Skip to content Skip to footer

Is Ohio part of a winning strategy in 2020?

Article published:
Working America canvassers in Columbus, OH.

Ohio has traditionally been seen as a crucial swing state, but a closer look reveals that other states should take priority to defeat Trump in 2020.

Left-wing politics in the United States have changed dramatically since 2016. The remarkable success of the Sanders primary campaign revitalized the left. Then, of course, Trump’s victory (in part a consequence of his ability to position himself as a populist outsider against a Washington “establishment” candidate) brought a white nationalist to power. This defined the 2018 midterm elections as a referendum on Trumpism; the large turnout in that race signaled the importance of this question to the U.S. electorate.

In this article, I argue that the left should approach the 2020 election by carefully distinguishing the tasks in the primary and general election phases. While we are able to fully air alternative policy proposals in the primary phase, once a general election candidate emerges, the need for a broad united front against Trumpism should be emphasized. This united front will necessitate compromises—we need to be able to work with people who we agree with only on some issues. The successful implementation of this orientation should strengthen the confidence and fighting spirit of the left and labor.

But there are electoral and demographic realities that must be confronted in the general election. Traditionally, Ohio, Pennsylvania and Michigan have been seen as politically similar Midwestern swing states. On closer inspection, this is far from the case. This article takes a close look at the three states and contends that Ohio has long been demographically distinct from the others. The differences (weaker unionization; smaller Black and Latino populations; and relatively greater portion of southern migrants) have made Ohio more conservative in presidential elections for more than a generation. Consequently, in a close 2020 general election, Ohio should not be a priority for resources.

The predicament of the U.S. left

A historical weakness of American national politics has been the truncation of our politics—particularly the closing off of alternatives to the left. In contrast to other developed democracies, we have lacked a significant social democratic force to contest elections and advocate for policies such as national health insurance. For the left, the alternative has long appeared to be either to strike out on our own—to run third party campaigns without a hope of victory, in order to raise issues or stand on the principle of independent political action—or else to collapse our efforts into the campaigns of mainstream Democratic Party candidates. In that case, participating in electoral politics meant accommodating to political outlooks far from our principles.

Your inbox needs more left. Sign up for our newsletter.

That conundrum of either principled abstentionism or accommodative collapse was broken with the remarkably successful Sanders campaign in 2016. Thirteen million voters in the primaries registered their support for a clear left alternative. Since 2016, a new, coherent left has emerged, with the growth of DSA as an organization and the election of capable, charismatic representatives of the left to Congress.

Without entering into Democratic politics, this rise of the left would have been more difficult. The challenge has long been for the left to break out of the cul-de-sac of speaking to and organizing in a small world of those already, essentially, convinced. Sanders raised issues that made it possible to put universal health care into the national debate. Because he was a legitimate candidate for the nomination, even those on the right had to take him, and the ideas, seriously.

Nevertheless, let’s keep in mind how far we have to go in order to be able to legislate from the left. Today, only a handful of democratic socialists hold seats in Congress. But the Senate remains Republican and conservative are for the most part unified. We will need to win dozens upon dozens of Congressional campaigns in coming years.

Building a broad front against Trumpism

Part of the rise of progressives since the Sanders campaign has been a reaction against Trump in the White House. The president’s open appeals to the racist right, his coarsening of political dialogue, his lack of personal integrity all contributed to the growth of an amorphous “resistance” that has swept millions of people into action. Thousands of women in Ohio, for instance, participated in local and national demonstrations against Trump—many engaging in political action for the first time. And they translated this into electoral significance in the 2018 midterm election.

Peter Olney and Rand Wilson laid out an orientation for the left in 2018. While rejecting whitewashing the Democrats on policy grounds, Olney and Wilson advanced the idea that the central task in the 2018 elections was to build a united front against Trump’s bloc, a “dangerous far right nationalist movement led by Trump and his backers.” Focusing on flipping the House to the Democratic Party did not mean forestalling tasks associated with building the left. On the contrary, we must also develop potential progressive candidates, work to strengthen an embattled labor movement, and more. But Olney and Wilson stressed the key task—opposing Trumpism—and connected it with a winnable immediate objective, namely, taking the House of Representatives for the Democratic Party.

Our question going forward is how to put this orientation into practice during the primaries and the general election in 2020. Several conclusions can be drawn from the 2018 elections at the federal level.

2018 was a referendum on Trump’s presidency

Two-thirds of voters responding to exit polls said that their vote was influenced by their view of President Trump. Thirty-nine percent said they cast ballots to demonstrate opposition to the president; 26 percent said they were showing support. Voters chose the issue of health care as their top concern (about 41 percent, nearly double those who listed the economy or immigration, the next highest concerns) said by a 58-34 margin that the Democratic Party was more likely to protect coverage for people with pre-existing conditions.

