On May 27, 2011, Texas Governor Rick Perry signed into law Senate Bill 14, which requires that voters show government-issued photo identification at the polls in order to cast a ballot. Because Texas is a “covered” jurisdiction under Section 5 of the Voting Rights Act, the Department of Justice must “preclear” Senate Bill 14 before the law can take effect. In other words, Texas must demonstrate that the law has neither a discriminatory purpose nor a retrogressive effect on minority voters. After Texas’s initial submission urging the Department of Justice to preclear Senate Bill 14, the Justice Department requested more information from the state, including data – broken down by race – detailing the number of registered voters without ID. The data that Texas provided in response to the Department of Justice’s request clearly demonstrates that Senate Bill 14 has a harmful and discriminatory impact on Latino voters.
Texas sent the Justice Department a spreadsheet detailing the number of registered voters statewide who neither had an ID issued by the state’s Department of Public Safety (DPS) nor matched an ID record in the DPS database. The State also provided the number of Spanish surname registered voters who didn’t have ID and who didn’t match any record in the DPS database. Because the State doesn’t collect data on the race or ethnicity of registered voters, Spanish surnames serve as the best available proxy for Hispanic or Latino origin. Absent data on the race of registered voters, however, the State was unable to estimate the number of black and Asian voters who didn’t have a DPS-issued ID.
Texas’s data shows that there are currently 174,866 registered Latino voters who do not have a DPS-issued ID and who do not match an ID record in the DPS database; this is 6.28% of all registered Latino voters. By comparison, just 4.29% of the state’s non-Latino registered voters did not have a DPS ID and did not match an ID record the DPS database. This strongly suggests that Texas’s Latino voters would be adversely impacted by Senate Bill 14.
Because the State does not report rates of ID possession among non-Latino registered voters of different races, there is no way of knowing exactly what the rate of ID possession is among white voters. Therefore, at first glance, it might seem impossible to definitively prove that Senate Bill 14 has a discriminatory impact on Hispanic voters. But despite the obvious limitations associated with the information Texas provided to the Justice Department, the discriminatory impact of voter ID laws on the state’s Latino population is unmistakable.
Using data from the November 2010 Current Population Survey (CPS) Voting and Registration Supplement, we calculate the percentage of non-Latino registered voters in Texas who are white. We found that 77.8% are white. Applying that estimated percentage to the data provided by the State, we estimate that there were 7.8 million white registered voters in Texas. As noted above, 6.28% of registered Latino voters did not have a DPS-issued ID and did not match an ID record in the DPS database. For the percentage of registered white voters without DPS-issued ID to be equal to the percentage of registered Latino voters without DPS-issued ID, about 490,000 white registered voters must not have a DPS-issued ID. But as Texas’s own data demonstrates, there were only 429,026 non-Hispanic registered voters without DPS-issued ID in Texas!
In other words, it is impossible for white voters to lack ID at the same rates that Latino voters lack ID. Even if every non-Latino person of color had a DPS-issued ID, the percentage of white registered voters without ID would still be lower than the percentage of Latino registered voters without ID: just 5.5% of white registered voters would not have ID, compared to 6.3% of Hispanic voters. Given that several national studies have found that African-American voters are significantly less likely to possess a state-issued photo ID than white voters, this disparity is almost certainly much larger in reality.
When Texas filed its preclearance submission with the Department of Justice, the state ostensibly wanted to show that Senate Bill 14 would not have a discriminatory impact on people of color. But the State’s own data demonstrates exactly the opposite.
 The 95% confidence interval for the estimate ranges from 76.0% to 79.7%.
 The 95% confidence interval for this estimate ranges from 7.61 million to 7.99 million.
 The 95% confidence interval for this estimate ranges from about 478,000 to 501,000.
 The 95% confidence interval for the estimated percentage of white voters ranges from 5.37% to 5.63%.