The Pennsylvania Department of State has released a voter registration tally for the May 2017 primary, which was held earlier this week. The percentages are: Democratic 47.94%; Republican 38.29%; Libertarian .55%; Green .15%; other and independent 13.08%.
In November 2016, the percentages were: Democratic 48.34%; Republican 37.85%; Libertarian .56%; Green .16%; other and independent 13.09%. Thanks to Michael for the link.
Interesting that both minor parties and independents took slight hits, DEMs down 0.4 and REPs up 0.44. I wonder why the swing toward the GOP.
This tally shows a lower number of registered voters than there were in November 2016. That means the government purged the rolls somewhat. Democrats tend to be more mobile so they tend to lose registrants at a higher rate than Republicans, after a purge.
The Republican registrations are fresher. From November 2015 to April 2016, R +146, D +89, O -44, reflecting registration to vote in the closed primaries.
From April 2016 to November 2016: R +175K, D +155K, O +119K, buoyed by enthusiasm for Trump and the presidential election. The independent share of the increase was almost twice their share of the total registration. If you are pretty apolitical, you might register if someone shoves a card in your face, but you won’t choose a party, and you might not vote. This surge in independent/indifferent voters happens every election year, particularly presidential election years.
Comparing 2008 to 2012, the Democratic vote total dropped by -286K, while the Republican vote increased by only +25K. Obama enthusiasts were dropping out, they weren’t switching to Romney.
From 2012 to 2016, the Democratic vote total dropped by only -64K, while the Republican vote increased by 290K. Again it was not people switching, so much as Trump support from voters who did not regularly vote in the past.
It is hard to remove voters from the rolls. If they die in Pennsylvania, elections officials will likely know about. But if you move, there is a very strong chance that they won’t be aware of. Pennsylvania uses the list of change of address filed with the USPS, but most person don’t bother. And even then you have to contact the voter and confirm that they are living elsewhere. If they fail to make contact, they have to wait two federal elections before they can be purged. So you are seeing people who registered, perhaps for the first time, to vote for Obama. If election officials lost track of them, and they hadn’t voted for two elections, they are now being purged.
If someone registered to vote in 2016, they likely voted in Pennsylvania. If someone didn’t vote in 2016, but were registered in the past, they likely moved and/or died.
Democratic registration peaked at 51.22% in November 2009, when Republican registration had a trough of 36.93%. Since then, Democratic registration has dropped to 47.94%, while Republican registration has increased to 38.29%. The increase in Democratic registration in 2008 was massive (+597K), that of the Republicans in 2016 merely huge (+321K).
Pennsylvania does use changes in address from driver’s license to track inter-county movers. If someone has changed their license to a different county, they may be contacted to switch their registration. This is a fairly new program.
Richard:
Was is it that the Constitution Party registration is not listed separately from the Other voters?
Pennsylvania election officials do keep track of the number of Constitution Party registrants, but they don’t reveal it publicly because it doesn’t have the same status as the Green and Libertarian Parties. Those latter two parties each had a statewide nominee who met the vote test. The Constitution Party didn’t meet the vote test. The vote test doesn’t actually put a party on the ballot but it does give it certain privileges.
Pennsylvania does maintain party registration. If you look at historical registration numbers, the Green and Libertarian parties disappear, and the “Other parties” gets a bump, and then a few years later they get listed again, with the numbers roughly the same as before.