Email No Panacea for Our Military

The MOVE Act is getting  a lot of attention of late, primarily for the failure of a number of jurisdictions to comply with its provisions.   The biggest obstacle to compliance was the primary schedule that shortens the window between certification of the ballot and the sending of ballots overseas.

One element of the MOVE Act that received a lot of attention was the provisions to allow for the emailing of ballots.  I really wasn’t sure how this would be received.  The early results for us, about three weeks after ballots have been mailed and emailed, is that the email is not increasing the number of votes cast.

For military voters, we emailed 42 ballots and just a single one has been returned.  A rate of 2.4%.  On the other hand, we have mailed 204 ballots, and already 33 have been returned.  A rate of 16.2%.

Now other factors certainly enter into this.  Many of these military ballots are sent within the country.  The final determination of the effectiveness of this method will be decided two weeks after the election when we do our final canvass.

But delays of three weeks or more in sending out ballots, either through the mail or email, are inexcusable.   Those justifying those delays by pointing to email delivery are relying on an unproven and untested method.

Voter Residency in Illinois

We get lots of questions from people about what constitutes “residency” for purposes of voting and for running for office.  As this is a University community, some people have concerns about students who are here temporarily but voting here.  The law is far from clear, but for students in Illinois, it is generally the case that they can declare either their campus or “home” address as their residence.

Now we have questions about the residency of Rahm Emanuel and whether he qualifies as a candidate for Mayor of Chicago.  As above, the definition of residence can be murky in Illinois law.  However, the discussion in the media has generally avoided the important factual determinations that have to be made.

In our circuit and appellate district, we had an interesting case involving residency;  People v. Baumgartner.  Since it’s a recent case, 2005, I think it gives critical guidance to this issue as it affects Emanuel.  The appellate court overruled the circuit court decision and ruled that Baumgartner did not commit perjury when he declared his Moultrie County address to be his “residence” despite owning a home in Champaign County.

Here is the paragraph in the appellate opinion that best applies to the Emanuel case.

Two elements are necessary to create a “residence” for voter registration purposes: physical presence and an intent to remain there as a permanent resident.   Delk v. Board of Election Commissioners, 112 Ill. App. 3d 735, 738, 445 N.E.2d 1232, 1235, 68 Ill. Dec. 379 (1983).   Residence is lost upon abandonment; however, “‘an absence for months, or even years, if all the while intended as a mere temporary absence for some temporary purpose, to be followed by a resumption of the former residence, will not be an abandonment.'” Stein v. County Board of School Trustees, 40 Ill. 2d 477, 480, 240 N.E.2d 668, 669 (1968), quoting Kreitz v. Behrensmeyer, 125 Ill. 141, 195, 17 N.E. 232, 253 (1888).  [*12]  “Where a person leaves his residence and goes to another place, even if it be another state, with an intention to return to his former abode, or with only a conditional intention of acquiring a new residence, he does not lose his former residence so long as his intention remains conditional.” Pope v. Board of Election Commissioners, 370 Ill. 196, 201, 18 N.E.2d 214, 216 (1938). To change residence, “there must be, both in fact and intention, an abandonment of the former residence and a new domicile acquired by actual residence, coupled with the intention to make it a permanent home.” Welsh v. Shumway, 232 Ill. 54, 77, 83 N.E. 549, 559 (1907).

As Emanuel apparently still owns his home and has a longstanding residence in Chicago, I think he’ll have no problem declaring and demonstrating that he had an intent to return.

This issue is of some importance here as professors on sabbatical or otherwise away have been known to rent out their homes while they are gone.  This does not take away their right to vote.

Preparing The Ballot This Year

A few things have changed in the election law that are making the job of preparing ballots a little easier.  At the same time, the number of referenda on the ballot is adding some complications.

