Today’s Must-Read: Justin Levitt’s Testimony on the Census Adding the Citizenship Question

You can find Justin’s testimony here, from yesterday’s House Oversight hearing. From Justin’s testimony:

The testimony above demonstrates that existing data have been largely sufficient to enforce the Voting Rights Act. But — contrary to what we know from past practice — even if there were many marginal cases, and even if ACS survey data were not up to the task, and even if other means of establishing the electorate like SSRV data were unable to compensate, it is still extremely unlikely that adding a question on citizenship to the decennial enumeration in this climate would improve any entity’s ability to enforce the Voting Rights Act. Adding a question to the decennial enumeration may give the illusion of increased precision and greater statistical power. But in this climate, that illusion likely arrives with a far greater cost. There are known statistical means of compensating for the low response rates in a survey like the ACS, to account for those who refuse to participate. In contrast, there are few permissible means to compensate for low response rates in the enumeration.88 If a question on citizenship in the enumeration causes response rates to drop, that loss cannot be repaired. And in addition to the impact on overall census enumeration mentioned above — that is, in addition to the impact on the Census Bureau’s one inviolate constitutional duty — the inaccuracy produced by refusal to participate directly impacts the ability to enforce the Voting Rights Act. At the moment, the ACS data arrive with a minor margin of error. In the ostensible name of reducing that margin of error, adding a citizen question to the decennial enumeration may create a far larger one, by driving an even greater nonresponse rate. As explained above, vulnerable communities — including minority communities who might otherwise be seeking protection from the Voting Rights Act — are already chronically undercounted, and with the addition of a citizenship question, will be that much more likely to go untallied. 89 Which means that the enumeration is likely to systematically undercount precisely the people who most need the VRA. If the problem with the ACS survey is that it occasionally leaves doubt whether a population is sufficiently sizable to merit VRA protection, asking the question on the decennial enumeration will likely drive down participation so that it appears certain that the population is not sufficiently sizable to merit VRA protection. And because of the undercount, that certainty will be false. Adding the citizenship question to the decennial enumeration in this climate is likely to lead to an undercount of the true Latino citizen population of jurisdictions like Farmers Branch, and others on the cusp of VRA protection. And this predictable undercount means that, contrary to the Department of Justice’s assertions, adding the question in this climate is not likely to increase any entity’s real enforcement capacity. An ostensible performance-enhancing drug that cripples the patient does not enhance performance.


Comments are closed.