North Carolina offers its residents the opportunity to cast early in-person (EIP) ballots prior to Election Day, a practice known locally as “One-Stop” voting. Following a successful legal challenge to the state’s controversial 2013 Voter Information and Verification Act, North Carolina’s 100 counties were given wide discretion over the hours and locations of EIP voting for the 2016 General Election. This discretion yielded a patchwork of election practices across the state, providing us with a set of natural experiments to study the effect of changes in early voting hours on voter turnout. Drawing on individual-level voting records from the North Carolina State Board of Elections, our research design matches voters on race, party, and geography. We find little evidence that changes to early opportunities in North Carolina had uniform effects on voter turnout. Nonetheless, we do identify areas in the presidential battleground state where voters appear to have reacted to local changes in early voting availability, albeit not always in directions consistent with the existing literature. We suspect that effects of changes to early voting rules are conditional on local conditions, and future research on the effects of election law changes on turnout should explore these conditions in detail.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
McDonald’s United States Election Project is hosted at http://www.electproject.org/early_2016 (last accessed May 18, 2017).
See (Herron and Smith, 2015) as well as declarations and reports by plaintiffs’ experts at http://moritzlaw.osu.edu/electionlaw/litigation/documents/League1551.pdf and http://moritzlaw.osu.edu/electionlaw/litigation/documents/League1553.pdf and http://moritzlaw.osu.edu/electionlaw/litigation/documents/League1555.pdf (last accessed July 4, 2017).
The Fourth Circuit’s opinion is available at http://electionlawblog.org/wp-content/uploads/nc-4th.pdf (last accessed May 18, 2017).
2008 and 2012 election results for North Carolina, see coverage in the New York Times, available at http://www.nytimes.com/elections/2008/results/states/north-carolina.html and http://www.nytimes.com/elections/2012/results/states/north-carolina.html (last accessed May 18, 2017).
For coverage of this race, see “NC governor’s race could be nation’s most competitive,” The Charlotte Observer, available at http://www.charlotteobserver.com/news/politics-government/election/article66261897.html (last accessed May 18, 2017) and “In North Carolina, a Governor’s Race Is Too Close to Call,” The New York Times, available at https://www.nytimes.com/2016/11/09/us/politics/governors-races-statehouses-eric-holcomb.html (last accessed May 18, 2017).
See “Appeals court strikes down North Carolina’s voter-ID law,” The Washington Post, available at https://www.washingtonpost.com/local/public-safety/appeals-court-strikes-down-north-carolinas-voter-id-law (last accessed May 18, 2017).
See “NC Republican Party seeks ‘party line changes’ to limit early voting hours,” The News and Observer, available at http://www.newsobserver.com/news/politics-government/election/article96179857.html (last accessed May 18, 2017).
See Zach Roth, “North Carolina GOP Out to Limit Crucial Early Voting Period,” NBC News, available at http://www.nbcnews.com/politics/politics-news/north-carolina-gop-out-limit-crucial-early-voting-period-n633571 (last accessed May 30, 2017).
In the federal litigation surrounding VIVA, North Carolina contended that the “same hours” provision of the legislation would be “materially the same as the early voting opportunities before the bill was enacted.” The Fourth Circuit noted in its opinion, however, that, “A critical problem with the State’s argument is that the law provided that any county could waive out of this requirement, and, in 2014, about 30% of the counties did waive out of the requirement.” See United States Court of Appeals for the Fourth Circuit, http://moritzlaw.osu.edu/electionlaw/litigation/documents/Opinion72916.pdf, p. 52.
See “Early voting reduced in 23 NC counties; 9 drop Sunday voting after NCGOP memo,” The News & Observer, September 6, 2016, available at http://www.newsobserver.com/news/politics-government/election/article100284162.html (last accessed June 1, 2017).
All county-level changes to EIP availability are drawn from a document prepared by the North Carolina State Board of Elections (NCSBE), available at http://dl.ncsbe.gov.s3.amazonaws.com/One-Stop_Early_Voting/2016/2016_Early_Voting.xlsx (last accessed July 4, 2017).
A full description of the changes made to EIP offerings by each county in North Carolina can be found in Online Appendix Table 29. Online Appendix Table 30 additionally includes each county’s partisan, racial and socio-demographic makeup.
In a few cases there are discrepancies between North Carolina voter history files and absentee files. For our purposes, a voter is only considered to have voted EIP in a given election if this voter is listed as having done so in the election’s associated absentee file.
