Association for Public Policy Analysis & Management

The recent propagation of voter I.D. laws has continued with 17 states requiring photo identification to vote as of 2018. Advocates claim it boosts accuracy and confidence in election outcomes, while others caution an additional cost to voting. I examine the effects of one of the strictest voter I.D. laws in the nation, Texas S.B. 14, which first went into effect in 2014. Since this law required one of seven forms of photo identification to be able to vote, it seems most likely to see an effect in this context, especially given Texas’s large and diverse population.

I proceed by first matching individual-level data on state-issued driver’s licenses and identification cards to voter history 1996-2016. To identify effects, I then compare people with and without permissible photo I.D. over time. Results show that registered voters who had not been issued a photo I.D. were 2 percentage points less likely to vote in the 2014 midterm election. In the 2016 presidential election, the effect is more than 13 percentage points.