The 2012 election has come and gone but debates about "strict" photo voter ID laws seem to be here to stay. Lawmakers in some states are already signaling their intention to pursue photo ID laws in their new legislative sessions. Despite a setback at the ballot box in November, the architect of Minnesota’s photo ID proposal, Senator-elect Mary Kiffmeyer, remains committed to enacting an ID requirement in her state. Republicans in neighboring Wisconsin have also pledged to redouble efforts to implement photo ID, as have lawmakers in the newly-GOP-dominated Arkansas General Assembly.
The photo ID debate is often framed as a simple tug-of-war between fraud and disenfranchisement. This framing belies the complexity of the issue. First, it is surprisingly difficult to measure and compare fraud and disenfranchisement. Second, a range of other issues—from economic to legal—introduce added complications.
This month’s Canvass provides a brief overview of five key issues to consider when evaluating photo ID proposals.
Photo ID laws seem like common sense. What better way to ensure that a potential voter is who he says he is than to ask for an ID with his picture on it? The catch, according to opponents of such laws, is that not all eligible voters have photo IDs, and IDs can be difficult or expensive to obtain. Photo ID opponents argue that a strict ID requirement is tantamount to disenfranchisement for any eligible Americans who don’t have, and can’t get, an ID.
How can policymakers assess this claim? It’s fairly easy to find out how many people do vote; we can check the election returns. But how do we know how many people would be prevented from voting due to ID laws?
Researchers tend to address this question in one of two ways, asking either how many people lack photo ID (and how hard is it for them to get it) or what effect ID laws have on voter turnout.
Their answers to these questions have been mixed at best. An oft-cited report by the Brennan Center for Justice says that up to 11 percent of Americans lack photo ID but other studies offer lower estimates, some as low as 1 percent. Researchers have also reached wildly different conclusions about the impact of ID laws on voter turnout, ranging from a negative effect to a positive effect to no clear effect at all.
Some of the variation can be traced to differences in research methodology. For example, some researchers compare voter rolls to motor vehicle agency databases to estimate the number of people without ID. Others use surveys, sampling either a) all Americans, b) registered voters or c) likely voters, each with different results.
The sheer complexity of the subject matter also accounts for some of the disparity. For example, voter turnout is influenced by many different factors—from a recent move to the weather on Election Day—and disentangling these influences can be tricky.
Concern about election fraud is the most commonly-cited reason for backing photo ID measures. Much like "stealing," ‘election fraud’ is a category of crimes rather than a single action. Photo ID laws are directed at a subset of these crimes—most notably, impersonating another voter at the polls. So far, evidence of such fraud has been elusive:
Opponents of photo ID laws take this to show that voter fraud is not widespread. Supporters offer a different analysis. Far from establishing that no voter fraud problem exists, they say, the lack of evidence actually highlights another fraud-related problem: our lack of tools for detecting fraud. Without ID laws, they argue, we can’t even detect, let alone deter, voter fraud.
Regardless of the impact on fraud itself, photo ID proponents maintain that ID laws are necessary to combat the perception of fraud. They say that a widespread belief that fraud is rampant has caused a crisis of confidence in the electoral system. This, in turn, is driving down political engagement and participation. Like the link between photo ID laws and voter turnout, however, the relationship between perceptions of fraud, electoral confidence and political participation may not be as straightforward as it seems.
(continued next column)
Like most public policies, photo ID proposals carry costs—both economic and administrative. Potential expenses associated with photo ID include:
- Voter education
- Additional poll workers and training materials
- Free photo IDs
Many states that have considered photo ID legislation have produced "fiscal notes," or estimates of the laws’ financial impacts. These estimates range from ‘no fiscal impact’ to over $7 million. (Links to fiscal notes are available here.)
Reports from states that have implemented photo ID laws provide actual costs:
- Georgia: $1.6 million ($770,000 for purchase and maintenance of new voting technology and $840,000 for voter education)
- Indiana: $12.2 million ($2.2 million for voter outreach and education between 2005 and 2010 and $10 million for free photo IDs from 2007 to 2010)
However, these reports might not include all expenses associated with photo ID implementation, such as the cost of defending the legislation in court.