North Carolina voter files contain a variable named voter_status_des. We disregard all registered voters for whom this variable is not equal to “ACTIVE.” There are 6,016,209 and 6,010,025 active voters in our 2012 and 2016 North Carolina voter files, respectively. Our voter file merge preceded as followed. We merged 2012 and 2016 voter files using North Carolina voter identification numbers that are contained in a field called ncid. The 2016 General Election absentee file contains this identification field, which we used to merge the file with our two voter files. However, the 2012 General Election absentee file does not have this field. To merge this file, we first formed an identifier using a county-specific registration number (voter_reg_num), voter name, and county. We use this identifier to merge the 2012 absentee file with our voter files. We dropped all records that had an absentee field, either 2012 or 2016, of ballot_rtn_status that was one of the following: “Cancelled,” “Spoiled,” “Wrong voter,” “Not vote,” “Duplicate,” “Returned undeliverable,” “Witness info incomplete,” “No application,” “E-transmission failure,” “ID not provided,” “Rejected,” “Returned after deadline,” “Signature different,” “Signed by other,” “Pending,” or “Voter signature missing.” We ignore all 2016 registered voters who were not registered in 2012 as well as all non-active voters and all Latino registered voters. Overall, our analyses that link 2012 and 2016 voter turnout levels use 3,291,118 records that are either Black or white and affiliated with the Democratic or Republican parties. One caveat to the data which we are unable to comprehensively address relates to individuals who moved between counties between 2012 and 2016. There is no way to identify individuals who moved more than once, and one-time movers are a small fraction of the voter file. Further, movers only pose a threat to the analysis to the extent that moving was somehow correlated with the treatment, changes to EIP laws. We have no reason to believe this is the case. So, while movers introduce noise into the data, we do not anticipate that it poses a significant threat to the overall analysis.
Kaplan and Yuan (2017) employ a similar design leveraging geography in their paper on the impact expanded EIP opportunities in Ohio. While they do not examine differences among racial subgroups, they do demonstrate that women, individuals of working and childrearing age, and Democrats are most impacted by changes to the number of days available to vote early in person. To determine the casting of EIP ballots, Weaver (2015) estimates EIP voting rates, by race, in Ohio—three largest counties in the 2008 and 2012 General Elections using ecological inference models and by leveraging spatial distributions of minorities in relatively racially homogeneous geographies.
The critical assumption in this research design, then, is that individuals living along a shared county border are comparable, or more comparable than they are to individuals not living along that same shared border. To assess the validity of this assumption with census data, we identified census tracts along a subset of county borders and compared the means of key demographics across counties. We then compared the difference in means to the difference in means between a county in the pair and a randomly selected census tract elsewhere in the state. On balance, border tracts in the paired counties were more comparable to one another than they were to a randomly selected census tract. This analysis is described in the Online Appendix and displayed in Online Appendix Tables 1–10.
We focus on Black voters, as compared to whites, because Black voters comprise the largest minority group in the state and were the focus of recent conflicts in the courts over attempts to alter the composition of the North Carolina electorate through changing election rules. For this same reason, we focus on individuals who are registered to vote with one of the two major parties. It may be the case, however, that Latinos, a growing sector of the population in North Carolina, and voters who are not affiliated with one of the two major parties, were also impacted in important ways by changes to EIP voting. The same can be said of Black Republicans. We conducted additional analysis among these latter party and race subgroups, and our findings largely corroborate those presented in the body of the text. As such, we elected to include them in Online Appendix, and additional tests among all groups are displayed in Online Appendix Figs. 1–10.
Although we are wary of drawing conclusions from regression models of the EIP voting data that we use in the body of our analysis, we nonetheless consider regression models of 2016 turnout among individual voters as a robustness check on our results. Our models use all active 2016 voters statewide who were also eligible to vote in 2012 in North Carolina, and the supplemental regression analysis can be found in Online Appendix Tables 18–22. The results are mixed, corroborating the matched analysis.
Related to this, we cannot rule out the possibility that individuals did not know about changes to EIP offerings in 2016, nor can we identify the extent to which voters were or were not aware of changes to the law. This would only bias our results if lack of knowledge about the law were correlated with types of changes made by counties, which while a possibility, we have no reason to believe is the case. Nevertheless, the possibility that individuals were unaware of the treatment in our quasi-experiment remains a limitation of the research.
This table is based on files downloaded from the NCSBE website and not on our two voter files. For example, data on absentee voting details for 2012 General Election was retrieved from http://dl.ncsbe.gov/index.html?prefix=ENRS/2012_11_06.
