Class, Race and Political Strategy in the Rust Belt
“This used to be a thriving hillbilly town”
1. Elections in abnormal times
Elections are complicated Rorschach tests for voters, particularly in abnormal times. We like to reduce them to a single issue; we use locutions like “this election was about X,” as if there were a single factor determining which candidate voters selected. But in political analysis too-quick generalizations can be misleading.
One doesn’t have to be especially sophisticated about politics to see that candidate Trump was a crude, bombastic, least-common-denominator showman. When candidate Trump’s negatives piled up—to his jaw-jutting hubris; openly delusory prevarications; venomous xenophobia concerning immigrants was added in early October “grab ’em by the pu**y” misogyny, many commentators believed that Clinton’s victory was assured.
But voters had other things on their minds. How else do we account for the significant number of Obama voters who swung to Trump? Were they cosmopolitan globalists in 2008 and 2012, only to turn reactionary white nationalist in 2016? Or is a better account this: For a significant slice of voters in key swing states, 2016, like 2008, was a referendum on the status quo. The status quo wasn’t working for them in 2008; nor was it working in 2016. When voters in the rust belt said in 2016 that “the Clintons had their chance,” they were referring to two terms in the 1990s that brought NAFTA which, plausibly, accelerated deindustrialization.
2. Pollster fail
The pollsters didn’t help: Failing to measure intensity of support, they predicted a more than 80 percent chance of Clinton’s victory. Better polling, in principle, could have shown a closer race. Those without access to expensive large-N methodology but who had their finger on the pulse in the rust belt were generally unsurprised by the outcome.
3. Mistakes of the Clinton campaign
The Clinton campaign failed to excite and mobilize (and in some cases even speak to the concerns of) hundreds of thousands of voters in the traditional Democratic base. It is arguable that the campaign had—and the corporate wing of the Democratic Party has—little to offer voters in the rust belt. The Employee Free Choice Act, organized labor’s key legislative ask, has not even come to the floor for a vote, even in the 111th Congress (2009-2011)—with strong majorities in both houses and a Democratic supporter of the measure in the White House. Elements of President Obama’s signature legislative accomplishment, the Affordable Care Act, such as the pre-existing conditions and age 26 provisions, are popular. But by permitting insurance companies to write a bill that failed to cap costs, the bill can arguably be portrayed as burdening working families without solving the most important problem. No wonder working and middle-class voters are suspicious of corporate Democrats!
The Clinton campaign failed to focus on a ground game—mobilizing voters in crucial cities and towns in key states. They were certainly warned of the danger of neglecting this work. Anecdotal evidence suggests that the campaign was so convinced of an overwhelming victory that they did not believe such a mobilization was necessary. Anecdotally: Apparently one major union that offered full-throated support for Clinton sent seven buses full of volunteers to Detroit on the eve of the election for GOTV (get out the vote—that is, eleventh hour mobilization). The Clinton campaign diverted the activists to Iowa, as a feint, to force the Trump campaign to devote resources to the Hawkeye State. Knowing as we do now that Clinton lost Michigan by just 11,000 votes, was this a miscalculation? Did it stem from overconfidence?
After her defeat, Clinton suggested that the campaign had no response to FBI director Comey’s announcement of the revival of the email investigation at the end of the campaign. If this is true—and there is no reason to doubt it—it is an indictment of a professionally run presidential campaign, which should have had solid and well-developed contingency plans for half a dozen potential “October surprises.”
4. Trump’s victory
Certainly candidate Trump’s xenophobia appealed to a hard core of far-right white nationalist supporters. But his victory by mostly razor thin margins in key swing states—Pennsylvania, Ohio, Michigan, Wisconsin—resulted from a broader combination of factors.
First, traditional Republican voters did not boycott him in the general election. Romney voters from 2012 for the most part came out for Trump. Noted historian Mike Davis is right to emphasize the importance of the fact that the predicted aversion of “normal” Republican voters to the candidate at the top of their ticket never materialized. (1) One way to interpret this would be to say that they voted for Trump’s xenophobia and misogyny.
Second, the Clinton campaign failed to run up necessary majorities among traditional supporters of the party. There was not a significant gender gap in the outcome.
Third, a small but significant slice of working class rust belt voters voted against business as usual. Again, as Davis points out, “several hundred thousand white, blue-collar Obama voters, at most, voted for Trump’s vision of fair trade and reindustrialization, not the millions usually invoked.” In a narrow race, that provided the margin of defeat.
