[EL] ELB News and commentary 6/24/11

[Joseph Lorenzo Hall]

On Fri, Jun 24, 2011 at 8:21 AM, Smith, Brad <BSmith at law.capital.edu> wrote:
>
> It seems to me that the Brooklyn case is a pretty compelling piece of evidence for voter ID. The question is whether the problem is really widespread enough to justify the solution. I generally think that laws based on zero tolerance or isolated scandals, no matter how compelling, tend to produce more harm than good.

A corollary is to what extent is the solution effective.

Much of the voter ID debate surrounds burdens to obtaining ID or the
documents required as a prerequisite to getting an ID.  There hasn't
been so much on whether or not voter ID requirements actually "work"
in that they accomplish some objective.  I've worked the polls in CA
in a number of elections and, at first, when people pulled out their
IDs, I had my pollworkers say, "Please put your ID away. ID is not
required to vote in California."  However, I stopped asking
pollworkers to do this as it was clear that the interaction between a
voter and pollworker when the voter is trying to spell their name or
street address aloud can be particularly inefficient compared to the
pollworker simply using the ID to look up the voter's name and address
in the poll book.  That's a real benefit of ID, but not one I think
should be part of any mandate.

I haven't seen a voter ID law that mandates pollworkers receive
additional training on 1) how to verify that a piece of government ID
is authentic and 2) how to observe voters presenting ID for signs that
they may be trying to associate themselves with an incorrect identity.
 The TSA trains its agents to spot both of these kinds of features:
they train agents to know security features are present in common
state DLs and passports, to use a UV/Blacklight source to illuminate
holographic laminates, to use magnification to look for inkjet dots (a
sign of a cheap fake ID) and for microprinting security features.
This training and measures tend to reduce false positives (where
someone erroneously claims to have a certain identity and passes the
ID check).

Of course, there are also issues of false positives (not to mention
not checking IDs of certain voters, say VBM).  I haven't seen a voter
ID law that specifies what voter's recourse when a poll worker falsely
claims their ID is fake or that they are not who they say they are.  I
suppose in many jurisdictions this would just fall back onto any voter
challenge procedures or require the person to submit a provisional
ballot.  Still, vote-by-mail will remain the loophole of all loopholes
for either fraud, coercion or vote-buying... that horse has already
left the barn and I don't see much that is politically feasible to get
it back in.

(And signature checking, the dominant form of authentication used in
VBM and provisional balloting, is not a particularly sophisticated
operation, in election administration.  While I'm aware of (and
concerned about) some jurisdictions using automated signature
verification tools and algorithms, many simply have temporary staff
that do a visual comparison... and many instruct those staffers to err
on the side of enfranchisement, knowing that people's signatures
change and that some people do things that an automatic verification
algorithm will not see as valid, like signing with their initials.)

best, Joe

--
Joseph Lorenzo Hall
ACCURATE Postdoctoral Research Associate
UC Berkeley School of Information
Princeton Center for Information Technology Policy
http://josephhall.org/