Last year Congress passed legislation that requires military ballots to go out at least 45 days prior to the election.  In the past, that was a goal of ours, but often one that we couldn’t meet because of the certification date for state candidates.  Thankfully, the legislature amended the Election Code to move that certification date.  Even with the law change, we still only received the final certification last Friday.  We are still waiting for the final certification of local ballot questions as well as write in candidates.  That deadline is Thursday, September 2.

Those local referenda are especially important for us this year.  We already have a State Constitutional question on the ballot, as well as one for the County.  Of particular concern are voters in the Condit township portion of the Village of Fisher.  They will be presented with two lengthy tax cap referenda.  As it stands now, those 138 voters will have the longest ballot in the county, one which will barely fit on the front and back of a legal sheet of paper.

Across the country, election jurisdictions like ours are preparing ballots well in advance, even before what is on the ballot is finalized.  That’s why I was surprised at hearing the election director in the state of Washington suggest that preparing ballots after their September 7 certification for mailing by September 18 was not feasible.

I think it probably is, but I think it definitely has the ability to put a burden on election offices.  I know we’ve had days that we’ve worked a weekend or evening to get military ballots out in time.  In this particular instance, for Washington, that amounts to an unfunded federal mandate, facilitated by the lack of action by the Washington legislature.  At the same time, the mandate is pretty clear and states and local jurisdictions ought to be looking at ways to get this particular job done.

I’ll post when our ballots go out in the mail.

Auditor Referendum Will Not Be on November Ballot

I have determined that the County Board’s action to place a referendum on the ballot eliminating the elected County Auditor’s office does not conform to the requirements of Illinois Election law and will thus not be on the November Ballot.

This is a difficult decision because refusing to certify a question or candidate is an extraordinary remedy.  However, no other one exists.

The certification to the Election Authority (my office) of questions by public bodies is required under the law to include “the form of the public question to be placed on the ballot, the date on which the public question was initiated by either the filing of a petition or the adoption of a resolution or ordinance by a governing body,” (10 ILCS 5/28-5).  That the certification is statutorily required to contain these elements strongly suggests that the Election Authority is to examine the certification to ensure that the elements are present and that they are in apparent conformity with the relevant statutes.  The Election Code requires that public questions be passed by governing bodies no more than a year before the election.  It is reasonable to assume that it is my statutory obligation to make sure that every filing of a public question by a governing body conforms to this requirement.

It is important to note that there would be no other statutory means to prevent the placement of a referendum on the ballot that was passed illegally by a governing body.  Unlike public questions resulting from petition drives, there is no mechanism for citizens to challenge the validity of a referendum passed by resolution of a governing body.  In fact, the Election Authority is the sole individual in a position to prevent an illegal question from appearing on the ballot.

I have written in the past about the principle of apparent conformity.  This decision I am making today comports well with what I have written in the past.  Of particular interest here would be People ex rel Giese v. Dillon in 1914.  (266 Ill. 272).

“The town clerk is a ministerial officer. It was his duty to examine the petition, and if upon its face it appeared that it was not in compliance with the requirements of the statute he was not required to submit the question for a vote at the ensuing election.”

The Dillon case is cited in virtually every case regarding apparent conformity.  In the 2008 Haymore decision, the Appellate Court once again reaffirmed that Dillon is “sensible and relevant.”
In the case at hand, the resolution of the County Board was passed on August 20, 2009, over 14 months before the November 2, 2010 election.  The Election Code is clear however that no question can be submitted more than a year before the election.

“However, no petition, resolution or ordinance initiating the submission of a public question, other than a legislative resolution initiating an amendment to the Constitution, may specify such submission at an election more than one year, or 15 months in the case of a back door referendum as defined in subsection (f), after the date on which it is filed or adopted, as the case may be. A petition, resolution or ordinance initiating a public question which specifies a particular election at which the question is to be submitted shall be so limited, and shall not be valid as to any other election, other than an emergency referendum ordered pursuant to Section 2A-1.4.  (10 ILCS 5/28-2)

As I noted above, refusing to certify a question is an extraordinary remedy that I don’t take lightly.  I have never done it before and hope it never happens again.  I also want to make it clear that I take this action as the Election Authority for Champaign County and not as Clerk of the County Board.