Findings do not differ when all tests are plotted. We include only the top 40 tests in order to improve the clarity of the presentation of the results. Plots of all tests can be found in Online Appendix Figs. 1–10.
As the preceding and forthcoming graphs demonstrate, it is difficult to say definitively how changes to EIP offerings impacted voters in North Carolina in any broad sense. Instead, the exact effects are context dependent. Yet, one may wish for a more universal assessment of what, on balance, the series of tests we conducted convey. We therefore counted the number of pairwise comparisons that are either positive or negative by type of institutional change among race and party subgroups, in order to offer a more succinct account of the impact of reforms. The results of this global test are as decidedly mixed as the display of individual tests presented here suggest they would be. This additional analysis is located in Online Appendix Tables 11–16.
Alvarez, R. M., Bailey, D. , & Katz, J. (2007). The effect of voter identification laws on turnout. Working Papers 1267 California Institute of Technology, Division of the Humanities and Social Sciences. https://ideas.repec.org/p/clt/sswopa/1267.html.
Amos, B., Smith, D. A., & Claire, C. S. (2017). Reprecincting and voting behavior. Political Behavior, 39(1), 133–156.
Anderson, C. J. (2007). The interaction of structures and voter behavior. In R. J. Dalton & D. K. Hans (Eds.), The oxford handbook of political behavior (pp. 589–609). New York: Oxford University Press.
Bentele, K. G., & O’brien, E. E. (2013). Jim crow 2.0? Why states consider and adopt restrictive voter access policies. Perspectives on Politics, 11(04), 1088–1116.
Berinsky, A. J. (2005). The perverse consequences of electoral reform in the United States. American Politics Research, 33(4), 471–491.
Biggers, D. R., & Smith, D. A. (2018). Does threatening their franchise make registered voters more likely to participate? Evidence from an aborted voter purge. British Journal of Political Science forthcoming.
Bowler, S., & Donovan, T. (2012). The limited effects of election reforms on efficacy and engagement. Australian Journal of Political Science, 47(1), 55–70.
Bradner, E. (2016). Why North Carolina is so important in 2016.
Burden, B. C., Canon, D. T., Mayer, K. R., & Moynihan, D. P. (2014). Election laws, mobilization, and turnout: The unanticipated consequences of election reform. American Journal of Political Science, 58(1), 95–109.
Burden, B. C., & Gaines, B. J. (2015). Presidential commission on election administration: Absentee and early voting: Weighing the costs of convenience. Election Law Journal, 14(1), 32–37.
Davidson, C. (1992). The voting rights act: A brief history. In B. Grofman & C. Davidson (Eds.), Controversies in minority voting: The voting rights act in perspective (pp. 7–34). Washington, DC: The Brookings Institution.
Del Pinal, J. (1995). Hispanics-Latinos: Diverse people in a multicultural society. Washington, DC: Bureau of the Census.
Delgado, R., & Stefancic, J. (1998). The Latino/a condition: A critical reader. New York, NY: NYU Press.
Erikson, R. S., & Minnite, L. C. (2009). Modeling problems in the voter identification—voter turnout debate. Election Law Journal, 8(2), 85–101.
Giammo, J. D., & Brox, B. J. (2010). Reducing the costs of participation: Are states getting a return on early voting? Political Research Quarterly, 63(2), 295–303.
Graham, D. A. (2016). Have North Carolina republicans found a way to reinstate discriminatory voting rules.
Graham, D. A. (2018). North Carolina’s landmark ruling against partisan gerrymanders.
Grimmer, J., Hersh, E., Meredith, M., Mummolo, J., & Hall, C. (2017). Comment on voter identification laws and the suppression of minority votes. Unpublished working paper. https://stanford.edu/~jgrimmer/comment_spsfinal.pdf.
Grofman, B., Handley, L., & Niemi, R. G. (1994). Minority representation and the quest for voting equality. New York, NY: Cambridge University Press.
Gronke, P. (2012). Early voting: The quiet revolution in American elections. In M. Streb (Ed.), Law and election politics: The rules of the game (pp. 134–148). New York, NY: Routledge.
Gronke, P., Galanes-Rosenbaum, E., Miller, P. A., & Toffey, D. (2008). Convenience voting. Annual Review of Political Science, 11, 437–455.
Hajnal, Z., Lajevardi, N., & Nielson, L. (2017). Voter identification laws and the suppression of minority votes. The Journal of Politics, 79(2), 363–379.