5. Keys to the rust belt
The rust belt—the Upper Midwest—remains of vital political importance. As demographic and political patterns presently hold, a handful of crucial states (actually, a handful of vital counties inside those states) hold the keys to presidential electoral success. It is only a slight exaggeration to say that Allegheny County in Pennsylvania (Pittsburgh); Cuyahoga and Lorain Counties in Ohio (metropolitan Cleveland); Wayne County in Michigan (Detroit); and Milwaukee County in Wisconsin now determine the outcome in close presidential elections in America.
Through much of the United States, Washington is viewed with deep suspicion. When unconnected with a higher purpose, said Augustine, the state is nothing more than highway robbery on a larger scale. (2) When was the last time the American state seemed connected to higher purpose? This sensibility is acute in the rust belt, with our towering hulks of shuttered steel mills, machine shops, auto assembly plants and so on. (3)
What, indeed, is rust belt experience of their national state? The US Postal Service: cool efficiency. Social security? Ditto. The military? “Thank god it is there because since the plants and mills closed, it is one of the few decent tickets out of this dying town.” But for the rest—the expensive machinery of the federal state is a distant, inefficient, unresponsive bureaucracy. You cannot control it; you cannot even get a response from it.
6. Rust Belt Anger: A Moment or a Secular Shift?
What are these “Trump Democrats” angry about? I am asked this question time and again, by serious people, activists and analysts and academics, based in Brooklyn, Washington, Paris. What were white, working and middle-class rust belt voters thinking?
A dramatic transformation of the economic landscape has been underway for a generation. The American economy emerged from World War II in a distinctive position. Productive plants in Europe and Japan was devastated by wartime bombing; meanwhile the “arsenal of democracy” meant that American industry was built up during the war. The United States held a dramatic hegemony in industrial production for a generation after the war. Meanwhile, CIO-led organizing, which accomplished the organization of basic industry only on the eve of the war, gave distinctive bargaining strength to U.S. workers. Consequently, from the 1940s to the 1970s—with up to about 35 percent of the workforce unionized—workers’ real (inflation adjusted) wages overall doubled or tripled.
But U.S. economic hegemony could not last; by the 1970s key industries such as steel and automobiles were being challenged by European and Japanese rivals. U.S.-based manufactures spread out to recovering and developing global markets. Global trade in general increased—from about 10 percent of U.S. economic activity in 1960 to about one-third by the turn of the century. Relatively higher wages in American industry put firms, now exposed to global competition, at a disadvantage. The openness of U.S. industry to intense international competition was one factor that led to the broad front attack on unionization; union density has declined to below 7 percent in the private sector. One consequence was that the real wage peaks achieved in the 1970s have been eroded in the decades since, overall by about ten percent.
Still, as late as 2000, there were still more than 17 million manufacturing jobs in the United States. Both before and after the Great Recession (2007-2009), these jobs disappeared at an astonishing rate. 30 percent of manufacturing jobs have been lost since 2000. Of course, there was a long-run trend away from agricultural jobs and manufacturing jobs, toward what is termed the service sector. But the 17 million job plateau, achieved during the Vietnam War, was not “permanently” lost until 2001, the year of the 9/11 attacks.
The dominant narrative is that globalization and “rotten trade deals” were the culprits. The notion is that “politics” is responsible for globalization and that a different political orientation could protect or even return manufacturing jobs to the United States. This dominant narrative has been pressed by the old-line manufacturing unions in auto, steel and related industries and was of course vital to the message of candidate Trump.
A counter-narrative is that technological change is responsible. (4) The soundbite (ITALICS)du jour is that new technology is responsible for 85 percent of manufacturing jobs lost in the United States since 2000.
In fact, it is not easy to ascribe job loss to (BOLD)either the richer articulation of the economy on a global scale (BOLD)or technological change. Changes in transportation and communication technology—such as containerization; computerized inventory control; and ever-more-efficient supply chains—were necessary preconditions for rapid-paced globalization. It isn’t “globalization” or “technology” but both. (5)
Nevertheless, Trump’s effective message on trade resonated with enough voters in the old industrial heartlands of the United States to help swing a close electoral college victory away from Hillary Clinton, whose husband’s name is associated with NAFTA, the 1994 trade agreement. Clinton = NAFTA = devastation of our communities is etched-in-granite common wisdom in the rust belt.