More Outcome Oriented Research

I can’t say I know much about race relations in Boston.  But I know a little about election administration.  Enough to know that the draft paper by Profs. Rachael Cobb, James Greiner, and Kevin Quinn,  is long on conjecture and short on facts.  Perhaps election judges in Boston are a bunch of racists.  One thing for certain, we’ll never know from this research paper.

The researchers determined that voter ID laws were not administered in a race neutral manner in the November 2008 Presidential election in Boston.  Unfortunately, their research fails in a fundamental area.  Additionally, this research falls into the all-too-common category of projects that miss an opportunity to base their findings on facts rather than statistical analysis.

First, these researchers leads you to believe, and the synopsis from Pew in their electionline report two weeks ago directly states as much, that the only reason a person in Massachusetts needs to show ID is if they are a first time voter under the Help America Vote Act who failed to provide ID at the time of registration.  That’s just not true.  A couple questions to election administrators in Massachusetts and a review of the Secretary of State’s website could have told these researchers as much.

Here is the link to the 2008 election guide sent out by the Massachusetts Secretary of State’s office.  On page 5, in the Massachusetts’ Voters’ Bill of Rights under item 8.

You have the right to vote but must show identification if: you are a first-time voter who registered to vote by mail and did not submit identification with the voter registration form; or your name is on the inactive voter list; or your vote is being challenged; or if requested by a poll worker.

And again on page 3 of the same guide.

To cast a provisional ballot, you must execute a provisional ballot affirmation before a precinct officer at the polling place declaring that you are a registered voter in the city or town and reside within the geographical boundaries of said precinct. You must also show suitable identification.

Here in Champaign County, we also have comparable regulations for ID.  One example of people who end up having to show ID on election day are those whose signature doesn’t match the voter registration record.  The most common reason for that is a person’s signature changing.  In the November 2008 election we literally had hundreds of people showing up to vote who hadn’t voted in over a decade.  Some had been registered back in the 70s but still had never cast a ballot.  It is hardly surprising that those people’s signatures had changed and that they were subsequently asked to show identification.  I would not be surprised to find that the same thing happened in Boston.  A number of other things can happen to cause a signature to not show up properly on the voter’s poll record requiring the judge to ask for identification.

But as glaring as the faulty legal foundations of this research is the methodology.  If you read the research, you’ll see a lot of fancy statistical analysis that purportedly makes the case for this trio.  But one wonders why it was necessary to do much in statistical analysis at all.  These researchers could very simply have shown up at the polling place and asked to observe.  This is from the Secretary of State’s Election Day Legal Summary.


To achieve the legal requirement that the election be held in public view, observers shall be allowed inside the polling place,  outside the guardrail, unless they are disorderly or obstruct the  access of voters.  Observers may keep notes including marked  voting lists.  If there are so many observers in the polling place  that they obstruct voters, they may be asked to cooperate in collecting information.  The warden may exclude from the  polling place any person who is disorderly or who obstructs  the access of voters.

This research surveys every 16th voter regarding whether they were asked for ID, presumes that the respondent is answering accurately, and then further presumes that they were either asked properly under HAVA or they were asked improperly.

By placing observers in the polling place, these researchers could have gotten far more accurate information that would have obviated surveys and statistical sampling entirely.  Each and every voter in a number of polling places could have been observed to determine their apparent race, whether they were asked for ID, and the reason they were asked.  Actual observation could have avoided all the accuracy problems associated with surveys.

I’ve got my doubts about the value of voter ID laws.  But those doubts have nothing to do with the ability of election judges to administer them fairly.  This research is faulty and its conclusions unsupported.  If opponents of voter ID laws are truly concerned about this issue, they should go back into the field in November 2010 with actual observation of the election process that does justice to the question.  Further, charges of racism against election officials should not be launched with such sketchy evidence.