4. Legal issues
Photo ID laws are almost guaranteed to come under legal scrutiny. Three of the four states that had photo ID in place in November faced legal challenges at some point (Georgia, Indiana and Tennessee, but not Kansas). Cases have been filed—or are anticipated—in all five of the states that are currently actively pursuing photo ID (Mississippi, Pennsylvania, South Carolina, Texas and Wisconsin). Therefore, both supporters and opponents of photo ID should be aware of the legal issues involved.
Some of these legal issues are common to all states. Challenges under the U.S. Constitution—such as the Equal Protection Clause-based argument raised by plaintiffs in Indiana’s 2008 case, Crawford v. Marion County Election Board—could surface in any state. (The court permitted Indiana to proceed with its strict photo voter ID law.)
Other issues vary from one state to another. Per Section 5 of the Voting Rights Act of 1965, states with a history of discriminatory voting practices must get approval from the U.S. Department of Justice for election law changes. States not covered by Section 5, such as Indiana and Pennsylvania, don’t have to clear this hurdle. Similarly, some states' constitutions include an affirmative right to vote. In these states, stricter voter ID laws face additional legal hurdles.
5. Public Opinion
Legislators who are considering photo ID legislation might be interested in their constituents’ thoughts about such laws. Nationally, ID requirements are very popular. National polls conducted before the 2012 general election found support for photo ID hovering between 70 and 80 percent:
Enthusiasm for photo ID laws may vary somewhat by region. A survey conducted around the 2006 general election found that support for photo ID was highest in the South and lowest in the Northeast:
- South: 81 percent
- West: 76 percent
- Midwest: 76 percent
- Northeast: 68 percent
Support can also vary from state to state. Mississippi voters passed a photo ID amendment with 62 percent of the vote in 2011. In Minnesota, on the other hand, support for a proposed amendment dropped from 65 percent in January 2012 to 46 percent on Election Day.
In the coming months, NCSL will release a series of briefs on these five voter ID-related issues—the potential for disenfranchisement, fraud and its prevention, implementation costs and considerations, legal issues and public opinion—with the aim of helping lawmakers engage in careful discussion of 2013's photo ID proposals. In the meantime, please see our Voter ID web page for information about legislative and legal action on voter ID.
Author’s bio: Karen Shanton is a legislative studies specialist and ACLS public fellow at NCSL. Contact her at email@example.com or (303) 856-1481.
The election may seem like old news to the media now, but NCSL continues to get many calls about party control of legislatures. Are you interested in:
- How many D's and R's serve in each chamber?
- Which states have veto-proof legislatures?
- The results of ballot measures across the nation?
If you’re looking for that kind of detail, check out NCSL’s StateVote election coverage for results. The commonly requested items are our party composition chart and map.
For analysis, NCSL’s blogs, The Thicket and Prop*50, deliver continuing coverage of legislative elections and ballot measures, respectively.
ONE-MINUTE ELECTION ANALYSES
"In one minute or less, what was noteworthy about how the election was run in your neck of the woods?"
That is the question we asked a variety of election experts a few days after the presidential election. (The New York Times did a similar wrap-up a few days later, even talking to a couple of the same people.) This month we're replacing two regular interview features in The Canvass with these one-minute analyses. Previous Q and A's with committee chairs who are steeped in election policy and administrators who put that policy into practice are archived for your reading pleasure, too. These answers are listed in alphabetical order, by state.
—“This year voters had an easier time than ever getting the information they needed, such as where to vote and what to bring to vote, thanks to technology advances. More voters than ever before were able to get registered easily, too, through online registration. We still see room for improvement for future years in terms of running really good elections; some of that will develop as states begin to collect and evaluate data about how the processes worked this year.”