Both Democrats and Republicans are mobilizing

The Democratic Party’s retaking the House of Representatives is the single most important aspect of the “blue wave” quality of the 2018 midterms. Not only did the Democratic Party retake control of the House; the popular vote for the party’s candidates reached an unprecedented level.

Sources: U.S. House of Representatives, History, Art & Archives, “Election Statistics, 1920 to Present,” at (acc. November 9, 2018). 2018 partial statistics from “House election results: Democrats take control,” New York Times, November 9, 2018, at (acc. November 9, 2018).

Yet, we should read these numbers carefully. The shift in percentages toward Democratic congressional candidates is significant. But the absolute numbers of votes are, too. Not only are these numbers the basis for the 2018 Democratic recovery of the House, they are also the basis for the mobilization that could defeat Trump in 2020. However, at the same time, we should also note the dramatic rise in Republican votes in the midterms. In the 2018 midterms both Democratic and Republican turnout rose. In the general election in 2020, should the Democrats and their allies fail to mobilize (and do so in a focused way), these numbers foretell the reelection of President Trump.

Two stages of electoral engagement: The primaries and the general

I would like to suggest that we distinguish sharply the two phases of the presidential election campaign—primaries and the general election. The primary phase, as Tom Gallagher so eloquently puts it, is for advocacy of fundamental ideas. The Sanders campaign was a way to present policy proposals from a democratic socialist point of view to a wide audience. As Gallagher has written, “In the process of taking some unusually ‘big issues’—universal health insurance, a minimum wage that is actually a living wage, the history of U.S. overthrows of democratically elected governments, etc.—right into America’s living rooms in the debates, the Sanders campaign arguably revolutionized the entire process…” But once the Democratic Party settles on a nominee, the task will change. Whether this candidate is more or less appealing to the left will become, at that moment, secondary to the necessity of uniting against Trumpism.

We can anticipate that this will be a difficult argument for many on the left. The process of competition among Democratic primary candidates tends to harden partisan views. How many supporters of Sanders in 2016 could not countenance supporting Clinton in the general election? Nationally, some 12 percent of Sanders primary supporters voted for Trump in the general election.

But it will be precisely here that the left ought to make sophisticated arguments for engagement. While it might be difficult to swallow the many compromises required to support a candidate from the corporate wing of the Democratic Party, the necessity of pushing back effectively on the right becomes paramount. For example, how many advocates of basic reproductive rights will, in the wake of Alabama’s draconian new anti-choice laws, see this as a crucial difference between the parties? To be in a position to build the left in the longer run, I contend that we need to engage in the issues of the day—and the fight against Trumpism will be the determining question during the 2020 general election.

Being part of a broad movement against Trumpism, no matter who the candidate, presents us with positive opportunities for building meaningful left politics, both during the campaign and subsequently. The general election season can be a time for us to build progressive bases, to argue for our policy proposals, and to expand our organizations. We can gain the respect of fellow activists as we work side-by-side in the electoral campaign. Principled dialogue on issues is integral to our efforts to advance a progressive alternative which, in any case, is a longer-run project.

Thus, a vital transition will be necessary once a clear Democratic Party frontrunner emerges. There is nothing wrong with an honorable and respectful conflict between wings of the front. But we need to unite to register and mobilize voters for the general election no matter who becomes the candidate at the top of the ticket. The youthful core of Sanders supporters, for instance, can play a decisive role in the 2020 general election, by working against Trump.

If we were operating in a parliamentary system with proportional representation, this turn would not be necessary. In that case, we could ride with our party’s candidates into the general election; the task of compromising would fall to the legislative members of our party—the process of setting differences aside, or not, in order to form a government. But in our system, the question will be for or against the president in the general election.

How things look from Ohio

In Ohio, national and international media attention has focused on the closure of GM’s Lordstown complex. But there is nothing new about large factories closing in Ohio. It is as if a long series of economic explosions has occurred. Economic transformation appears as almost a natural phenomenon. Whatever different politicians have said, there is a sense that politicians can do nothing positive to help the local economy.

Source: Bureau of Labor Statistics, State and Area Employment, Hours and Earnings, at (acc. June 4, 2019). Data are for April of each year.

As with the United States as a whole, even after the deindustrialization of the 1980s, manufacturing employment remained relatively stable through the 1990s. As late as 2000, there were still more than one million manufacturing jobs in the state. In two phases—before and then during the 2007-9 Great Recession—manufacturing employment dropped off significantly. It was below 800,000 by 2007 and hit its low point in 2010, at 615,000. Since then, there has been a gradual rise, but manufacturing employment remains below 700,000. In other words, it is about one-third below what it was as late as 2000.

