Where the Obama Coalition Failed Clinton

It’s a well-established but little know fact that Obama would have won in 2012 even if the electorate resembled 2004 in demographic and partisan categories.  Indeed, Obama’s 2012 reelection charted a different path than his 2008 election bid.

Obama did what few Democratic Presidents could do in 2008.  He carried states as diverse as California and as white as Ohio.  His unlikely victories in Indiana and North Carolina were fueled on the back of robust black turnout.  In his 2012 reelection bid,  he charted a decidedly populist tone on inequality and corporate greed to maintain the allegiance of Northern whites who had drifted away from the party in 2010.

In hindsight, it is easy to see why Clinton was vulnerable to falling short against Mr. Trump.  Long before James Comey’s letter to Congress, WikiLeaks leaked emails about the DNC and even before Clinton’s deplorable comments polls showed Mr. Trump consistently ahead of Mr. Romney among blue-collar whites.  Clinton, however, was not matching Obama’s numbers among black voters.

In truth, the core of the Obama coalition was not the “Ascendant Majority” or the “Rising Electorate” but rather an unlikely alliance between urban blacks and Northern whites.  They voted for Obama for different reasons but in the end their mutual partisan preferences gave the President over 100 Electoral College votes and the Presidency.  In 2016, that alliance fell apart and the “real” Obama coalition crumbed as did the “Blue Firewall.”

The Faltering Coalition

Towns along the Mississippi River or industrial counties along Lake Erie are not the kinds of places one envisions supporting Obama.  These blue-collar, majority-white towns seem custom made for GOP victories.  While they gave the GOP solid margins in 2010 they sided with President Obama in 2012.

This runs counter to the conventional narrative that Obama won because of Hispanics and large numbers in urban locales.  The President would have been reelected even if Romney had garnered George Bush’s margins among Hispanics.

Consider these facts.  Obama did better among non-college educated whites than Al Gore and John Kerry did, 34 percent, according to Voter Files.  That 34 percent gave Obama more votes than his lopsided margins among black and Hispanic voters.  It certainly gave him more electoral votes.  In a wide swath from Oregon to the coasts of Maine Obama ran better than any Democrat since Bill Clinton.

Four years later, Mr . Trump painted these areas red.  From Luzerne County, PA to Kenosha, WI to Trumbull, OH, Trump garnered massive margins.  No bastion of blue-collar, Democratic dominance was immune.  For example, Erie County, flipped from red to blue and Clinton’s margins in Scranton, PA was thousands of votes shy of Obama.

Unlike 2012, the partisan shift among blue-collar whites was not limited by region.  While Romney did better than McCain and Bush among these voters it was largely limited to the South.  He did worse or no better among Northern whites.  But this go-round, exit polls from CBS show Trump actually did better among down-scale whites (defined as making $30K or below) than among influential whites.

As recently as a few months before November it was assumed Clinton would actually do better than Obama among Northern whites.  The assumption was these voters would remember her from her 2008 primary bid and support her.  Most of these voters were older and distinctly non-Southern, attributes sure to help the moderate Clinton.

Except not only did Clinton do worse among older, Northern whites but also younger, Northern whites.  Whereas Obama made gains among both relative to 2004, Clinton gave up those gains and more to Trump.  For example, Clinton lost whites without a college degree by over 20 points and barely won over whites 18-29.

Turnout did not Decide 2016

Despite turnout lagging below 2012 it did not decide the election.  The USC Daily Tracking poll, composed of a rolling sample of 2012 voters, found a significant number of Obama voters backing Trump.  Unlike standard surveys, USC’s numbers were not impacted by turnout.

As more Voter File data becomes available we should be able to get a fuller picture of who voted in 2016.  But, what data is available suggests turnout did not drive margins among Northern whites.   Few, if any more, conservative whites showed up relative to 2012.

While not part of the North, data from North Carolina and Florida shows turnout among white Democrats and white Republicans increased at about the same rate.  The Florida Panhandle might have become even redder under Trump but it was not because white Republicans showed up in massive numbers.  Rather, Trump increased GOP margins among Independent whites.

Exit polls, despite their issues, also tell us about Trump’s gains.  If you parse through the data, you’ll find  ,”Trump won 19 percent of white voters without a degree who approved of Mr. Obama’s performance, including 8 percent of those who “strongly” approved of Mr. Obama’s performance and 10 percent of white working-class voters who wanted to continue Mr. Obama’s policies.”  In an era of polarized politics this is unprecedented.

Trump/Obama Voters

I have written about Trump/Obama voters and where they mattered. For Americans with fixed ideological views it seems shocking that an Obama voter could back Trump.  Of course, the same could be said for a Romney/Clinton supporter (there were a lot of both this cycle).

But, both Obama and Trump ran on similar, if not exactly alike, campaign themes.  Obama in 2012 made inequality and corporate greed a centerpiece of his campaign pitch.  Indeed, nothing better illustrated this than Obama crowing about the success of the auto bailout while Romney was best known for the headline in the New York Times, “Let Detroit Go Bankrupt.”

Obama went out of his way to paint Romney as a heartless plutocrat.  While the ad accusing Romney of killing a factory was absurd and wrong it underscored the point that a President Romney would leave the struggling middle class prey to outsourcing and globalization

Romney, and McCain for that matter, did find success in one bastion of blue-collar white America.  Due to the Obama administration’s “War on Coal” Romney did better in West Virginia and Southern Virginia than any Republican nominee since Reagan.  These gains would be a harbinger of 2014 and 16.

