Ohio and Realignment
A Look at Ohio’s Political Patterns in 2008 and Beyond
By Daniel Coffey, Ph.D.
Who will win Ohio in 2008? Well, unfortunately, the answer is not contained in this article. But, we might ask a different question: will the results in 2008 provide any information about where Ohio is trending politically? While individual elections are often hailed as a “mandate from the people” for the winning candidate, political scientists have long recognized that some elections are more meaningful than others. Will 2008 be such an election for Ohio?
This leads to another question: what is Ohio? Is it a red state? A purple state? Almost a blue state? In order to answer the question of where Ohio is headed, we must take a step back and explore where Ohio has been. In this article, I will take a look at Ohio’s recent voting history and explore what factors might be driving potential change. Political realignments, political scientists believe, occur with regularity in American politics, about once every 30 years, or once a generation. During these periods, there is usually one critical election that decisively gives one party control over national political institutions. Leading up to the realignment is a period of unusually intense polarization on a few big issues on which the parties take clear positions.
Strong public support, usually along regional lines, provides the winning party with the ability to implement a dramatic set of policy changes, instituting a new governing philosophy.
There have been five major realignments that scholars agree upon in American history, with some debate over the most recent realignment. The alignments occurred in 1800, 1828, 1860, 1896 and 1932. It makes the most sense, applying realignment theory, to define two realigning eras since the New Deal: 1964 and 1994. In 1964, the “Solid South” cracked and, despite a massive landslide, Barry Goldwater took five deep southern states from Texan Lyndon Johnson as civil rights issues began to supplant the New Deal alignment centered on economic concerns.
While some scholars consider Ronald Reagan’s election in 1980 as a realigning election, detailed analysis of election data indicate that it failed to alter patterns of voting in congressional and state elections, which continued to favor the Democrats into the 1990s. In 1994, however, during the “Republican Revolution,” the GOP picked up 52 seats in the House, eight in the Senate and 472 state legislative seats. Newt Gingrich was elected Speaker of the House, and he pushed for the passage of the GOP’s Contract with America, a document that largely borrowed from Ross Perot’s 1992 Reform Party platform, calling for lower taxes and abolishing many federal programs. The GOP successfully reached out to the new voters that Perot engaged in the political process. Importantly, party positions on religious and cultural issues began to harden after the 1994 realignment.
What about Ohio? State patterns are more variable and often out of step with national patterns. Ohio, however, has a fairly good track record in matching national patterns. For example, as noted above, the Reagan Era did little to upset Democratic dominance in Congress or in state elections and, in Ohio Democrats, had one of the party’s strongest periods of dominance, controlling state government for most of the 1980s. In 1994, however, Ohio exactly followed national trends as the GOP took control of the State House and all elected offices statewide. This period of Republican dominance ended in 2006, when Ohio and national politics coincided again. While Democrats won back control of Congress, in Ohio, state Democrats took control of nearly all statewide executive offices including winning their first gubernatorial election in 20 years.
Ohio County Voting Patterns
To more fully explore Ohio voting patterns, it is useful to look at county voting patterns because there is more variance to explore and it is a more sensitive instrument for analyzing trends.
County voting patterns are, in part, a function of the demographic makeup of their populations. What characteristics stand out among the Republican counties? According to data from the U.S. Census, counties that have voted for Republican presidential candidates since 1980 are overwhelmingly white, have higher median household incomes, and are more rural than the state average. In contrast, Democratic counties tend to have substantially higher proportions of citizens living in poverty, lower median household income levels, are more racially diverse (18 percent are African-American), and more urban (91 percent) compared to the rest of the state.
These divisions are consistent with what we observe at the national level. Contrary to some arguments, most notably that of author Thomas Frank in “What’s the Matter with Kansas,” working class and poor whites are not voting more Republican today than in the past. In fact, the exact opposite is true. Political science studies are consistently showing that over the last two or three decades, income is becoming a stronger predictor of voting patterns. County-level patterns hide variation among individuals, but using data from the U.S. Census and surveys of individual voters from exit polls, in Ohio we find higher levels of income predict higher levels of casting a Republican vote. The other patterns are also consistent with the conventional wisdom about American politics. Race is a major factor in predicting voting patterns as to whether an individual lives in a rural versus urban area.
The next question is whether there is any evidence of political change in Ohio. County patterns, especially in presidential elections, demonstrate remarkable stability. For example, of the 72 counties won by President Bush in 2004, 71 were also won by Reagan in 1980 (Reagan won six counties that Bush did not win.) In fact, the correlation between the Bush county vote in 2004 and the county two-party vote is statistically significant in every election since 1980.