In response to this structural upheaval, individuals tend to put their heads down, focusing on their own situation. If things go well, it feels like hitting in the lottery. If disaster strikes, a crisis has to be dealt with. But there is little confidence that any assistance will come from above. A demoralized sense of being politically lost prevails.

Many who voted for Obama because they desperately wanted change, and then switched to Trump for the same reason, now feel utterly at sea. By my count, eight counties—Montgomery, Sandusky, Stark, Ottawa, Wood, Portage, Erie, and Trumbull—voted for Obama in 2012 but switched to Trump in 2016. With the exception of Montgomery, which includes Dayton, these are countries that stretch from Youngstown to Toledo, along the northern third of the state. Northern Ohio is historically the home of heavy industry—steel in cities like Youngstown and Canton; cars in Lordstown; and countless smaller machine shops or plastics or chemical processing facilities. That is, these Obama-Trump counties are precisely the places hit hard by industrial dislocation.

As one former Lordstown worker put it, “I don’t know where to go. It seems like no matter what [Trump] does or tries to do, it doesn’t work out. Well, now what? What the heck do we do? Do we go back to beating our heads against the wall? Or do we try something different?”

As dislocating as things are for manufacturing and other workers who are attempting to navigate the post-industrial job market, things may be worse in the countryside. Ohio agriculture is heavily dependent on soybeans and corn. Soybeans are the most important crop—5 million of 13.9 million acres are planted in them. In value terms, soybeans ($2.53 billion) and corn ($2.25 billion) far outpace all other agricultural products (the next highest is hay at $420 million). For most of the period since 2014, both commodities have been priced below a reasonable break-even point$4/bushel for corn and $10/bushel for soybeans. Low prices are a product of global oversupply, fueled in part by competition from such countries as Russia and Brazil. But Trump’s trade disputes, particularly with China and Mexico, have pushed many farmers over the edge. “Prices for soybeans and hogs plummeted after those countries retaliated against U.S. steel and aluminum tariffs by imposing duties on U.S. products like oilseeds and pork,” Jesse Newman and Jacob Bunge report in the Wall Street Journal.

Trump’s bailouts for farmers hurt by retaliation in the trade wars—$12 billion in 2018 and another $16 billion announced in May 2019—are a necessary lifeline for farmers throughout the Midwest. But a position of dependency upon the federal government is far from acceptable to farmers.

Although the lifelines keep some farmers from the worst, they cannot save others. Throughout the Midwest, farm bankruptcies are now occurring at levels not seen in a decade—since the Great Recession. According to the U.S. Department of Agriculture, farm debt has risen to $409 billion, the greatest figure since the 1980s.

In Ohio and among all farm households, more than half are now losing money. In 2018, the average farm household’s farm income nationally was negative $1,548. Consequently, the generally low price of soybeans, exacerbated by the trade war with China, means that many farmers are dependent on the government’s bailouts to survive.

One Ohio farmer, Christopher Gibbs explained to CNN that he took the money in order to keep his farm in the black. But, he said, the need for such supports is “an indication that the president’s policy wasn’t working. Or certainly wasn’t working fast enough.” He also underscored that the tariff monies do not come from the countries with which Trump is engaged in a trade war, but rather, from American importers—who then pass the increased costs along to consumers. While farmers appreciate the supports, says Gibbs, “we would rather have trade. We would rather have our markets back.”

Source: USDA, National Agricultural Statistics Service, 2017 Census of Agriculture—State Data (Ohio), at,_Chapter_1_State_Level/Ohio/st39_1_0005_0006.pdf (acc. May 26, 2019).

Farmers also express the worry that, with American access to important markets like China blocked, other producing countries—like Brazil and Argentina—might “permanently displace American suppliers.”

The question of course is what impact these developments will have politically on farmers and farm communities. Even though the number of farmers is relatively small, the lives of small towns are dependent on farm performance. While many farmers will bravely seek to tough out the “short term” pain of the trade war—and continue to support the president—others may turn against his approach.

Incidentally, along with West Virginia, Ohio is ground zero for the drug abuse crisis—the spike in fatalities from overdoses. The easy availability of powerful opioids is a necessary condition for this epidemic. But the devastation of personal lives—the loss of hope and a sense of control over one’s own economic destiny—that stems from decades of economic shocks—is also responsible. This is another symptom of the deep economic shock that has hit. “In 2017, Ohio had the second highest rate of drug overdose deaths involving opioids in the United States. There were 4,293 reported deaths—a rate of 39.2 deaths per 100,000 persons.” This figure is well over double the rate for the country as a whole.

