Young people are marching after #Parkland, but are they registering at a higher clip in Florida?

Let’s take a deeper dive into new registered voters in Broward County, Florida. According to a story in the New York Times by & , young people are registering. It’s hard to tell from their graphic (below) whether the total new registered voters in Broward in the month of March is 3,416 (628 + 626 + 1575 + 587) or if the figures are for just 18-25 yr olds newly registering to vote in the county.

NYT Broward Young Registration

According to my count, which draws on the May 2018 statewide voter file, some 4,383 voters registered anew in Broward County in March 2018. Of those, 2,026 (46%) are 18-25 years old.

But what do these new registration figures mean?

Here are the comparable March 2014 new registrations in Broward County (that I’ve drawn from a June 2014 statewide voter file).  In March 2014 in Broward, there were 9,853 new registrations, more than twice as many my figures indicate were recorded in March 2018.  Of those in March 2014, 2,972 (30%) were 18-25 yrs old (as of June 6, 2014).

So, compared to four years earlier, a higher percentage of new registrants in Broward in March 2018 are aged 18-25, but that’s because half as many people registered in March 2018 compared to March 2014 in the county.  In March 2014, nearly 1,000 more 18-25 year olds than in March 2018.

Of course, there’s always some slippage when dealing with voter files. My snapshot for 2014 is from early June; it’s certainly possible that some individuals who registered in March 2104 and who were living in one of the state’s other 66 counties moved to Broward in April or May, upping Broward’s June count. It also appears that the March 2014 count of new registrations in the statewide voter file includes thousands of “pre-registrants,” that is, individuals who were 16 or 17 years old who registered to vote that March. Those individuals don’t appear in the March 2018 statewide voter file.

What about the partisan breakdown of younger registrants across the four year time span?

Surprising no one, in Broward County this past March 2018, of the more than 1,800 newly registered voters 18-21 years old, 48% registered as Democrats, 37% as NPAs, and 12% as Republicans.

Statewide in Florida, in March 2018 some 10k 18-21 year olds newly registered to vote. Roughly 38% are Ds, 19% are Rs, and 40% are NPAs.

For a comparison of new party registrations, statewide in March 2018 there were slightly less than 11k new voters aged 45-64. Of these older, new registrants, 26% Ds, 33% Rs, and 38% NPAs.

None of this is news. Younger voters register to vote. They are more likely to register as Democrats. Older people newly register, too. In substantial numbers. They are more likely to register as Republicans.

So, a couple more comparisons of new registrations in Florida in 2018 versus 2014.

Between January 1 and April 30, 2018, there were a total of roughly 155k new registrations statewide. Of those new registrants, 50.2k are 18-29 year-olds and 44.6k are 45-64 year-olds. Of the 18-29 group, 32% are Ds, 20% are Rs, and 45% are NPAs. Of the 45-64 group, 26% are Ds, 34% are Rs, and 37% NPAs.

Compare these figures with first four months of 2014. Between January 1 and April 30, 2014, there were 166.6k new registrations in the state. Of those, 50.2k were 18-29 years old.  Another 22.4k were 45-64 years old. Of the 18-29 group, 28% were Ds, 19% were Rs, and 48% NPAs. Of the 45-64 group, 29% were Ds, 33% were Rs, and 33% were NPA.

One last point about the supposed causal relationship between the uptick in voter registration among younger individuals and Florida and the #Parkland tragedy.  The Times reporters write:

In Florida, voters under 26 jumped from less than 20 percent of new registrants in January and February to nearly 30 percent by March, the month of the gun control rallies. That ticked down to about 25 percent in April, as the demonstrations subsided, but registration of young voters remained above the pace set before 17 students and faculty were killed at Marjory Stoneman Douglas High School in Parkland.

It’s amazing that the exact same pattern existed four years earlier, over the first four months of 2014, when Florida did not have a school massacre and there weren’t mass demonstrations against gun violence and the #NRA.

The percent of total new registrants aged 18-24 statewide in Florida in January and February 2014 was 20.6% and 20.3%, respectively. In March, 2014, the percentage of all voters registering in the state aged 18-24 jumped to 34.5% — even higher than the percentage in 2018 following #Parkland and student mobilization.  Then in April, 2014, the 18-24 group’s percentage of new registrants dropped down to 24.6% of all new registrants, which was “above the pace set before…….”  You get the picture.