Picking up on this Trump did not just promise to protect blue-collar America but restore it to its former glory.  He railed against NAFTA and shipping jobs overseas.  He promised to stand up to companies like Carrier shipping jobs overseas.  More importantly, he warned about the excesses of Wall-Street and criticized America’s urban, PC culture.

Whether these themes gave Trump his winning margins among Northern whites is debatable.  But, what is not debatable is that Trump painted Clinton as a tool of the global elite and somebody who would sell out the middle class.

Prior Republicans had tried to appeal to Northern whites via gay marriage, abortion and tax cuts.  They refused to yield on free trade, immigration or the social safety net.  Instead of talking about reforming government many Republicans simply preached, “cut, cut, cut.”

Trump found success running on the dangers of the regulatory state, radical Islam and unrestricted free trade.  He swung right on the issue of guns while talking about reasonable restrictions and laid claim to the law-and-order mantle.

None of this is to say that turnout did not impact the election.  Rather, it simply did not swing white, blue-collar voters further to the right.  For turnout among a key Democratic constituency did matter and probably gave Trump the election.

Where Turnout Mattered

Situated along the Great Lakes, Cuyahoga County anchors Ohio Democratic politics.  It’s the base of Democratic votes and gives the party hundreds of thousands of votes to play with.  Yet, this go-round the county failed to perform.  The 30 percent black county spit out 37,000 fewer votes overall than 2012 and Clinton garnered almost 50,000 fewer votes in the county than Obama.

While not the North, indications black turnout dropped almost uniformly can be seen in the South.  Voter registrant forms from the South where voters indicate their race show black turnout lagged.  In North Carolina, black turnout fell from 23 percent in 2012 to a likely less than 21 percent in 2016.  In Georgia, black turnout fell almost 2 percent and in Louisiana it fell a point and a half.  Florida seems set to see similar numbers.  Black turnout is still set to exceed 2004 but it will fall short of Obama’s 2012 mark.

Clinton needed these voters.  Her losses among blue-collar whites meant she needed a new source of votes and she did not get it.  Indeed, for while Clinton garnered 17,000 more votes than Obama in Philadelphia that gain came almost exclusively among majority-white precincts.  Turnout in black precincts dropped significantly in the city.

Notably, black turnout seems likely to drop uniformly across the country.  Whether a state had new voter ID laws or not does not seem likely to impact the raw numbers.  This is bad news for advocacy groups seeking to overturn such laws.

Pinning Clinton’s loss on black voters is decidedly a one-sided analysis. Obama would have won reelection even with current black turnout levels.  He would still have cruised in Pennsylvania, Wisconsin, Ohio and Michigan.

The Clinton Coalition (ie Rising Electorate) Fails to Deliver

The data driven Clinton campaign knew they would likely see black turnout drop.  But, they were confident the “Rising American Electorate” composed of Hispanics and college educated women would see them through.

To a degree such an assumption was proven right.  Clinton made significant gains among affluent, college educated voters nationally.  She won traditionally Republican Fort. Worth, Texas and Orange County, California.

But these two counties illustrate her campaign’s foolhardy gamble.  They came in non-competitive states.  Clinton ran strong in the Georgia suburbs but she still lost the state by 5.  She won California by 30 points and lost Texas by 9 points.  Her gains in these places meant little in the grand scheme.

Data from Fivethirtyeight and other sources show education correlated closest with support for Trump and Clinton.  In this case, Trump benefited from the Midwest having fewer college educated whites while Clinton simply could not get the margins she needed in red Sunbelt states to flip Electoral votes.

Exit polls say Trump carried college-educated whites nationally, and he very well might have, but even if Clinton did it would not have mattered.  Clinton’s margins in urban enclaves were fueled by her strong margins among the affluent in Seattle, Boston, LA and NYC.  Her gains even went into strongly or formerly Republican suburbs.


In only one suburb did Trump improve on Romney’s performance.  While Trump did better in rural areas he did so by giving up the affluent to Clinton.  Inevitably, this is probably because these areas have benefited from diversity and trade.  Both things that Trump ran against and that have not helped the Midwest.

But, not all suburbs are created the same and the narrowness of Clinton’s gains among the affluent gave her significant margins only on the Coasts and  the noncompetitive Sun-Belt.  For example, in Lake County, Ohio, a suburb of Cleveland, swung to Trump by 15 points.  Clinton’s massive margins among those making $250K and above did not swing the county because it does not have many of these voters.  Likewise, neither does Kenosha, Wisconsin or Macomb, Michigan.

Even while Clinton made gains among the affluent she might have bled not just down-scale white but down-scale black and Hispanic support.  Exit polls show Trump won 8 percent of black voters.  In heavily Hispanic Southern TX, Clinton’s 66 percent support among the group in national exit polls manifested itself with smaller margins in counties along the Rio Grande.  She might have improved on Obama’s performance in the state by 10 percent but it was not because Hispanics came out in force for her and against Trump.  Indeed, analysis by Harry Enten at Fivethirtyeight shows that of the 24 counties with a Latino population of 75 percent or more Clinton ran behind of Obama in 18 of them.

Clinton won the popular vote because she reassembled the Obama coalition in non-competitive states by winning affluent Republicans and the well-educated.  But, when it came down to winning the voters she needed to take the Electoral College, she fell short.














3 thoughts on “Where the Obama Coalition Failed Clinton

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s