leftI can’t resist giving this claim out of North Carolina special prominence.  It has to be a contender for the phony baloney vote fraud claim of the year.

Among the changes being proposed in NC is an end to same-day registration during “one-stop” in-person absentee voting.

State Senate Pro Tem Phil Berger is quoted in the Greensboro News-Record supporting this change because of concerns over vote fraud (boldface added):

Senate President Pro Tem Phil Berger, R-Rockingham County, wants to look at shrinking early voting periods, which he said last week put stress on election workers and campaign coffers. He also wants to look at ending Sunday voting and same-day registration. Same-day registration raises concerns about voter fraud, and Sunday hasn’t traditionally been a voting day, he said.

The problem with this claim is that North Carolina already has in place the most fraud-proof system I am aware.

According to the State Board of Election website, and conforming to my own recollection, if a citizen shows up to a one-stop voting location and wants to register or change their registration and case at ballot at the same time.  I quote from the SBOE website (the last sentence is key):

Within two business days of an one-stop registration, the county board of elections will attempt to verify the registrant’s North Carolina drivers license or Social Security number, update the statewide registration database and search for possible duplicate registrations, and proceed to verify the registrant’s address. The registrant’s vote will be counted unless the county board determines that the person, for some statutory reason, is not qualified to vote.

It bears repeating: in North Carolina, an early vote cast after a same day registration is not counted unless the county board of elections can verify the validity of the registration.

I would hope the SBOE is calling up Sen. Berger and informing him about the procedures in place in his own state.

A decade of reform since HAVA has led to many positive changes in American elections, not the least of which is that there are solid empirical data available to provide guidance to legislators and election officials who want to improve performance and conduct more efficient elections.

This is one of the goals of Doug Chapin and the team at the Election Academy and this is the topic of the Election Data Dispatches coming out of the Pew Center on the States.  Charles Stewart of MIT and Barry Burden of Wisconsin have been leading an initiative on Measuring State Election Performance that puts many of these indicators to use.

I realize that election reforms are seldom if ever non-partisan.  I am not naive about the intentions of political actors.  But I would at least hope that if a legislator is proposing a major change to the election system in a state, he or she would try to at least pretend to have some facts and not just make things up.

Unfortunately, fairy land seems to have descended upon the Tarheel State this past week. I suppose it’s all because Duke, UNC, and NCSU are out of the NCAA tournament.  (OK, that last comment was grossly unfair.)

Let’s take a selected list and ask if this claim can be substantiated:

Claim: State Sen. Bill Cook (R-Guilford) believes that few North Carolians vote in the first week of early voting:

Cook says his campaign took him to early voting sites in his rural district where there was rarely a line.

“I don’t think we need (the extra days),” he said. “The first week, you get lots of folks. The second week, nothing. It’s almost a desert.”

Fact: 58% of early voters, and 30% of all voters cast an early-in person ballot in the first week in 2012.

Claim: Senate President Pro Tem Phil Berger (R-Rockingham County) says that “Sunday hasn’t traditionally been a voting day.”

Fact: No day has “traditionally been a voting day” other than Tuesday.  12 of the 32 states that provide early in-person voting mandate some voting on either Saturday or Sunda, while most other states allow local officials to make this choice.  It’s impossible without collecting detailed information at the county level to know just how many jurisdictions nationwide allowed Sunday voting, but it’s certainly not uncommon.  (The GAO issued a report on weekend voting just a few years ago on the topic of weekend voting.)

Claim: Greg Steele, chair of the NC Federation of College Republicans, says that cutting early voting in half will save money and 

would encourage people to research candidates before they vote.

“It would energize people into taking more time to be invested into what they’re voting for,” he said.

Fact: State Elections Director Gary Barlett takes on claim 1: “There is not going to be any savings at all by reducing the early voting period,” he said.

Claim two is hard to parse; apparently, people who vote in the first week of early voting don’t have enough time to research candidates, but voters in the second week do have enough time.  Steele expresses no concern about no-excuse absentee voters whose ballots are mailed out 45 days before the election.  What we do know from extensive research into the early voter is that these are voters who have already made up their minds and show no lower levels of political information or interest than Election Day voters (in fact, in most cases, quite the opposite).

Early voting is no an unalloyed good.  There are valid arguments that can be made against early voting.

The problem is that most of these reasons apply mainly to no-excuse absentee voting, not to early in-person voting, and the NC legislature seems unconcerned about by-mail ballots.

I tried to be nice the first time to True the Vote, but as Fairvote pointed out to me on Twitter, their revised report (linked here: http://www.truethevote.org/news/true-the-vote-report-proves-widespread-claims-of-voter-suppression-false) only makes things worse.

All the evidence on turnout (debunked first by me then more devastatingly by FairVote on the Election Law listerv) have been removed, yet the claims of turnout effects remain on pg. 4 of their report:

Pg 4: “Further, voter turnout rates should have demonstrated drops in participation.In both cases, our research proves otherwise.”

 Necessary correction: The research as been removed and the previous research has been shown to be inaccurate.  

This comment about lines in Florida may have been in the first draft of the report, but is also inaccurate.  This is from pg. 5 of the report:

Pg. 5: “the combination of roll purges, early voting changes and photo ID requirements did not manifest any clear negative impacts in 2012. While some voters complained of long lines, especially during early voting, the average wait time was 50 minutes for Floridians.”