This is really a story of omitted variable bias. There are lots of other reasons why young people register in March — starting with ramped-up voter registration drives by the state’s 67 Supervisors of Elections in High Schools and at community events. Many of the students and other young people in Florida who were understandably and genuinely mobilized to register after the Parkland shootings would have registered anyway at one of these events, just as many of their older brothers and sisters did four years earlier.

In short, it’s still way too early to draw conclusions about the supposed registration effect of #Parkland.

Of course, turnout of younger voters in a midterm election is another issue altogether, but that story will have to wait for another day.

Think those last-minute registrants who vote in the upcoming election will turn out in future elections? Think again.

My latest research, coauthored with UF PhD candidate, Enrijeta Shino.

Timing the Habit

Online version available here.

 

Will Influx of Puerto Ricans flocking to Florida hurt Trump or Republicans?

I merged information about the birthplace of the 13m registered voters in Florida with precinct results in the 2016 General Election.  The following figure reveals the two party vote for Trump and Clinton in those Florida precincts that had at least 100 Puerto Rican-born registered voters who voted in the 2016 General Election. Each of the precincts’ voters were at least 50% Hispanic, and at least 1/5 of those Hispanics who voted were Puerto Rican-born.  In other words, these are about as Puerto Rican of precincts in Florida as it gets.

For example, in these majority-Hispanic voter precincts, in which, say (on the X-Axis), 30% of Hispanics who voted were Puerto Rican-born (and that have at least 100 PR-born voters), Trump won only 30% of the vote, on average. He did very poorly in all of the predominantly Puerto Rican precincts.

PR 2016 Vote

Does this mean that Republicans are in trouble in FL if the exodus from PR to the mainland happens? Puerto Ricans, certainly compared with other Hispanic groups in Florida, have weak voter turnout.  Of the more than 180k PR-born voters in my database, only 112k of them voted in 2016 (62%). That’s much lower turnout rate than, say, Cuban-born voters in FL. In 2016, 242k of the 325k Cuban-born naturalized citizens in my database turned out; that’s roughly 75% turnout.

A Manic Morning ElectionSmith Exclusive: Party, Race/Ethnicity, and Age Breakdown of >90k Floridians Who Successfully Registered to Vote During Extended Week following FDP Lawsuit

With an extra hour from daylight savings, I was able to crunch some new numbers…

According to the Florida Department of State, Division of Elections transparent by rather cryptic website, over 107k voters registered between October 11 and October 28.  According to the DSDE’s website, “Pursuant to a court order, Florida’s voter registration deadline for the 2016 General Election was extended from October 11 to 5:00 PM on October 18. The Department of State’s Division of Elections is committed to ensuring that all eligible Floridians who submitted a complete voter registration application by the extended voter registration deadline can vote.”

As the expert who crunched the numbers for the plaintiffs in that federal lawsuit (Florida Democratic Party v. Richard Scott) I’m obviously interested in the numbers of new registrants.

For some reason, DSDE’s website of daily activity reports provides only the “total number of voter registration applications verified and made active, and voter registration applications in the verification process, on or after October 11.”

Here’s the final running count provided by DSDE.

Date Applications Verified & Active (Registered Voters)
10/11/16 – 10/28/16 (4:12 PM) 107,085

I’m not sure why October 11 registrations are provided as the start date of the running total, as that was the regular final day (bookclosing) to register to vote in Florida prior to the court’s extended week of VR.  The DSDE’s processed voter registration dates also run through October 28; these voters who were processed after the October 18 extension should be eligible to cast regular ballots in the November 2016 General Election, as the DSDE states that, “The Department will be providing updated daily numbers until the state provides statistical bookclosing (voter registration) reports for the upcoming General Election.” [emphasis added]

Below are the party and demographic breakdowns of the 90,695 Floridians who were successfully processed by the state’s 67 county SOEs and the DSDE and recorded in the statewide voter file as being registered to vote on October 12 (the first extended day per the court’s order) through October 18, 2016.