Hanmer, M. J., & Traugott, M. W. (2004). The impact of voting by mail on voter behavior. American Politics Research, 32(4), 375–405.
Hasen, R. L. (2005). Beyond the margin of litigation: Reforming US election administration to avoid electoral meltdown. Washington & Lee Law Review, 62, 937.
Herron, M. C., & Smith, D. A. (2014). Race, party, and the consequences of restricting early voting in Florida in the 2012 general election. Political Research Quarterly, 67(3), 646–665.
Herron, M. C., & Smith, D. A. (2015). Race, Shelby County, and the voter information verification act in North Carolina. Florida State University Law Review, 43, 465–506.
Hicks, W. D., McKee, S. C., & Smith, D. A. (2016a). A bipartisan election reform? Explaining support for online voter registration in the American States. American Politics Research, 44(6), 1008–1036.
Hicks, W. D., McKee, S. C., & Smith, D. A. (2016b). The determinants of state legislator support for restrictive voter ID laws. State Politics & Policy Quarterly, 16(4), 411–431.
Highton, B., & Wolfinger, R. E. (1998). Estimating the effects of the National Voter Registration Act of 1993. Political Behavior, 20(2), 79–104.
Hill, D. (2003). A two-step approach to assessing composition effects of the National Voter Registration Act. Electoral Studies, 22(4), 703–720.
Hood, M. V., & Bullock, C. S. (2012). Much ado about nothing? An empirical assessment of the Georgia voter identification statute. State Politics & Policy Quarterly, 12(4), 394–414.
Kaplan, E., & Yuan, H. (2017). Restrictive voting laws, voter turnout, and Partisan vote composition: Evidence from Ohio. Unpublished manuscript.
Larocca, R., & Klemanski, J. S. (2011). US state election reform and turnout in presidential elections. State Politics & Policy Quarterly, 11(1), 76–101.
Leighley, J. E., & Nagler, J. (2014). Who votes now. Princeton, NJ: Princeton University Press.
Neeley, G. W., & Richardson, L. E. (2001). Who is early voting? An individual level examination. The Social Science Journal, 38(3), 381–392.
Oliver, J. E. (1996). The effects of eligibility restrictions and party activity on absentee voting and overall turnout. American Journal of Political Science, 40, 498–513.
Rigby, E., & Springer, M. J. (2011). Does electoral reform increase (or decrease) political equality? Political Research Quarterly, 64(2), 420–434.
Southwell, P., & Burchett, J. (2000). Does changing the rules change the players? The effect of all-mail elections on the composition of the electorate. Social Science Quarterly, 81, 837–845.
Southwell, P. L., & Burchett, J. (1997). Survey of vote-by-mail senate election in the state of Oregon. Political Science & Politics, 30(01), 53–57.
Stein, R. M. (1998). Early voting. The Public Opinion Quarterly, 62(1), 57–69.
Stein, R. M., & Garcia-Monet, P. A. (1997). Voting early but not often. Social Science Quarterly, 78, 657–671.
Timpone, R. J. (1998). Structure, behavior, and voter turnout in the United States. American Political Science Review, 92(01), 145–158.
Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and equality: Civic voluntarism in American politics. Cambridge: Harvard University Press.
Weaver, R. (2015). The racial context of convenience voting cutbacks: Early voting in Ohio during the 2008 and 2012 U.S. Presidential Elections. SAGE Open, 5(3), 2158244015591825.
Williams, M. S. (2000). Voice, trust, and memory: Marginalized groups and the failings of liberal representation. Princeton, NJ: Princeton University Press.
Wolfinger, R. E., & Hoffman, J. (2001). Registering and voting with motor voter. Political Science & Politics, 34(1), 85–92.
Wolfinger, R. E., & Rosenstone, S. J. (1980). Who votes? (Vol. 22). New Haven, CT: Yale University Press.
The authors would like to thank Matt Barreto, Loren Collingwood, Mackenzie Israel Trummel and Tye Rush for their support and feedback on this project. We are also grateful to our anonymous reviewers. Replication files are available here: https://doi.org/10.7910/DVN/2JCJLX.
Electronic supplementary material
Below is the link to the electronic supplementary material.
About this article
Cite this article
Walker, H.L., Herron, M.C. & Smith, D.A. Early Voting Changes and Voter Turnout: North Carolina in the 2016 General Election. Polit Behav 41, 841–869 (2019). https://doi.org/10.1007/s11109-018-9473-5
- American politics
- Political behavior
- Election reform
- Racial and ethnic politics