Michael Thielen digs himself deeper into a hole. The released chapter of my forthcoming book discusses the best academic studies available on voter i.d.  The simple reason my book does not cite any evidence from the Carter-Baker commission is its age: it predates the Indiana voter i.d. law (the first of the stricter voter id laws implemented).  Instead, I focus on the actual empirical work on voter id, including important work by Pitts, Erikson and Minnite, and Ansolabehere and Persily.  There is no original empirical research in the Carter-Baker report: only (unsubstantiated) claims about voter confidence, claims which Ansolabehere and Persily tested in an article published later in the Harvard Law Review.

The truth?

Virtually no evidence that voter i.d. laws prevent the kind of fraud (in person, impersonation voter fraud) which is a serious threat in American elections.

No evidence of a link between voter i.d. laws and voter confidence.

The chapter explains these points in detail.  I have yet to see good empirical evidence refuting either of these two points.  Certainly nothing in Carter-Baker does.

And, for the record, I have come out before and in the last chapter of the book I will come out again in favor of a national voter identification card, with all the costs paid for by the government, and registration conducted by the government (not third party groups) and the optional use of a thumbprint instead of the voter identification card for voters who choose to use it.  This would be part of a package of fundamental reforms to make our election system more uniform, efficient, and fair.