PARTY

Democrats: 28,357 (31%)

Republicans: 23,410 (26%)

NPAs: 36,414 (40%)

3rd Parties: 2,514 (3%)

 

RACE/ETHNICITY

Black: 10,782 (12%)

Hispanic: 19,239 (21%)

White: 45,050 (50%)

All Others: 15,624 (17%)

 

AGE

18-29: 32,448 (36%)

30-44: 22,939 (25%)

45-59: 18,042 (20%)

60-105: 13,810 (4%)

Other (Pre-Registrant; Over-105; Redacted): 3,456

 

These October 2016 figures comport to my expert report in the FDP v. Scott lawsuit, where I find that in 2012, the more than 116,000 individuals who successfully registered to vote in the final six days before the October 9, 2012 bookclosing were disproportionately more likely to be racial/ethnic minorities and younger (I was not asked to look at the party of the registrants by the plaintiffs).

Gaining a hour of sleep (work) feels pretty good, but not nearly as good as I feel having done my part to extend the voter registration deadline in Florida.  Pretty amazing what a little tweet can do…

scott_vr_matthew

The Last Word on Voter Registration Numbers in Florida

On December 31, 2012, a month and change after President Obama’s re-election when he carried Florida a second time, of the 12.6m registered voters in Florida (including Active, Inactive, and Pre-registered 16 & 17 year-olds ), Democrats comprised 40%, No Party Affiliates (NPAs) made up 21%, and Republicans accounted for 35% of the electorate.

Flash-forward to August 31, 2016. Democrats now make up only 38%, NPAs are up to 24%, and Republicans continue to comprise roughly 35% of the overall statewide electorate, which now weighs in at a hefty 13.7m registered voters.

And there was much rejoicing among Republicans…as the gap between registered Democrats and Republicans had narrowed, from roughly 595k to just 419k of all Active, Inactive, and Pre-registered voters on the statewide voter rolls, as of August 31, 2016.

But as I’ve tirelessly Tweeted and conversed with the media, this narrowing (by 176k voters) does not mean that Republicans have out-registered Democrats over the past 3 1/2 years.

As I’ve repeatedly shown, Democrats have actually out-registered Republicans over this period of time. 

Below I report the 2.32m newly registered voters, by party, of those who registered between January 1, 2013 and August 31, 2016 and who are still in the voter file. These are not voters who registered prior to January 1, 2013 and who subsequently changed their party registrations. No, these are NEWLY registered voters over the 3 1/2 years as of August 31, 2016.

fl-2013-2016aug-new-registrations-by-party

As the table shows, of the 2.32m voters who registered during the 44 month period and who are in the statewide Florida voter file (as of August 31, 2016), 31% are registered as Democrats, 38% are NPAs, and 28% are Republicans.

In real numbers, over 80k more Democrats than Republicans are in the voter file who registered during the 3 1/2 year period. More precisely, for those of you who really like facts, as of August 31, 2016, there are 80,329 more Democrats who registered since January 1, 2013 than Republicans who registered over the same period of time and who are in the statewide voter file as of August 31, 2016.

So, with Democrats out-pacing Republicans on new registrations over the past 44 months, how is it possible that the the registration gap between the Democratic and Republican parties has closed by some 176k registered voters, with the relative percentage of Republicans growing and the percentage of Democrats decreasing?

The main reason is that the attrition rate over the past 44 months of previously registered Democrats is larger than that of Republicans who were registered prior to January 1, 2013. Some 1.24m voters have ‘exited’ the voter file over the 3 1/2 years.  Some 514.9k Democrats who were registered prior to 2012 are no longer on the rolls; 435.8k Republicans who were on the books back on December 31, 2012, are no longer registered to vote.

That’s a Democratic disadvantage of 79k voters, and it largely accounts for the Republicans closing the overall voter registration gap in the Sunshine State.

The balance –some 97k voters — accounting for Republicans closing the statewide voter registration lead held by Democrats is due to party-switching.

Of the 11.3m voters in Florida who were registered on January 1, 2013 and who remained on the voter rolls on August 31, 2016, some 216k former Democrats are no longer (as of August 31, 2016) registered with the Democratic Party. There’s also been party-switching among formerly registered Republicans, but it has not been as drastic: some 120k registered voters who were registered as Republicans on December 31, 2012 are no longer registered Republicans (as of August 31, 2016).

I’ll have fuller examination of  party-switching in a subsequent post, as I’m currently working on an academic paper on the topic that delves down into much greater detail.

So there you have it. The great mystery of the narrowing voter registration gap between Democrats and Republicans in Florida has been solved.

No conspiracy. No skewing of the data. Just the facts.

Yes, I know facts are tough to swallow for some, especially in these heated times.

But if you just chew on them long enough, you really can digest them. Really. Just try.

I still have hope that the United States can become a Mastication Nation.