Sean Greene, Pew Election Initiatives (District of Columbia)
—“The first thing that comes to mind is Florida's restrictions on early voting locations. State law does not allow the use of facilities with ample space, such as community centers or recreation centers, to process the volume of voters who want to vote early. That's why we have lines. Next, the legislature added 11 amendments to our ballot. Two of them were so long that they took up an entire page combined. (Two pages, if you are producing multi-lingual ballots.) This means it takes longer to vote and increases the physical load on poll workers.”
Lori Edwards, Supervisor of Elections, Polk County (Florida)
—“The only issues we experienced in Idaho seemed to be relatively minor and localized. There was an issue with a bond election in one city with split precincts and it appeared that some people may have been given the wrong ballots. About 25 percent of the Idaho electorate voted early or absentee, so that helped keep the lines down.”
Gary Moncrief, Political Scientist, Boise State University (Idaho)
—“Presidential elections expose all the stress points in our election administration because that’s when the system is most pressure-tested. In this year’s election, I think we saw process issues specifically around advance voting. Expectations have increased, which means we should consider having additional advance voting locations. There’s a cost with that, in people, rent, and supplies, so we’re going to need to model all of that out.”
Brian D. Newby, Election Commissioner, Johnson County (Kansas)
—“In Maine we saw very high, at times record-setting, new registrations on Election Day and a huge influx of independent expenditures, especially from national labor groups that targeted first-time incumbents. It was an election like Mainers have never seen and I feel we lost a large part of what has made our state special: local candidates who focused more on retail politics. Maine has always been a state where campaign spending was fairly low. Now the landscape has been changed by special interest funding.”
Representative Andre Cushing III (R, Maine)
—“One factor about this election stands out. Despite the increased opportunities for voter registration, extended early voting, huge increases in mail absentee ballots, and other factors designed for the convenience of the voter, neither voter registration nor turnout seems to have been affected very much.”
Dick Smolka, Editor, Election Administration Reports (Maryland)
—“Guilford County, North Carolina, opened 16 early voting sites in the middle of October and an additional six the last week of the month. The polling places were open seven days a week, with limited hours on the weekends. On the last day of early voting, lines in any location did not exceed more than one hour and 45 minutes. We had 162,000 early votes cast, which helped to reduce voting lines on Election Day. An additional 80,000 voters voted at 165 polling places on Election Day with no voting lines reported anywhere.”
George Gilbert, Director of Elections, Guilford County (North Carolina)
—“The 2012 election went remarkably smoothly in North Dakota. There were no complaints about the voting process; the only stories were about winners and losers in the election. A twist occurred in this year’s election regarding campaigning on Election Day. Federal district judge Daniel Hovland ruled against a North Dakota law that had banned the practice, saying it was unconstitutional and violated free speech. Therefore, people could campaign and keep lawn signs out on Election Day.”
Senator Ray Holmberg (R, North Dakota)
—“Virginia faced several challenges in this year’s presidential election. Turnout was a bit higher than in 2008, but officials had fewer resources given recent challenges for local budgets. The number of provisional ballots was fairly high, due to uninformed voters as well as redistricting. Many applications for ballots and actual ballots arrived late because of Hurricane Sandy. Finally, the election official (poll worker) shortage is beyond acute.”
Cameron Quinn, Election Administrator, Fairfax County (Virginia)
—“Virginia voters experienced long lines because there seemed to be too few machines, inefficient check-in procedures such as too few voting officials checking people in, machine malfunctions, etc. Additionally, it was quite cold in Virginia (one of the first cold weeks this fall), so the perseverance of voters in the face of such long lines is a real take-away from this election. One thing I was surprised not to see much of is voter integrity advocates challenging voters, particularly in minority precincts.”
Rebecca Green, Co-director of the Election Law Program at William and Mary Law School (Virginia)
—“The state of Washington ran its first all-mail election in a presidential election year and it appears that we are headed for turnout near 85 percent statewide.”
