US Presidential Elections 2016

Ohio | Wisconsin | Pennsylvania

In this post, I attempt to analyze the results of the 2016 presidential elections by focusing on three keys states Obama won in 2008 and 2012: Ohio, Wisconsin and Pennsylvania. Together they represent 48 electoral college votes which would have been sufficient for Hillary Clinton to pass the 270 mark.

There may be numerous reasons that are hard to decouple behind Trump’s victory (or rather behind Clinton’s loss). By focusing on three states where Obama won the electoral college votes twice, I would like to narrate a story that is inline with Robert Reich’s opinion piece in the Newsweek. here

1990’s style center-leftism (Clinton-Blair left) is not able to attract the mid-income working class voters to the Democratic Party (nor to the Labor Party in the UK). While Obama was able to energize the voters by offering a promise of change, as he is ending his second term, the income disparity has not shrunk but increased. One can argue that Hillary Clinton was a candidate closer to Mitt Romney than Barack Obama. Although a message focused on economic policies to improve equality and quality of life of the lower and middle income classes may not be able to attract the religious conservative votes, it may be the message needed to bring low-income to middle class votes back the Democractic Party in states such Ohio, Wisconsin and Pennsylvania. It is an embrassment to the Democratic Party that a billionaire from New York was to convince millions of disgruntled voters that he would be a better representative and fighter for their welfare and interest than the Democratic Party.

In the below graphs, y-axis shows change in the vote share of the Democractic and the x-axis represents individual counties. Counties are ordered by different variables in each graph per state: 1) Unemployment rate 2) Median-income 3) Percentage of white 4) Percentage of the non-college educated population. Colors of the bubbles indicate the value of the variable used to order the counties in the x-axis. Sizes of the bubbles are indicative only and try to indicate the number of people corresponding to the loss/gain in the vote share change as: (2016 DP Vote Share - 2012 DP Vote Share) * County Population / 1000.


Obama won in Ohio by 4.6% and 3% over McCain in 2008 and Romney in 2012 respectively. According to the currently available results, Clinton trailed Trump by 9% points. In almost every county, the DP vote share declined compared to the 2012 elections including Cuyahoga (Cleveland), and the decline seems to be correlated to the unemployment rate and college-education.

In Cuyahoga, a county that is ~30% non-white, Clinton got 383,974 votes vs Obama’s 447,273 in 2012. On the other hand, Trump got 190,660 votes vs Romney’s 179,894. This may indicate that besides losing the white votes, Clinton might have lost the support of the non-white voters expressed a lower turnout.

Clinton Obama Trump Romney
Cuyahoga 383,974 447,273 190,660 179,894

Ohio, Unemployment

Ohio, Median Income

Ohio, College Educated