 Necessary correction: The average wait time in Florida was the longest nationwide by far.  It was nearly four times the national average of 14 minutes, and 50% higher than the next state on the list (Maryland, at 28.8 minutes).  

Data are reported here http://www.nytimes.com/interactive/2013/02/05/us/politics/how-long-it-took-groups-to-vote.html.

If this is evidence of no “negative impact,” I’d hate to see what a negative impact looks like!

What makes this frustrating for an academic is that the impact of voter ID requirements on turnout and on wait time remains an open question, at least as far as I read the scholarship thus far.  The discussion isn’t helped along by badly misleading and poorly documented reports.

True the Vote provides a nice report with information on voter ID laws and the politics surrounding these laws.  They do a service by aggregating the number of voter suppression reports by states and county.

That’s the good news.  The bad news is that their conclusions about the relationship between voter ID and turnout are wrong, or more generously, unsupported by the evidence.

The report compares turnout in states in 2008 and 2012 and attributes EVERY CHANGE to the existence (or not) of a voter ID law.

Nothing else is considered, including those things that every observer knows are the primary drivers of voter turnout: the battleground status of the state, the competitiveness of other campaigns in the state, and overall campaign spending and election activity.

Compare two of the states they highlight: Colorado, Florida, Nevada, North Carolina, and Ohio.   In every state, total campaign spending was substantially higher.  In every state save one (Nevada) the ratio of campaign spending was much closer, in some cases dramatically so.

Campaign spending in 2012: http://www.cnn.com/ELECTION/2012/campaign-tracker/ and in 2008: http://www.cnn.com/ELECTION/2008/map/ad.spending/

I don’t know what the impact of voter ID laws were on turnout, but I am confident that the True the Vote report doesn’t shed much light on the question.

Hat tip to Rick Hasen for this story, Rep. Candice Miller, chair of Government Operations, who says the Federal government has no role in fixing problems with our voting system: http://www.dailytribune.com/article/20130226/NEWS01/130229602/miller-blasts-obama-s-plan-for-election-standards

Nice piece by Scott Adler and John Wilkerson, showing how the “reversion point” (sequestration) may be the preferred outcome for both congressional parties at this point.


Jeff Mapes, the dean of political reporters in Oregon (Jeff–“dean” in this context refers to your influence not your age!), has an interesting column on how ending Saturday mail delivery could impact Oregon’s vote by mail system.

Director of State Election Steve Trout rightly notes that Oregonians are not supposed to try to mail their ballots on Saturday anyway–the state urges voters to put them into the mail no later than Friday.

The problem arises not only among those voters who don’t realize this, but among others who don’t realize the last time for pickup at mailboxes and for those voters who live in rural counties and who don’t have easy access to drop boxes at libraries, other locations, and the county elections office.

But the problem could not just turn on urban/rural.  Compare, for example, the number of drop boxes in Multnomah County, which provides more than two dozen drop boxes distributed throughout the county for their 367,992 voters ( a Google Map of drop box locations in Multnomah County) to the  two locations provided by Clackamas County for their 190785 voters (perhaps Clackamas County provided more outdoor drop boxes for the 2012 election, but I can’t find any reference to this on their website).

I have been fielding a lot of inquiries about early voting after the 2012 election, and one of my pieces of advice is to make sure, as much as feasible, that all voters have equal access to the polling place, whether that be the early in-person satellite location or election day polling place.  I realize that Oregonians can just mail their ballot in, but data from the past has shown that as many as 25% of the voters choose to drop off their ballots.  The kind of discrepancy shown above is a point of concern.

Oregon will need to think long and hard about this one, and the legislature may need to pony up resources allowing Steve and Secretary Brown to install many more drop boxes around the state and may need to mandate some number of drop boxes based on population size and geography.

Image courtesy of Channel 13 News, Orlando FL

This report came across my news feed, courtesy of Channel 13 news in Orlando. I have looked but cannot find the referenced report at the Division of Elections.

One point of interest in the report is the longstanding concern of many political scientists with the way turnout is reported in this country.  When 64,000 voters choose not to vote the top of the ticket in battleground state during a highly contested presidential contest, it’s pretty obvious that turnout should include all voters who showed up and cast (or attempted to cast) a ballot.

“Top of the ticket” totals are inaccurate.

I hope that that the report includes a breakdown by mode of balloting.  There is reference in the report to higher undervote and overvote rates among absentee voters, something Charles Stewart and I found in a 2010 report.  More coming soon!

Congratulations to Zach Markovits, Charles Stewart, and the Pew team that has been working on the Election Performance Index for a number of years.  It’s been a long road, and there will inevitably be complaints from states who think they are ranked unfairly, and from advocates who think the rankings are insufficiently detailed or use the wrong measures.

From my perch inside this machine, it has been educational–and frustrating–to discover how difficult it still remains to collect comparable, consistent, and relatively complete measures of election performance for 50 states and the District of Columbia.  Many of the rankings are driven upwards and downwards not by performance but by simple reporting.  That will remain the case until states feel some pressure to collect and report a relatively complete set of indicators after each election.  (This kind of pressure is one of the rationales behind the index first proposed by Heather Gerken.)

If anyone wants a sense of how much forward progress has been made in a few short years, compare this report to the first exploratory effort, the Data for Democracy report that I assembled along with Dan Seligson in late 2008.  It is encouraging to see how much progress has been made between the two presidential election cycles.