternatives. And the reason the grassroots could build them so quickly is that they were organizing in communities where the range of potential Democrats look a lot like them in priorities and preferences—and were already connected to them through informal networks from swim team carpools to county bar foundations.

Just as it would be wrong to read the number of wealthy suburban seats flipped blue as evidence of a leftward ideological swing, it would be wrong to read the Democratic row-office collapse in the old-industrial northeast and west as evidence of a rightward one.

I noted above that the conservative Democratic incumbents who the remaining machine insiders backed came up short, and the far more diverse (racially, generationally, ideologically) newcomers who grassroots-linked activists had recruited did too. Equally striking is that candidates of both kinds came pretty close to winning. In Beaver, Washington, Westmoreland, and beyond, almost all Democrats on the ballot in 2019 outran Hilary Clinton by eight to ten points or more. They just didn’t outrun her quite enough to win. So one reading would suggest that as politics become fully nationalized here, the state of play looks more like a two-to-four point Democratic deficit than a Trump-sized Republican blowout.

Once more, it matters that political geography is fractal. These countywide totals mask trends pushing in different political directions. The suburbs within the Middle Suburbs—those subdivisions built by the men who moved up and out of factory towns and away from the Democratic Party, which today are home to not only the final few of that generation but to those among their children and grandchildren who have prospered enough to stay, but not enough to move away—remain deeply skeptical of the Democratic Party. Poverty rates here are half the Beaver County average, population totals are relatively stable, and locals want to keep it that way. While 2018’s gubernatorial and senate results indicated a willingness to split ballots for familiar and competent centrists with a D after their names, 2019’s losses by county-level candidates in places like these across the state suggest the Democratic brand has not yet hit bottom with this slice of Rust Belt voters.

The places in Beaver and similar counties where Democrats did post improved numbers in 2019 are the upscale, dense, micro-cosmopolitan neighborhoods where doctors, lawyers, and other professionals had been largely Republican up through 2016. Places like these tend to be home to members of groups like Beaver County Voice for Change: the “middle aged women, professionals or former professionals who are now local Democrats’ top volunteers,” according to Beaver County Democratic Committee co-vice chair Andy Bosh (himself a post-2016 newcomer).

That is, the activated-against-Trump new grassroots are as hard at work here as elsewhere. In places like Beaver County they know that means building coalitions with people who are not just like them. But their dance of connection with the potential Democrats around them is complicated because the universe of potential Democrats in places like this is complicated. You have organized labor and the building trades strongly supporting fracking and the ethane “cracker” plant—and environmentalists and worried moms opposing them. You have African-American voters in former factory towns who know they’ve rarely had a voice within county leadership—and occasional-Democrats out in the subdivisions who are invested in a story of self that denies white racism altogether.

What it adds up to are regions that aren’t “coming back” to Democrats in their topline vote totals any time soon: not because the Rust Belt is a political monolith, but precisely because it isn’t. Each county is internally diverse in consistent ways, with moving parts headed in opposite directions. The new Democratic coalitions whose outlines are emerging are both awkward and incipient, and for now they are not growing fast enough to compensate for the final crumbling of the old one around them.

The union fall-off; the rural divide; the narrow boundaries of belonging embraced by some within a generation that inherited a unique conjuncture of U.S. global power and labor bargaining power, and who live today in regions hammered by the loss of both. Donald Trump created none of these trends, yet he turned out to be a perfect politician to exploit them. November 2016 was the result. None of the underlying drivers of these trends have dissolved over the past four years, and it should not surprise us that Trump’s potential route to victory has not either.

This does not mean that the Rust Belt and rural communities of the old industrial Midwest should be written off by Democrats or progressives. Rising leaders are building new stories there right now. The clock will not stop at 2020’s particular balance of forces, any more than it stopped two generations ago when the world looked finally ready to provide reliable reward for the sweat of working people in these steel valleys—and then didn’t.

Tina Shannon, Connie Ruffner, Darcelle Slappy, and their impassioned fellow volunteers are acting with the conviction that new political coalitions are achievable in the former industrial heartland. Sooner or later they will be right.