🧭 What Was Tested: Whether state-level strict voter identification laws reduced voter registration or turnout in the United States, or produced unequal effects across race, gender, age, or party affiliation.
📊 Nationwide Panel and Natural Experiment: Uses a difference-in-differences design on a nationwide panel of voter records covering 2008–2018 with 1.6 billion observations. A large range of specifications is reported; the most demanding model includes state, year, and voter fixed effects plus state- and voter-level time-varying controls.
📈 Key Findings:
- No negative effect on registration or turnout overall or for any subgroup defined by race, gender, age, or party.
- In the most demanding specification, point estimates are -0.1 percentage points for both overall registration and turnout (95% CIs: [-2.3; 2.1] and [-3.0; 2.8], respectively).
- The estimated effect on nonwhite turnout relative to whites is +1.4 percentage points (95% CI: [-0.5; 3.2]).
- The lack of a negative turnout effect cannot be explained by backlash: campaign contributions and self-reported political engagement do not indicate a voter reaction against the laws.
- Party mobilization appears to have increased: the probability that nonwhite voters were contacted by a campaign rose by 4.7 percentage points, suggesting outreach may have offset modest barriers.
- No detectable effect on fraud, either actual or perceived.
🔍 Why It Matters: Across a massive, nationwide dataset and rigorous controls, strict voter ID laws do not measurably reduce registration or turnout and do not increase fraud. These results imply that resources aimed at improving election outcomes may be better spent on other reforms than on strict ID requirements.