Representative Sam Hunt (D, Washington)
ONE BIG NUMBER: 7.6 MILLION
That is the number of “lost votes” in the 2008 presidential election due to mail-in ballots. So says the report, Voting: What Has Changed, What Hasn’t, and What Needs Improvement, produced by the Voting Technology Project from the California Institute of Technology and Massachusetts Institute of Technology. Researcher and MIT professor Charles Stewart III says that the pipeline that moves mail ballots between voters and election officials is “very leaky.” Votes “leak out” if a mail ballot is requested but no ballot is received (3.9 million); ballots that were sent out are never received for counting (2.9 million); or ballots are rejected for counting for one reason or another. All told, 21 percent of requests for absentee ballots do not result in a counted vote.
BOOKMARK THIS: THE UNITED STATES ELECTION PROJECT
Election geeks always want to know what will happen to turnout, based on a change in election law. The United States Election Project, run by George Mason University professor Michael McDonald, is the place to go for this data. (A hint: the weather is a better predictor of turnout than any single election policy.) Within the webpage, go here for 2012 turnout data; it is being updated on an almost-daily basis.
Ballots in just two states this year asked voters to weigh in on election policy, and in both states voters were skeptical of change. Minnesota voters rejected a constitutional amendment that would have required voters to show photo IDs at the polls, and Arizonans rejected a proposal to shift to a “top two” primary system. ("Top two" is used in California and in Washington. Candidates, regardless of party, run on the same primary ballot, and only the top two advance to the general election.)
Contested elections did not play a big role in November’s election—but there is always the next election, and the next, and the next, where tight races could lead to recounts and litigation. Law professor Joshua A. Douglas has done us all a favor by writing Procedural Fairness in Election Contests, a scholarly paper backed up with the facts on how all states manage contested elections.
The Election Law Program at William and Mary Law School offers web-based lectures by national experts on key election law issues. From this link, choose "video modules." While state judges are the intended audience, legislators and other election geeks are most welcome to look and learn as well.
The Alabama case, Shelby v. Holder, is now on the U.S. Supreme Court docket. The justices will review this case in terms of the constitutionality of Section 5 of the Voting Rights Act. That’s the section that requires some states and jurisdictions to get pre-clearance for any and all changes to election law before implementing them, based on past histories of discrimination. http://moritzlaw.osu.edu/electionlaw/litigation/ShelbyCountyAlabamav.Holder.php makes available all the filings related to this case.
FAST and SIMPLE are the monikers of two bills dealing with elections that have been introduced in Congress in the post-election lame-duck session. FAST comes from the Senate, and SIMPLE comes from the House, and both are sponsored by Democrats. These are compared in Doug Chapin’s blog, The Election Academy. Both infringe on states' rights.
NCSL is building its YouTube presence. In terms of elections, after the last couple of years we’ve posted short interviews with a series of experts: John Lindback and David Becker talking about voter registration; Neil Simon pitching the idea of permitting international election observers; Zach Markovits discussing how we know when elections are run right; Shane Hamlin and Tammy Patrick explaining how online voter registration is handled in Washington and Arizona; and Doug Chapin addressing cost as a consideration in election policy.
Election law observers are familiar with NCSL’s databases on election-related legislation. One covers 2001-2010 and the second one covers 2011 through the present. You may be less familiar with the databases that follow ballot measures, legislation dealing with the initiative and referendum process and term limits legislation. All are open to the public and designed to be user-friendly. If you’d like a more personal introduction to these databases, please call us at (303) 664-7700.
FROM NCSL's ELECTIONS TEAM
At the moment, NCSL is experiencing the calm between storms—and we are pretty sure that our readers are, too. The storm that’s gone by is November’s election. The storm that’s coming will hit when legislatures begin to open for business after January 1st. Based on the nature and the number of in-depth information requests we're receiving from legislators and staff, we predict an interesting year ahead. If you have questions you’d like answered as you prepare bills, now is a good time to ask. We’ll be here through the holidays, ready to work with you on your projects.
Thanks for reading.
Jennie Bowser, Karen Shanton and Wendy Underhill