The elections are comparable, but there are indications that Ohio counties are becoming slightly more polarized in terms of how red or blue they are. For example, Bush won an average of 59 percent of the vote in the 72 counties that he won. In those same counties, Reagan won 58 percent. On the other hand, if we look at the counties each candidate lost, Reagan won 44 percent of the vote in the counties he lost while Bush won only 42 percent. Using an index of dissimilarity, commonly used in sociology and political science, we find that there has been a mild increase in dissimilarity. In 2004, Republican counties were more different (i.e., there were proportionally fewer votes for the losing candidate) from Democratic counties than in any election since 1984, a landslide election won by Reagan.
Some changes occurring at the micro level might provide clues about what the physical and metaphorical battleground of the future might be. The patterns are not the same across counties. That is, some counties have moved to become more Republican during the last quarter-century, while others have become more Democratic. In the figure below, I trace the Republican share of the vote in three selected counties. Warren is typical of one pattern as one of the fastest growing counties in the state. The previously rural county, located between Dayton and Cincinnati, cast approximately 35,000 votes in the 1980 election compared to just fewer than 100,000 in 2004. This remarkable growth has resulted in a county that is even more Republican than it was nearly a generation ago. In contrast, Franklin County, which cast the second highest number of votes in 2004, has trended in the opposite direction, becoming more Democratic during the last two decades. The break in the trends is most apparent in 2004, when the state split nearly evenly (as did Stark county which matches the state pattern almost perfectly) while Warren nearly matched the Republican rate in Reagan’s landslide election in 1984. However, George W. Bush won 20 percent fewer votes in Franklin.
What Will Define the Next Alignment?
So, what explains the trends that are evident in Ohio county voting patterns? The causes of political alignments and realignments are complex, but there are three general explanations. First, it might be driven by events. For example, the Great Depression ended nearly two generations of Republican dominance at the national level. On the other hand, major issues, such as slavery or civil rights, can shake up the political landscape like an earthquake, making allies of old enemies, as party reputations become fixed by divisions over new concerns. Or, change might be the product of slow demographic shifts when sociological forces (such as generational replacement) finally reach a breaking point in a critical election.
Exit poll data from recent elections would seem to indicate that all three explanations are evident in Ohio. Demographic change might be helpful in understanding why there has been so much stability, as well as understanding the change evident in Warren and Franklin counties. Why? Historically, Ohio’s economy was dominated by manufacturing and agriculture. Unions were a powerful force and Democrats performed well in most of Ohio’s eight central cities. This offset Republican strength among rural counties with agricultural economies, although Republican machines, most notably in Cincinnati, often tipped the balance toward the GOP.
In recent years, the American economy has moved away from a manufacturing base to a service-centered economy. The growth of Columbus best represents this shift from urban to suburban residence. Fueled first by the private automobile, then a declining manufacturing base, and most recently, the “Information Revolution,” suburban sprawl has carried citizens and jobs away from central cities. These cities lost population to less densely populated hinterlands, ironically repopulating nearby rural areas with far flung “exurbs.”
There are different theories about what this means for each party’s future. In recent years, this movement seems, at first glance, to favor the Republican Party. In 2004, Bush won 66 percent of the vote in the ten fastest-growing counties. Overall, Bush won an additional 97,000 votes in these ten counties between 2000 and 2004, which helped to ensure his re-election. This is substantial considering that the total population growth in these counties was 167,000 citizens overall, and Bush’s winning margin in the state was 118,607. As Matt Bai noted in a widely cited (New York Times magazine) article in 2004, this was due to an extraordinary effort using a volunteer GOTV network coordinated from the White House to individual precincts. This was not simply an effort to win, but to capture a certain kind of voter. As Kevin Philips argued in 1969 in the “Emerging Republican Majority,” properly understanding population movements provides an eye toward the future. Not all citizens have the same propensity to move, and those that do move to fast-growing areas reflect a dynamic that can be almost Hegelian in nature. Rove identified exurbanites to be wealthier, somewhat more likely to home school their children or send them to private school, and to be gun owners or at least supporters of gun rights. Such areas also have large populations of Evangelical voters, a core constituency of the Republican Party.
Alternatively, why have urban counties become more Democratic in their voting patterns? We might conclude such areas are more Democratic now because the Republicans have fled the suburbs, leaving a poorer, more racially diverse population. In part, this is correct, but it should be pointed out that in the relatively wealthy suburbs of Shaker Heights, Beachwood and Solon, Bush won only 23, 29 and 46 percent of the vote in 2004.
