Yesterday, the Washington Post released an internal Department of Justice memo in which career DOJ lawyers recommending rejecting the Georgia ID law because of its likely discriminatory effect on minorities (article here with links to the memo).
The memo and debate is interesting on several grounds.
First, most people have little idea how the preclearance process works, and this memo provides an illustration of the facts that are used in making these determinations.
Second, the case illustrates how important it is to think through the implications of changes in election laws. Here, when you make ID requirements very strict, you have to think through questions like: How accessible are driver’s licence offices? How costly is public transportation (both the nominal cost of the ride and the opportunity costs to the person who has to get the license)?
Third, the social science used in this work are very interesting. The DOJ uses very straightforward analyses of who is affected by the change. They found, for example, that minorities are more likely to live in counties without a driver’s licence office and also not have a car. However, the data on driver’s licenses for Georgia are abysmal according to the DOJ analysis (and are likely to be abysmal in most places). People who have moved, died, or otherwise should not be in the database remain there. (Clearly there is a need for states to exchange better data here!) This makes analyzing racial differences among people with a driver’s licence and without a driver’s licence.
Finally, the articles in the Thursday and Friday Washington Post also get at the political nature of elections. Politics is largely shaped by those who control the rules of the game. It should not be surprising that there are politics involved in this process. Of course, it remains disappointing that, once again, basic analysis of social science data gets trumped by the better political argument.