Early Votes Don't Mean What You Think
Media coverage of the midterm election is abuzz with early vote chatter – across the United States and even here in Louisiana. Early voters in the bayou state have already cast 308,631 ballots for the November 6th election, the third highest total on record in the state and highest for a non-presidential election. Surely all those ballots must mean something.
No. Not really.
1. Early votes tell us nothing about election outcomes
One of the things that early vote watchers love to do is to recite the number of votes cast by different groups, especially by Democrats and Republicans. Inferring anything about election outcomes from these breakdowns is a fairly sketchy practice anywhere, but it is especially problematic in Louisiana.
The early vote numbers, of course, do not include actual votes. They include information about voters that is included in their registration. This includes party registration, but party registration is a terrible predictor of actual votes in Louisiana. Whether in early voting or Election Day voting, a sizable share of registered Democrats rarely ever vote for Democratic candidates (I will write more about this in a couple of weeks).
The table below shows the share of early votes cast by individuals registered in each major party since 2008 as well as the share of total votes cast for Democratic and Republican candidates.
The most obvious takeaway is that registered Democrats always cast more early votes than registered Republicans do, yet Democratic candidates do worse than Republican candidates do in the overall vote (with one exception). In other words, the share of registered Democrats among the early vote greatly overstates Democratic candidates’ vote share in the final tally.
Even comparing elections – to see if changes in party share of early vote predicts changes in total vote share across elections – does not tell us anything useful. Consider the 2008 and 2016 presidential contests in Louisiana with the highest and lowest shares of early votes cast by registered Democrats, respectively. Surely, this massive shift – with registered Democratic early vote share falling from 58 percent to 44 percent – had a radical impact on how the vote ultimately turned out, right?
Nope. Not at all.
Although the early vote party registration margin fell from a 29 percent Democratic advantage to a six percent advantage, the margin in the total vote share for the Democratic candidate versus the Republican candidate did not budge.
Sometimes the early vote margin remains stable, but the overall vote margin shifts. Registered Democrats’ advantage in early voting did not change from October 2011 to October 2015, but the margin in the actual vote share going to Democratic candidates fell by more than half.
Even more troubling for using registration statistics of early voters to predict election outcomes, the registered Democrat share of the early vote was stable from the primary to the runoff in 2015, but the Democratic candidate’s share of the vote flipped from a 15-point deficit to a 12-point advantage.
These examples illustrate the broader point that there just is no relationship between the distribution of early votes cast by individuals registered in each party and the distribution of votes ultimately cast for candidates by the end of the election.
This graph shows how weak the correlation is. The dashed line would be a perfect one to one relationship, but the relatively flat solid line shows there is little correlation between early vote and election outcomes.
2. Early votes don’t tell us much about turnout either
Political watchers often take a boost in the number of early votes as an indication of heightened voter interest leading to higher turnout in the election. There is some truth to this interpretation. Generally, elections with a large number of early votes also tend to have larger number of total votes. Besides just being an insipid take (‘more votes means more votes’), the heuristic often breaks down when making inferences about specific elections.
The number of early votes and the number of total votes cast in each statewide election in Louisiana since 2007 shows just how often the heuristic fails.
Let’s start with the 2014 U.S. Senate elections. The number of early votes fell about 12,000 from primary to runoff, and overall turnout in the elections dropped by about eight points. So far, so good for the early-vote-as-proxy-for-turnout heuristic.
Contrast that with what happens in 2015. The number of early votes shoots up by about 35,000 from the primary to the runoff. That is nearly three times larger than the magnitude of primary-to-runoff change in early vote totals for 2014. Surely, we had massive boost in turnout for the 2015 runoff then, right? Right? No. Turnout barely budged.
Until this year, the November 2015 election held the record in Louisiana for most early votes in a non-presidential contest at 270,144. Compare that to the next highest, the November 2014 election, which had about 25,000 fewer early votes. Which election had higher turnout? Not the one with more early votes….and it is not even close. Turnout in November 2014 was 51.5 percent. Turnout in November 2015 was 40.2 percent.
Or, look at presidential contests. The number of early votes has steadily climbed from 292,213 in 2008 to 356,603 in 2012 to 531,555 in 2016. Turnout? Holds steady at 67 or 68 percent despite all those early votes.
Or, look at the trend in gubernatorial elections. Relative to 2007, voters cast more early ballots in 2011 and 2015, but overall turnout was lower in the latter elections.
I am intentionally cherry-picking elections that cast doubt on the ‘more early votes mean higher turnout’ heuristic, but it is not hard to find them because the relationship between early votes and turnout is weak and imprecise.
First, a big part of the relationship simply reflects differences in the office at the top of the ticket. Both early voting and overall turnout tend to be higher in presidential elections than, say, special elections for Lieutenant Governor or State Treasurer. Surprise, surprise. Sure, in October 2017 we could have looked at the number of early votes to say, “Wow, looks like fewer people are going to vote in this Treasurer’s race than voted in last year’s presidential election,” but we already knew that from looking directly at turnout across many presidential and lower office elections. The early vote totals would not add any new information.
Second, the change in early voting from election to election do not just reflect differences in voter interest. They also represent a secular increase in early voting that has nothing to do with variation in engagement across elections. Over the past decade, there has been a greater propensity for voters to take advantage of early voting. Early voting as it currently operates in Louisiana has not been around for all that long. Louisiana voters are still becoming more familiar it and using it more.
Here is the share of total votes that were cats early in all statewide elections since 2007.
The trend is clear. More and more voters are choosing to cast their ballots early rather than on election Day, even in low turnout contests. Indeed, some of the highest shares of early votes were in last year’s primary and runoff for State Treasurer when overall turnout barely topped ten percent.
This trend is even clearer when broken out by office at the top of the ticket.
A naïve glance at early voting numbers can cause people to misinterpret this pattern of habitual voters moving their votes from Election Day to early voting as a boost in enthusiasm about a specific election.
Pundits this year love to talk about early votes – here in Louisiana too. Yet, anyone making a claim that the early vote numbers in Louisiana reveal heightened enthusiasm needs to be able to parcel out how much of the rise in early votes is due to that enthusiasm and how much is due to this secular increase in early voting. I have yet to see anyone do this with any degree of precision.