Progress in one’s own life seems disconnected with politics. People may feel grievances against past politicians who made promises they couldn’t fulfill. But there is little sense of optimism or enthusiasm going forward. Instead, a sense of alienation pervades working class communities, a feeling of unjust arbitrariness.

Door-to-door canvassers in 2020 are going to confront hopeless lethargy about elections. Do elections have anything to do with the problems we confront? Some politicians or bureaucrats might screw things up. But the prevailing sense is that no one can point to any political figure or program that has helped them economically.

Mobilizing smart: Get out the vote in key swing states

It is remarkable that Trump’s overall approval rating, even with the economy booming, has hovered around 40 percent. Previous presidents’ approval, even unpopular presidents in difficult times, tended to be higher. That Trump’s approval would be so low even as the economy hums along at nearly full employment is a measure of how poorly his amalgamation of xenophobia, nationalism, misogyny and arrogance plays with the American people.

Yet, Trump should not be underestimated. We should acknowledge that Trump’s 2020 campaign is perfectly viable. He has the ability to raise a massive war chest from deep-pocketed donors; his approval among Republicans remains strong. Indeed, he has reshaped the Republican Party in his own image.

Many things went wrong for the Democrats in the general election in 2016; we would do well to avoid the mistakes made then. One of the vital errors, of course, was the failure to emphasize on-the-ground Get Out the Vote (GOTV) efforts. Thus, in a close election, key Midwestern swing states, Wisconsin, Michigan and Pennsylvania, went to Trump by a razor-thin margin of 77,000 votes.

We would compound the error if we failed to take into account the yawning differences between swing states. My view is that there are vital swing states in the Midwest—Pennsylvania, Michigan and Wisconsin—that should be a focus.

The longstanding Michigan-Pennsylvania-Ohio pattern

Although Ohio has gone to the presidential winner in every election since 1964, pollsters from both parties do not see it this way anymore. But this tactical observation can obscure the fact that for a whole generation Michigan and Pennsylvania have voted more Democratic in presidential elections than Ohio.

Sources: Federal Election Commission, Federal Elections 2016 (Washington, December 2017), at (acc. May 7, 2019); (acc. May 7, 2019); (acc. May 7, 2019).

This pattern, the long-term difference between Ohio on the one hand and Michigan and Pennsylvania on the other, is something of an iron law. In key elections going back a whole generation, a gap exists between Ohio and the other industrial states—Ohio has tended to be more Republican by three to seven percent. If this historical trend holds, in a close election, it will be easier to flip Michigan and Pennsylvania than Ohio.

The magnitude of the difference can be viewed this way: In 2016, Hillary Clinton lost Wisconsin, Michigan and Pennsylvania by thin margins—22,748, 10,704 and 44,292 respectively. In comparison, the Clinton loss margin in Ohio was 446,841. Had Clinton tallied only the 2012 Obama/Biden vote total in the three close states, she would have won all three states. Not so in Ohio—Obama’s 2.828 million votes in 2012 would not have overcome Trump’s 2.841 million in 2016.

Source: Results are reported at “Presidential Election Results: Donald J. Trump Wins,” New York Times, August 9, 2017, at (acc. May 19, 2019).

It is not only Michigan, Pennsylvania and Wisconsin that make more sense as targets for our electoral resources than Ohio. After the three key Midwestern states, it then makes more sense to prioritize Florida, Arizona and North Carolina as states that can be flipped before coming to Ohio. Even targeting Georgia would make more sense than Ohio—given the stark reality of Trump’s margins in 2016.

What accounts for the strong pattern that Michigan and Pennsylvania are more Democratic than Ohio?

Commentators speak of Ohio as having an aging, white population that is less educated than other Midwestern states. Let us examine these and other demographic factors, in attempting to account for a longstanding pattern that transcends the 2016 outcome.

An elderly population?

Age, alone, does not appear to correspond to the Michigan-Pennsylvania-Ohio voting pattern.

Sources: Author’s calculations from Resident population for selected age groups: 1980 to 1989 (in thousands), Population division branch, U.S. Bureau of the Census, March 1995, at (acc. July 8, 2019); Frank B. Hobbs with Bonnie L. Damon, 65+ in the United States, U.S. Department of Commerce, Bureau of the Census, Current Population Reports, Special Studies, P23-190, 1996, at (acc. July 10, 2019); U.S. Census Bureau, Quick Facts, at (acc. July 3, 2019).

Education level?

