Monday, March 03, 2008

Maryland Primary--Assessing PG County


Before we get into the results that will be coming from Texas, Ohio, Vermont, and Rhode Island, I want to spend a bit more time on some results from an earlier primary--Maryland. One county that's political history is quite interesting and that has been undergoing noticeable change in recent years is Prince George's County. Encompassing the suburban area directly east of the Washington, DC line, PG County is notable in that it is the wealthiest county in the country with a majority African-American population. Currently, in a population of close to 900,000, PG County is roughly 63% African American, 27% Caucasian, and 7% Hispanic. A recent Brookings Institution study gives a nice overview of the county. It is summarized in this Washington Post story.


As you would expect from these demographics, PG County is solid Democratic territory now, especially in presidential contests. In 2004, Kerry won 82% of the vote; in 2000, Gore got 80%. However, PG County wasn't always so Democratic in its leanings. Nixon got nearly 60% in 1972, for example and Carter received just a slight majority in 1980. So who gave these Republicans such numbers? The culprits would seem to be the county's white working class population. Cities like Bowie, Laurel, Hyattsville, and Suitland had sizable numbers of such voters going back generations. In fact, it was in PG County that George Wallace did particularly well--he received 19% of the vote in 1968 (compared to 13% nationwide). Moreover, Wallace won the 1972 Democratic primary in Maryland, despite being shot while campaigning in Laurel.


It is this background that got me wondering about what the recent primary results would show. More specifically, are there any remnants of this white working class community left? If so, how would they vote? Are their numbers sizable or concentrated? The natural heir to Wallace's economic populism would seem to be Mike Huckabee. Southern, culturally conservative, and anti-establishmentarian in temperament, Huckabee is in many ways a modern version (absent the racial appeals) of Wallace. So, what do the numbers show???


First off, we need to reckon with the vast disparity between the Democratic and Republican contests in terms of turnout. While some of this disparity was certainly due to the relatively settled nature of the Republican contest, the fact remains that county wide, 160,000 Democratic votes were cast compared to 12,000 in the Republican primary. Thus, this is an overwhelmingly Democratic county (as the presidential results cited above attest).


In the Democratic race, Obama received 79% compared to Clinton's 20%. Clearly, the demography of the county (African-American) propelled Obama's numbers as he outperformed his statewide total of 61%. On the Republican side, Huckabee out performed his statewide total of 29% by receiving 33% of the vote in the county. So does this mean that Huckabee has inherited Wallace's legacy? Not so fast. A few days ago, John pointed to a Washington Post article about Huckabee's success in Washington, DC's overwhelmingly African-American wards 7 and 8. Did the same thing happen in PG County? I decided to look at the Republican race more closely and there seems to be some reason to suspect this.


One thing that should be noted right off the bat is not only the small number of Republican votes county wide, but how small these numbers get when you go precinct by precinct. Across the 216 precincts in the county, there was no precinct in which more Republican votes were cast than Democratic. The average precinct cast 687 more Democratic votes than Republican. There were an average of only 56 Republican votes per precinct. What this does is not only amplify the party leanings of the county but it makes it difficult to make definitive conclusions about how strong a pro-Huckabee trend might be. Only a small handful of votes (oftentimes less than 5) determined whether Huckabee won a precinct versus McCain. Nonetheless, of the 216 precincts, Huckabee won 41 of them.


So what can we say about these precincts? To determine whether these were in mostly white or mostly African American areas, I decided to look at the Democratic vote versus the Republican vote in each precinct. Specifically, I compared the number of total Democratic votes with the number of total Republican votes per precinct and used this as a proxy for how "black" or how "white" the precinct is. I'm assuming that there are relatively few highly integrated precincts. This is obviously a crude measure but absent more detailed demographic portraits of each precinct, this will have to do for now. So, those precincts with a greater difference of D to R votes was deemed more "black" and those with a lower difference of D to R votes was deemed more "white."


I next divided these 216 precincts into 5 quintiles based on this continuum. What do we see about Huckabee's performance? In the "blackest" quintile, Huckabee got 9 of his 41 precinct wins; in the second quintile he got 12 of his wins; the middle quintile produced 8 of his wins; the fourth quintile produced 7 of his wins; and finally, the last and thus "whitest" quintile produced 5 of his wins. Thus, we generally see that the more Democratic the precinct was, the more likely it was to go for Huckabee. Why might this be? One thing we must realize is that whatever group of people we look at, there will be some political diversity. In a lot of these precincts, Huckabee is getting a dozen or so votes. You can find a dozen people to support just about anything. However, there might be some reason to expect some African American support for Huckabee. As the Washington Post story shows, Huckabee's evangelical, Baptist faith and social conservatism would be expected to play well in the African American community. The key is to find enough African American voters, in a vast sea of Democrats, willing to pull the Republican lever. For all practical purposes, it wouldn't really make sense for a Republican presidential candidate to campaign in PG County. However, digging into the numbers unearths some interesting tidbits.

No comments: