Tuesday November 13, 2012

Blue Cities, Red Cities

A professor's analysis of urban voters shows majority of cities went for Obama based more on identity politics than the economy.

An analysis of 2012 U.S. Presidential election results, by assistant professor of political science Josh Ambrosius

Bill Clinton sailed to victory in 1992 on the famous campaign mantra "It's the economy, stupid." No doubt Republicans and Mitt Romney believed the weak economy would be their ticket to unseating the nation's first urban president. But analysis of urban voters shows identity apparently trumped the economy in 2012 — and it was the strength of this urban advantage that carried Obama to an easy Electoral College victory and an eked-out popular vote in 2012.

This urban advantage creates an image most people are familiar with: A national map of county election results that is very red with pockets of urban blue. But just like states come in two colors, there are "blue cities" and "red cities." While Obama handily defeated Romney in the core central cities across the U.S., nearly one-quarter of these cities still supported Romney. This total carried by Romney, however, is down 45 percent from the number Bush won eight years ago. The demographics of red and blue cities are predictable, but Romney's red cities surprisingly had better economic indicators: lower unemployment and higher shares of manufacturing jobs. My research shows that it was the politics of identity — not economic performance — that produced an urban landslide for Obama.

The Republican Party predicated its election fortunes on its candidates' supposed economic smarts rather than direct appeals to identity, a strategy that failed to produce victory and will haunt the party for the foreseeable future. The Democrats too must worry about the urban appeal of the candidate who takes Obama’s mantle.

I analyzed 95 "core" counties — those containing the central cities of America’s major metropolitan areas with populations of 250,000 or more — in each of the past four elections.  The sample includes counties containing megacities like Chicago and Los Angeles, but also mid-sized cities like Louisville, Ky., and Dayton, Ohio. An analysis of core counties is not the same as restricting vote totals to within city limits, but election data is reported at the county level and other county-level data is easily obtainable. 

These 95 core counties accounted for just over 34 million votes in 2012, or shy of one-third of the electorate, of which about 21 million, or two-thirds, went for Obama. Remarkably, George W. Bush was able to carry 40 of these 95 counties with about 43 percent of the vote in both 2000 and 2004. In 2008, Barack Obama limited John McCain to just 18 counties and 38 percent of the vote. In 2012, Mitt Romney was able to increase this total minimally to 22 counties and 39 percent of the vote. 

Several core counties are reliably red across all four elections, including those of the Jacksonville, Fla.; Phoenix; and Salt Lake City metro areas. Counties that were red under both Bush elections but blue under Obama include the cities of Dallas, San Diego, and Cincinnati.  Ones that went blue in 2008 but red in 2012 were few, including Omaha, Neb.; and Grand Rapids, Mich. Despite a bad economy, Romney was only able to flip four counties from blue to red.

What are some of the differences between core counties won by Romney and those won by Obama? Not surprisingly, Obama's blue counties have higher levels of minorities, young people under 30, people with a college degree, and same-sex households. Blue counties also have lower proportions of veterans and followers of culturally conservative religious traditions, including evangelical Protestants and Mormons. Obama's counties, on average, boast nearly doubled populations and 12 percent higher median incomes.

Interestingly, Romney's counties have slightly lower September unemployment rates (7.5% vs. 7.9%), the last announced before the election, and higher shares of manufacturing jobs as a proportion of total employment (8.75% to 7.97%).  In other words, Romney's counties actually experienced better economic performance under an Obama presidency.

Statistical analysis shows that the socio-demographic and cultural "identity" variables outlined above steered the urban vote. Together these variables account for approximately three-quarters of the variation in the vote proportion for either candidate in these counties. The two economic variables, unemployment and manufacturing, together account for less than 8 percent of the variation in candidate support. (The manufacturing variable may itself be more of an identity variable too if it proxies union membership.) In other words, our identities color how we see the economic data and feel its effects when voting for president.

I draw three conclusions from this analysis:

1) The practice of identity politics is alive and well. The president identifies with the urban electorate in a way Republicans cannot and likely will not anytime soon. Obama's appeal to urban populations is greater than recent Democratic nominees Al Gore and John Kerry. Obama's identity cut the GOP's previous urban victories in half and increased the Democratic share of core county votes by nearly 5 million over Gore's total in 2000.

2) The urban electorate is identifying less and less with the Republican Party. President Bush was not a city guy by any stretch of the imagination, but he carried 20 more core counties than McCain and Romney. Is this demonstrative of a societal trend, evidence of Bush "architect" Karl Rove's genius, or simply because of Obama's urban appeal? Whatever the reason, the GOP must compensate for this deficiency or risk losing future elections by even greater margins in urban areas. It will require softening their platform's stances on social welfare policy, immigration reform and same-sex marriage and diversifying the top of the ticket. This latter strategy might be easier said than done considering former Republican National Committee chairman and African-American Michael Steele's pre-election comment that the GOP is "not ready for folks like me."

3) While Obama is a definite urbanite, he has not necessarily worked hard to earn the urban vote.  Obama attempted to appeal to the nation as a whole over his first four years — a clear, successful strategy to maintain second-term electability. Perhaps as the Republicans fear, he will use his second term to enact his "real" agenda, one that could conceivably include more emphasis on distinctly urban policy as called for post-election by the National Urban League.

If the Democratic Party harkens back to 2000 and 2004 in 2016 by nominating Vice President Joe Biden, or another similar white, liberal man, the party could sacrifice its urban gains under Obama — thus making real urban policy and urban economic performance more important to secure future urban votes. But in the face of Republican inaction to claim the cities as their own, the Democrats will continue to paint these towns blue — even if the shade is a bit lighter next time around. 


Joshua D. Ambrosius, Ph.D., is an assistant professor in the political science department and the Master of Public Administration program at the University of Dayton. He is also a Senior Fellow with the Center for Sustainable Urban Neighborhoods at the University of Louisville. He holds advanced degrees in public policy and administration, with emphases on urban policy, from the Johns Hopkins University and the University of Louisville. His publications have appeared in several top urban policy journals, including the Journal of Urban Affairs and Housing Policy Debate.

Graduate assistant Michael Spafford assisted with this research.

For more information, contact Meagan Pant, assistant director of news and communications, at 937-229-3256 or mpant1@udayton.edu.