While there is a small difference between the three states in educational level (measured as percentage of the population with a bachelor’s degree or higher) in 2010, this has not been historically the case.

Sources: U.S. Census Bureau, A Half-Century of Learning: Historical Census Statistics on Educational Attainment in the United States, 1940 to 2000, Table 6. Percent of the Total Population 25 Years and Over with a Bachelor’s Degree or Higher by Sex, for the United States, Regions, and States: 1940 to 2000, at (acc. July 8, 2019); U.S. Census Bureau, Quick Facts, at (acc. July 3, 2019).

Three other factors: Unionization; race and migrants from the south


One element of an explanation of Ohio’s Democratic lag could be differences in unionization rates. Through time, Ohio has a slightly smaller unionization rate (density) than Pennsylvania and Michigan. The difference is persistent through time. The detailed data provided by Hirsch and MacPherson shows this for the period since the 1980s. This is confirmed by earlier figures—in 1953 Michigan’s overall unionization rate was 44.6 percent; Pennsylvania’s 40.6 and Ohio’s 38.0. (See Leo Troy and Neil Sheflin, U.S. Union Sourcebook (West Orange, NJ: IRDIS, 1985), table 7.2.)

Source: Barry Hirsch and David MacPherson, Union Membership and Coverage Database from the CPS, at (acc. July 8, 2019).

Race? A complicated factor

A second factor, racial composition of the population, appears to contribute. Through time, Michigan has a higher proportion of African Americans than Pennsylvania and Ohio. But Ohio has also led Pennsylvania consistently.

Source: Campbell Gibson and Kay Jung, “Historical Census Statistics on Population Totals by Race, 179 0 to 1990…” Population Division, U.S. Census Bureau, Washington, DC, September 2002, Working Paper Series No. 56, from (acc. July 8, 2019).

But if we also consider the percentage of the population that is Latino, the pattern emerges. In the 2010 U.S. Census, the percentages are—

Source: U.S. Census Bureau, Quick Facts, at (acc. July 3, 2019).

Data suggests that race is part of the explanation for the Michigan-Pennsylvania-Ohio pattern.

The “Moonshine Line”? Ohio’s larger southern population

A third factor is the proportion of the population that stems from the south. Throughout the twentieth century, the industrial economies of Michigan, Pennsylvania and Ohio attracted population from elsewhere. But the portion of Ohio’s population from the south has been markedly greater than that of Michigan or Pennsylvania.

Source: Gregor Aisch, Robert Gebeloff, Kevin Quealy, “Where we came from and where we went, state by state,” New York Times, August 19, 2014 (updated), at (acc. July 1, 2019).

On this measure, Ohio looks more like Indiana than Michigan or Pennsylvania—in fact, through the whole post-WWII period there was a greater portion of Ohioans from the south than even Indiana.

In sum, Ohio’s unionization rate is lower than Michigan’s or Pennsylvania’s; its Black and Latino populations are smaller and its southern population larger than the others. These factors appear to contribute to the longstanding pattern of voting less Democratic in presidential elections.

Where should resources be focused?

Ohio was once vital to presidential election victory. In the whole post-World War II period, only Kennedy in 1960 lost Ohio but won the presidency. But two important changes make this axiom passé: First, Ohio’s share of electoral votes has dropped. In 1960, it was worth 25. Today, it is down to 18. Second, three states that Al Gore lost in 2000 shifted toward Democratic candidates consistently in 2008, 2012 and 2016: Nevada, Colorado and Virginia. These three states today are worth a total of 28 electoral votes (in 1960 together they counted for less than Ohio—21 electoral votes). In Nevada, the growing power of The Culinary, UNITE-HERE Local 226, in Las Vegas/Clark County, helps to account for the shift. In all three states, there is a significant and rising percentage of the electorate that is Latino. Holding the new blue states—Nevada, Colorado and Virginia—is a priority. So, of course, is flipping the closely contested Midwestern states—Wisconsin, Michigan and Pennsylvania.

I would contend that in the 2020 general election, resources should be focused in the three Midwestern states that Trump carried by razor-thin margins in 2016, as well as the three “newly blue” states of Virginia, Nevada and Colorado. Unfortunately, Ohio’s demographics make it more challenging for any Democratic presidential candidate. In a close general election, it should not be a priority.

Changing public opinion in Ohio, like changing the country as a whole, is a long-term task. Principled advocacy of progressive policies, such as universal health insurance and clean energy to combat climate change can eventually win, but not without years of painstaking work. We should approach the 2020 general election with urgency: Defeating Trump’s program of white nationalism is our critical present task. Succeeding will require a broad united front; and success will invigorate progressive forces for the longer-run struggle.


About the Author