Thursday, March 20, 2008

Obama and the Wealthy

After the Wyoming caucuses, John noticed Teton county's overwhelming support for Senator Obama and argued that this could be seen as indicative of Obama's support among the higher income stratas across the country. With each successive primary and caucus, exit polling has shown that Obama does best among the higher income Democratic voters while Clinton has done best among working class Dems. In struggling states like Ohio and the soon to vote Pennsylvania, this correlation between income and vote choice has worked to Clinton's advantage and offers her best (though slim) chance of overtaking Obama.

I thought I'd take a broader look at these trends by focusing on those high income counties across the country that have voted to this point. A recent study by the Bureau of Economic Analysis provides a listing of the wealthiest counties in America, measured by per capita personal income. I've gone through the top 100 counties and looked at how they voted in the primaries and caucuses.

Of these 100 wealthiest counties, there are 18 that are either in states that have yet to vote or are in the disputed Florida and Michigan. Thus, I haven't included them in my analysis--leaving a total of 82 counties. So, how did Obama and Clinton do??

Of these 82 wealthiest counties, Obama won 60 (73% of the total) while Clinton won the remaining 22. Not only did Obama win an overwhelming number of these counties, but his vote totals in these counties outperformed Clinton in the counties she won. In the Obama wins, he averaged 62% of the vote whereas Clinton averaged 53% in her counties.

I next wondered whether to some degrees these numbers were affected by broader statewide trends. For example, if a candidate won statewide, might this explain part of their success in a particular county? To answer this I looked at how many of these counties were won by the candidate who lost statewide. In such cases, the economic variable would be stronger. Here, Obama's performance is even more impressive. Clinton did not win a single wealthy county in a state she lost. Obama, on the other hand, won 18 wealthy counties in states he lost. Specifically, he won 5 counties in New Jersey, 4 in California, 3 in Massachusetts, 2 in Tennessee, 2 in Texas, and 1 each in Nevada and New Mexico. Finally, we can look at the importance of a "home state" effect. Of Clinton's 22 county wins, 6 were in her home state of New York while only 2 of Obama's were in Illinois.

Thus, when we look at those counties at the very top of the income distribution, we get further confirmation of Obama's attractiveness to high income voters in the Democratic races. One thing that we might raise as a question to consider as we move toward the general election (should Obama be the nominee) is how many of these voters--who have historically voted Republican--crossed over and voted in contests that weren't closed to registered Democrats. Will they stick with Obama in the fall or return to the Republican fold?


Anonymous said...
This comment has been removed by a blog administrator.
Geoff Robinson said...

That's the $64 question I think. McCain is likley to have an appeal to these voters.
Very interesting site. Key's State and Nation is one of the best books of political science ever written and it is good to see it being remembered.

Tracey said...

Am I correct in remembering that Clinton would do better against McCain than Obama would in popular election projections?

Given McCain's historical political positions and works(even honored in 2004 by the Concord Coalition with the Economic Patriot Award for his fiscally responsible policies)he could swing plenty of votes over if he plays his cards right.