Why might this be? John Judis and Ruy Teixeira’s book, “The Emerging Democratic Majority,” is based on the premise that the rise of a post-material society in the United States will largely benefit Democrats. A new set of workers who are mostly concerned with the production of ideas and services rather than goods should, the authors claim, be more attracted to the Democratic Party’s social liberalism and moderate-left economic polices, largely making irrelevant the old conflict between labor and management. Professionals, those with advanced and often specialized degrees (and distinct from managers and executives), have traditionally been Republican voters. They point out, however, that the traditional class of doctors and lawyers is increasingly made up of software engineers, teachers, social workers, designers, architects and academics.
And, there is some evidence to support their claim. While these professionals voted for Ford and both times for Reagan, they found only 37 percent of self-reported professionals voted for Bush in 2000. This might help to explain the pattern we saw in Franklin County, with The Ohio State University and many major corporations located in Columbus. Today’s white collar “knowledge worker” is more likely to be a Democrat than a generation ago. In fact, using an admittedly crude model of those counties, Governor Ted Strickland outperformed John Kerry’s 2004 performance in 2006. We see that one of the best predictors turned out to be the percentage of citizens with a college or advanced degree.
The Culture War in Ohio’s Suburbs
This dynamic is being played out in Ohio’s suburbs. As social scientists, we know that demographic data can only explain so much about individual behavior, and citizen attitudes are just as important as one’s race or income level. The suburbs are diverse and our surveys indicate that as many as one in five suburban voters are Evangelical Christians while one in five are self-identified liberals. The exit polls from 2004 revealed the parties are evenly balanced, with 35 percent of citizens identifying with each party and the remaining third of voters identifying themselves as independents.
Our survey data reveals that suburban voters were pulled in different directions by the complex issue environment of the 2004 election, while these same voters broke more heavily Democratic in 2006. Bush held a slight advantage among such voters in terms of issue concerns in 2004. Even in 2004, one in four voters in the suburbs was most concerned with the condition of the nation’s economy. Suburban voters were ambivalent about the economy as 30 percent said the economy was worse four years earlier and 30 percent said it was better. One in five suburban voters said moral values were most important and the same percentage said they were most concerned about terrorism.
In 2004, Bush generally was liked by suburban voters; 52 percent approved of his performance in office, nearly 60 percent approved of the U.S. decision to go to war in Iraq and 61 percent of suburban voters felt the U.S. was safer from terrorism. On the other hand, Bush was routed among another crucial swing group: independents. Suburban independents accounted for 21 percent of suburban voters and voted heavily Democratic in 2004, a sign of things to come in 2006, as Kerry won 63 percent of such voters.
Another look at the data reveals the degree of polarization in the suburbs is grounded in divisions over social and cultural issues. Liberal suburban voters strongly disapproved of Bush and the decision to go to war in Iraq. They rated the condition of the economy as much worse compared to other voters, and Kerry won nine in 10 suburban liberals. On the other hand, Bush won 96 percent of Republican votes, and 86 percent of conservative voters, nearly matching his rates among similar rural voters. Bush was the overwhelming choice among those who attended church at least once a week (64 percent), while Kerry won 60 percent of those attending religious services infrequently. Bush won 76 percent of suburban white Evangelicals. These are not trivial differences; each of the groups mentioned above are sizeable in the suburbs.
We expect then, in 2008, another closely divided Buckeye state. Our analysis of the data indicates that there is a high degree of stability in regional voting patterns, but that Republican and Democratic counties are becoming somewhat more distinct from each other over time. On the other hand, the suburbs remain a crucial battleground, in part because tensions over social and cultural issues are played out among the diverse group of voters who live in these areas. What we can say for certain is that one of the reasons Ohio is a preeminent battleground state in so many presidential elections is that so many of the dynamics of American social and political life are played out here in the Buckeye state.
Dr. Daniel Coffey can be contacted by e-mail at dcoffey@uakron.edu.
------------------------------
Dr. Coffey is the assistant professor of political science and a fellow in the Ray C. Bliss Institute of Applied Politics. His research and teaching interests include political parties, public opinion, state and local politics, campaigns and elections, and research methods.
He is the co-editor of “The State of the Parties”(5th ed.) with Dr. John Green, the director of the Bliss Institute. He has published articles and book reviews in (State Politics and Policy Quarterly and Party Politics.) He is the lead author of the forthcoming (Buckeye Battleground: Ohio, Campaigns, and Elections in the Twenty First Century,) published by The University of Akron Press, along with John C. Green, David B. Cohen and Stephen C. Brooks.