Note that the devastation of whole swaths of the rust belt does not hit white workers or a white “aristocracy of labor” alone. It affects whole communities. When a car plant, steel mill or shipyard is boarded up, all workers in the community—white, black, Latino; men and women—are affected. So are small businesses. Isn’t this one reason why so many blacks and Latinos in the rust belt voted for Trump?Top row L-R) Detroit, MICH. 1992. 1986: Duquesne, PA. 2nd row: Braddock, PA 1986. US Steel ET works in the background, left. Duquesne, PA., 1986 The Duquesne Steel Works, closed. 3rd row: Braddock, PA. 1986. Detroit, MICH. 1992. Chyrsler plant. All Photos: copyright Robert Gumpert
7. Wrenching adjustment
The wrenching adjustment to global competition has proved difficult throughout the United States. But in the rust belt, the adjustment has been acute. The closure of one or two major plants can devastate a rust belt community.
As a friend who commutes past Ford’s shuttered Lorain Assembly plant on the west side of a quintessentially rust belt community said to me recently, “This used to be a thriving hillbilly town.” The plant, which opened in 1958, attracted thousands of workers, many from the rural south. Some eight million vehicles rolled off its assembly lines before it was shuttered in 2005. Everything in the local economy was buoyed by the presence of the plant and its relatively decently paid unionized workforce. But the next generation faces a stark reality.
Joining a branch of the military is one ticket out of economic despair. Sports scholarships to college are prized because the high cost of education means that individuals and families who have to pay typically take on onerous debt. While a college degree opens some professional opportunities—work for big health insurance companies or banks or mortgage firms, or in high tech—it is hardly the all-but-guaranteed ticket it was a generation ago. Plenty of college graduates languish for years, now, in a string of part-time jobs, often defaulting on loans. Those who remain economically active often work multiple low-wage jobs to make ends meet. In Lorain, a bartender at Scorchers, a sports bar—one of the few going concerns on Broadway (the main street downtown)—told me she works four part-time jobs to make ends meet.
For the generation before the advent of the CIO, it was common to work 60 or 70 hours/week in the steel mills of towns like Lorain or Youngstown or Canton. (6) Now, with the demise of auto and steel, the political economy has come full circle: The grandchildren of those who formed the CIO unions now work, once again, 60 or 70 hours/week to make ends meet.
And those are the lucky ones: Every neighborhood knows people who have dropped out of the regular economy. (7) When the jobs on offer are for minimum wage without health or pension benefits, is it any wonder that a portion of the population turns to the illegitimate economy (drugs; prostitution; illegal gambling; small-time robbery, etc.)? Or that another slice self-medicates with drugs or alcohol to deaden the pain of failed dreams? In Ohio, even stable communities have witnessed a rise in drug use. It is common for high school students to know people who have died of drug overdoses. Ohio, the bellwether state, leads the nation in overdose deaths. (8)
8. A different model of politics
Beltway insiders in the Democratic Party are ready to throw over the traditional Democratic base in the rust belt. (9) They can dismiss working class Trump supporters as inveterate racists, as “deplorables.” (10) Their road to victory in 2020 is to mobilize women and African Americans and Latinos. They have the idea that the white working class—or the rust belt segment of the class—is beyond hope, beyond reaching. This dovetails with their GOTV view of “doing politics,” a view that I believe we must reject.
Beltway insiders see voters in general merely as a means to getting themselves elected. Business as usual politically has entailed bombarding citizens with slogans—a kind of least common denominator politics that, in the United States, is particularly personalistic. This works fine for the Democrats as long as they have a personally appealing candidate, such as Obama. But a candidate with baggage such as Clinton could not overcome the anger and disgust that many ordinary Americans have toward Washington (and Wall Street).
We need a different model of doing politics—an organizers’ approach. (11)
Analysis matters. If you believe Trump’s working class base voted for him (BOLD)because of his racist appeals, and all you are interested in is electoral victory, then looking away from this fraction of the working class makes sense.
This is certainly true of a portion of them but it is our contention that a significant element of that base voted against Clinton, against Washington, against business as usual. They were, in fact, so disaffected with Washington that they were willing to overlook Trump’s racism and misogyny: They voted Trump in spite of his crass scapegoating and hubris, not because of them.
But the ordinary Democratic Party, interested solely in winning electoral power, may well decide to look away from Trump voters. They reason that U.S. elections are merely about turnout. We can look forward to a strong argument for getting “back to basics” in 2020—a robust GOTV operation.
