About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« Unconventional Thoughts on the Democratic Primary | HorseRaceBlog Home Page | Obama's Success in Central Pennsylvania »

A Review of the Pennsylvania Primary

Hillary Clinton won a strong victory yesterday in Pennsylvania. As expected, her voting coalition was quite similar to the one she had in Ohio and in previous non-southern contests. This is another sign that the basic demographic divide separating Obama and Clinton remains in place, some three and a half months after voting began.

The following chart details this by comparing Clinton's performance among the select demographic groups in Ohio and Pennsylvania.

OH and PA Demographics.gif

By and large, we can see why Pennsylvania and Ohio produced similar results in the aggregate. Clinton did roughly as well these groups in both states. Obama, for his part, improved here and there on her best Ohio groups. For instance, he trimmed her lead among white men. However, Clinton minimized this by doing slightly better with some of Obama's best groups - like, for instance, the college educated. Overall, it added up to a roughly similar result. Clinton won Ohio by 10.3%. She won Pennsylvania by 9.4%.

What we see, then, is what we have seen again and again in this contest. Clinton continues to do well with "downscale" whites. Obama does well with "upscale" whites and African Americans. What is intriguing about this result is not just that it is similar to Ohio - but also that it is similar after seven weeks and millions of dollars in campaign expenditures. Clearly, these voting groups are entrenched.

Let's amplify this analysis with a look at how both candidates did in each region of this diverse state.

Clinton Share of PA Vote By Region.gif

These results are as we might have expected. Clinton dominated the western portion of the state - winning Erie in the northwest and Pittsburgh in the southwest. What is more, if you look at the counties in the far southwest corner of the state, you'll see that Clinton's margins were quite lopsided. For instance, Fayette County gave Clinton 78.9% of the vote. This, I think, is an indication of how West Virginia will go. Expect Clinton to win the state overwhelmingly. A 40-point victory does not seem unimaginable to me.

She ran strong in the northeast - winning Scranton and its surroundings. Furthermore, while the chart does not indicate it, Clinton performed extremely well in the "Middle T" of Pennsylvania - the great rural expanse that stretches across the center of the state. Obama only broke her winning streak in Centre County (home to Penn State), neighboring Union County, and Dauphin County (home to Harrisburg).

In the east, Obama ran strong in Philadelphia County, but the two split the five suburban counties. Clinton also won the Lehigh Valley. Additionally, the two split fast-growing York and Lancaster counties in the southeast.

All in all, it was a strong, and generally predictable win for Hillary Clinton. But what does it mean?

We can say the following. If the superdelegates had grown concerned after Ohio about Obama's ability to win lower income whites in the general election - these results will not alleviate their worries. Pittsburgh, Scranton, and Erie all swung decisively for Clinton. If Ohio had them worried, so will these results.

As for whether it will serve Clinton's short-term goal of spinning herself as still having a chance to capture the nomination - that remains to be seen. That will be a matter up to the media, potential Clinton donors, and superdelegates. As of this writing, she was up by 9.4% in the Pennsylvania vote. Whether this is "enough" is not a question I can answer.

Nevertheless, I can say the following. As of this writing, Clinton has netted 216,000 votes from the Keystone State. Last month, when working out my "Predict the Race for Yourself" spreadsheet, I inserted values that seemed to be favorable numbers for Clinton that were also doable. These were not meant as predictions, but rather as an illustration of how Clinton might come back to win at least one valid popular vote count. The number of votes inserted for Pennsylvania was 211,000. I took that to be a reasonably optimistic outcome for her. She basically met that goal last night. This implies that she is "on track" to catch Obama in one of the more valid vote counts. It won't be easy, but she can still do it. If she does, this could be a compelling argument for the superdelegates.

I hope to return later in the week with a more comprehensive analysis of how Pennsylvania voted. Stay tuned!

-Jay Cost