One of the more interesting observations in Florida’s modern political history was the performance of Socialist-turned-Democratic gubernatorial candidate Brian Moore during the 2010 Democratic primary against Alex Sink. Yes, Moore was beaten badly statewide. However, Moore was able to perform better in the northern part of the state, particularly the Florida Panhandle. If we look at Moore’s election results, 21 of his top 25 performing counties were in North Florida and the Panhandle. In Holmes County, Moore was within three votes of Democratic front-runner Alex Sink.Why was this the case?
For anyone who has been observing Florida politics (especially those who have been following me for a while on my other website), you know that North Florida has been shifting sharply toward the Republicans since the 1980s. Some of these counties used to be nearly as strongly Democratic as they are now strongly Republican. Still, as of 2016, the tide has turned.
So, why did Moore perform so well in these counties? Nobody has really tried to tackle this question. Yes, some people have noticed the trend, but they do not explain the Socialist phenomenon.
Looking at this odd case, we first need to ask ourselves “why” would someone vote for Moore. Let’s face it, nobody really knew about him. And when he ran for the Socialist Party for president, he did not perform well. So we know that his success did not have to do with his overwhelming popularity in North Florida. But did it have to do with the unpopularity of Alex Sink? Yes, these voters obviously did vote against Sink, and might have done so as a protest vote. But again, we don’t know why that would be the case. It is highly unlikely that a registered Democrat would take the time to vote against someone, especially if they are probably unaware that they have a primary opponent in the first place.
That last point is important…why did people go out to vote? As I have already stated, I highly doubt they went to vote against Alex Sink. Instead, voters probably went to vote in other races instead, but then cast a protest vote against Sink. In many of these rural North Florida counties, local races are usually more salient than state or federal races. Therefore, voters are not going to vote for the top of the ticket, but instead are voting for the bottom of the ticket. Maybe we can call this a “up ballot drop off”? It is unusual, but it could happen.
Anyway, because these counties are shifting Republican, many of these voters who are registered as Democrats probably vote Republican more than Democratic nowadays. However, they haven’t changed their registration, as local candidate might still run as Democrats or run in non-partisan elections, which does not require a change in registration. Therefore, these top-of-the-ballot races just become a nuisance to voters.
If this is the case, and voters in these counties where Moore did well really were more concerned about races further down the ballot, there should be a large number of under votes when it comes to the Democratic primary for governor. Therefore, I decided to run a simple regression on this hypothesis to see if it holds true. I collected all of the data I needed for this experiment, which was county-level election data for each county. The hypothesis states that if the percentage of under votes for the Democratic governor’s primary in a county increases, so does the percentage of vote for Brian Moore. Below is the scatter plot chart of the results.
As we can see, there is a significant relationship between the under vote percentage and Brian Moore’s percentage of vote in a given county. This seems to indicate that voters were going into the booth to primarily vote in another race. In many cases, voters left the governor’s race blank, while others voted for Moore. In counties where people actually turned out to vote in the Democratic primary for governor, not only do we see Moore’s totals decrease, we also see less under votes.
Using aggregate-level data to understand the “Brian Moore Phenomenon” in 2010, we see that North Florida and Panhandle Democrats are more concerned about local races than they are about races at the top of the ticket. Also, it is probable that many of these ‘registered Democrats’ now identify as Republicans, but do not change their registration because local races usually do not require registration change (as many in Florida are now non-partisan). So, when these voters are voting for governor in the primary, they usually do so as a protest vote or try to pick the worst general election candidate, and are probably supporters of the Republican nominee.
The reason this happens is primarily due to the fact that Florida is a closed primary state. If Florida were to open up its primary process, most of these voters would probably select a Republican ballot, and Democratic numbers would decrease drastically. However, if there had been an open primary in 2010, Alex Sink would have probably won these North Florida counties by a much stronger margin.
(Note: St. Johns, Hardee, and Glades Counties were excluded because I could not determine the Democratic primary vote total for these counties.)