But the GOTV orientation is insufficient for changing the political culture, which is what we must do if we are to build a social democratic current in American politics. For a whole generation, Democrats and labor organizations have come to working class America, on politics, only during GOTV. As long as Democrats and labor treat ordinary workers as walk-ons in a drama that has politicians and a handful of labor leaders or interest group heads at center stage, workers will feel alienated from the party, from their organizations and from the national state it self.
The different model would be an organizers’ approach to working class communities. Organizations such as the Progressive Caucus in Cleveland, a successor to the grassroots Sanders campaign in 2016, are involved in neighborhood-by-neighborhood organizing of a democratic membership formation, animated by broadly social-democratic politics. Such groupings can combine electoral aspirations with activist, direct action and civil disobedience style tactics. There is no reason that such formations cannot—once they become strong enough—establish workplace or industry-wide groupings based on the militant minority who are ready, today, to become organized and active.
9. An inside-outside orientation to the Democratic Party
No matter how disaffected with Washington, U.S. workers are not prepared to break with the two-party system. This system will for the foreseeable future necessitate that even radical reformers run campaigns within the broad arenas of the Republican and Democratic parties. This has hardly hindered the Tea Party and Freedom Caucus wing of the Republican Party; it certainly shouldn’t hinder the Sandersist wing of the Democrats. (12)
10. What kind of organization?
We should resist the blandishments of any leaders who contend that the be-all and end-all of our organizing should be re-electing ordinary Democrats (or even Sanders-style social democrats). To push back against global corporate power, the building of strong grassroots organizations is needed everywhere. (13) Whether they are activist unions or community organizations or proto-party style groupings like Progressive Caucuses, these grassroots groups need to be characterized by:
a. Regular meetings aimed at regular action (on issue campaigns such as the Fight for Fifteen or against environmental degradation);
c. An educational component;
d. Maintenance of a proper perspective on electoralism.
In short, we are advocating the creation of an organized social democratic (broadly progressive) force, both inside and outside the Democratic Party. This force should certainly participate in elections. But if it has one foot in Democratic Party electoral politics, it must have another foot firmly planted in movement-style activism: civil disobedience, direct action, workplace organization, etc. This stems from our theory of change—that significant political, social or economic reform does not come from elections but from direct action. Is this not the lesson of the CIO and the Civil Rights Movement?
One danger is that the Democratic Party and organized labor, even if they adopt some of Sanders’ rhetoric going forward, will not alter their approach to “politics.” We need to challenge this. We need to stop treating workers as means (to getting elected) and start treating them as ends. We need to be organizing all the time; and building active involvement from a significant minority of members and their families. This means that the center of gravity of our work will not be inside electoral campaigns–it will be going on all the time. We can and should be building organizations that meet regularly to determine a course of action—be it strengthening union locals or engaging in external organizing or supporting other groups of workers. Both union locals and the incipient Sandersist organizations need to be doing this.
1) “The ‘miracle’ of the mogul’s campaign…was capturing the entirety of the Romney vote, without any of the major defections (college-educated Republican women, conservative Latinos, Catholics) that the polls had predicted and Clinton had counted upon.” Mike Davis, “The great God Trump and the white working class,” Jacobin, February 7, 2017 (acc. April 22, 2017).
2) Augustine, The City of God, Books I-VII (Washington: Catholic University of America, 1950), Book IV, ch. 4. Augustine was the most influential of the western (Roman) church fathers.
3) Bill Fletcher, Jr., and Bob Wing, “Fighting back against the white revolt of 2016,” Verso Press, November 28, 2016, (acc. May 17, 2017), is a generally strong, insightful article, which should be required reading for anyone who cares about the progressive movement. In my view, it is, however, incorrect on this point: Fletcher and Wing contend that candidate Trump did not play to “legitimate concerns of the masses.” “Trump did not address the concerns of most voters. He addressed the fears of many white voters. Those fears…are both economic and racial. The economic fears focus largely on the potential for economic disaster.” As the following section contends, the fear of economic devastation is grounded in the rust belt reality of the past generation. The mills have been closing or radically downsizing since the late 1970s, a trend that accelerated after 2000.
See also Peter Olney, “Go red! Thoughts on the labor movement in the age of Trump,” Stansbury Forum, December 27, 2016, (acc. January 14, 2017), which takes issue with Fletcher/Wing’s use of wages as proxy for class and which advocates constructive engagement with a segment of the white “populist” working class as a task for the labor left.
4) Michael J. Hicks and Srikant Devaraj, “The myth and the reality of manufacturing in America,” Ball State University, Center for Business and Economic Research, June 2015, (acc. April 23, 2017); Federica Cocco, “Most US manufacturing jobs lost to technology, not trade,” Financial Times, December 2, 2016.
5) I owe thanks to Peter Olney for this insight concerning the logistics supply chain. See also Glenn Perusek, “Cleveland: City of Tomorrow?” Belt Magazine, March 2015, (acc. September 10, 2015) on the social consequences of the ongoing development of technologically-based unemployment.
6) In 1919, half of all workers in the steel industry in the United States worked 72 hours per week. See U.S. Senate, Report Investigating Strike in Steel Industries, 66th Congress, 1st Session, Senate Reports, vol. A, no. 289 (Washington: U.S. Government Printing Office, 1919), p. 14, cited in Samuel Yellen, American Labor Struggles, 1877-1934 (New York: Monad Press, 1974), 254.
7) Alarming statistics on the decline in workforce participation, particularly among men, are an indicator.
8) “Ohio leads nation in overdose deaths,” Columbus Dispatch, November 29, 2016, (acc. May 16, 2017). The broader case concerning “deaths of despair,” the stunning rise in mortality for white working class men, is put by Anne Case and Angus Deaton, “Mortality and morbidity in the 21st century,” Brookings, April 10, 2017, (acc. April 13, 2017).
9) I mean this term “Beltway insiders” to apply to the broad swath of Democratic staffers and those who share their worldview: the corporate wing of the party. Of course not all loyal Democratic staffers, whether or not they literally work in Washington, deserve the following characterization.
10) A strong statement of the position: Mehdi Hasan, “Top Democrats are wrong: Trump supporters were more motivated by racism than economic issues,” The Intercept, April 6, 2017, (acc. April 8, 2017).
11) GOTV is frenetic grassroots “get out the vote” efforts by rank-and-file activists and volunteers in the stretch-run of campaigns—knocking on doors, handing out flyers at workplaces, and making calls from phone banks to critical voters and districts. The premium is on talking to as many people as possible, having the thinnest, most superficial discussions possible (“don’t get into it with anybody”). Is it any wonder that ordinary workers feel abused by such a system?
12) A side point is the issue of naming things. I myself am convinced that “Labor for Our Revolution” won’t fly in the heartland. It is an ultra-left slogan. Yes, workers are angry. Yes, they want dramatic change. But they are not ready to vote for a left alternative outside of the two major parties. We have ample demonstrations of this—in, for instance, Green Party candidacies. Today we have a chance to build mass-based social democratic organization. We will squander this opportunity with ultra-left slogans and organizational names.
The adoption of “Our Revolution” and the talk about a political revolution in America is woefully naïve—ahistorical. Does it mean a political revolution like 1776? Is the rewriting of the U.S. Constitution hinted at? Of course not. Sandersism is an effort to revive American style social democracy.
In a blush of enthusiasm for Sanders’ success in the 2016 primaries, let us not forget decades of anti-communism. This has had an effect, and not only among those who came of political age before 1989. We cannot defend “Labor for Our [Socialist] Revolution.”
13) There is of course a tendency of long-standing that insists on remaining outside the Democratic Party on principle. The Democratic Party is the “graveyard of social movements” and the pull of electoralism is inexorable. To change a system dominated by the titans at the head of the firms that dominate global capital, the ultra-lefts contend that a new party must be built (or that protest politics are all that is possible). This is noble purism, honorable but wrong, predicated on a refusal to take responsibility for overall policy. I believe we can hold ourselves to a higher standard. We should be able to do better than merely to decry “corporate welfare” or environmental degradation. We should be able to offer both an immediate program of policy reforms and advance a vision—a maximum program—of a decent social-democratic society. Those who would build a labor or left party outside of the Democratic Party (that is, reducing a tactical question to the level of principle) commit an error of orientation. It is a tactical matter to establish a pole of attraction inside a wing of the Democratic Party. In contrast, purist abstentionists want to stand on the sidelines. See Glenn Perusek, “Between Abstention and Accommodation: Progressives and the Democratic Party in the General Election and Beyond,” Stansbury Forum, July 2016.