Ulric B. and Evelyn L. Bray Social Sciences Seminar

Abstract: Eligible voters sometimes return deficient mail ballots that will not be counted because they do not conform with particular procedural requirements. Vote curing policies allow such voters to still cast a valid ballot. We characterize vote curing along two dimensions: notice of a deficient ballot and the opportunity to correct it. We evaluate the effects of vote curing policies in Pennsylvania's 2024 general election because it featured both across and within-county variation in notice and correction opportunities. Individual-level data on voters who returned deficient mail ballots reveal that the likelihood such voters ultimately have their votes counted increased by about 25 percentage points (p.p.) when notice of the deficiency was provided before versus on Election Day, about 10 p.p. when voters had any opportunity to correct the deficiency before Election Day, and about 25 p.p. when deficient ballots were automatically returned. These large effects support emerging legal arguments challenging the lack of notice based on procedural Due Process principles or non-uniform correction opportunities based on the Equal Protection principle in Bush v. Gore. We conclude by considering how to structure vote curing to expand voter access without harming election integrity or exacerbating concerns about uniformity.

Joint with Michael Morse and Liz Stark