The only thing this paper accomplishes is further diminishing the confidence of voters in the election process.  That’s very unfortunate, and unfortunately predictable in these days of agenda driven research.

146 Illegal Votes Identified in November 2008

I am often asked about the possibility of illegal votes cast in Champaign County.  The questions are usually about either the possibility of double voting or the possibility of people voting here who are living somewhere else.

We’ve never had the resources to do the exhaustive study of some of these issues that we’d like.  However, this last semester, we had an intern in our office who dived into the project of determining whether some people in Champaign County voted illegally.

Our research identified 146 illegal voters.  The research is available here and the actual data here. The research paper itself puts the issue in the context of the controversy surrounding so called caging.  Caging is an attempt to identify voters who no longer reside at an address by sending them nonforwardable mail and then challenging their right to vote on Election Day.

In the time shortly before and shortly after the November 2008 election we had 653 pieces of mail sent to voters in the November 2008 election that were returned by the post office as undeliverable for some reason.  We then wrote the property owner of the voter’s claimed residence and asked whether that person resided at the address on Election Day November 5, 2008.  In 146 of those instances, the voter did not live at the address that they claimed on their application to vote and their voter registration yet they still voted a full ballot.

Without further research, it is impossible to know if these are voters who moved within Champaign County and should have voted a Federal Only Ballot or whether these voters had left the county entirely.  In all but two instances, the voter’s identification (either Driver’s License or Social Security number) was validated on the State Board of Elections statewide database.  None of the 146 voters are shown to have voted in another jurisdiction.

Much more research needs to be done yet on this project, including determining why so much mail gets returned when in fact the individuals in question actually reside at the residence they claim.  Of course, the research was conducted only for voters for whom mail was returned.  Most voters never received a mailing from our office after August 1, 2008.

As I write in the paper, the key element here from my perspective is that voters who tell the truth vote a federal only ballot.  Voters who are willing to lie about their residence, are rewarded with a full ballot.  In the November 2008 election, that meant they were able to vote on such things as County Board (including a race decided by fewer than 200 votes ) and the local sales tax referendum, which lost by just 246 votes.

There’s much more detail, thoughts, as well as possible solutions to this issue in the research paper.

Further, while I realize that this is not the highest priority for law enforcement, I have turned this information over to an investigative authority.

Seating a Presumptive Winner

Much is being made of the possible special election that a federal court may force upon Illinois to fill the vacancy created by the Presidential election of Barack Obama.   I wrote last week about how that could be done for little to no cost.

However, another problem raised by opponents of the special election needs to be addressed,  the short period of time in which the Senator would serve.  In determining the length of that time, people are using the final canvass of votes as the determinant for when the new Senator would take office.

There is precedent though for the presumptive winner of  a special election to be sworn in before the final canvass is conducted.  In 2008, Bill Foster was elected on March 8 to fill the vacancy created by the retirement of Denny Hastert.  He was sworn in just 3 days later, despite the official canvass not happening for weeks afterward.

The minutes of the March 7, 2008 State Board of Elections meeting give some idea of the reasoning.

Discussion then moved to the request from the Clerk of the U.S. House of Representatives for unofficial election results of the March 8 special election. Member Brady indicated that he spoke with the Clerk’s office and it is their practice to evaluate unofficial results. If it is clear from the unofficial results who the victor is they would act immediately and seat the member. However, if it is a close election and they were unable to determine the results the Clerk’s office would not act on unofficial results.

If the same process were to be followed in the Senate, a special election winner in November could be seated by the end of election week.

DOJ NVRA Memo Seeks to Gut HAVA Fraud Protections

In December of last year and March of this year, I wrote about the suggestions from many that the National Voter Registration Act (NVRA), passed in 1993, disallows removals from the voter file based on a match within the statewide voter registration file created by the Help America Vote Act (HAVA), passed in 2002.  It’s a weak argument, that to date has been pushed mostly by voter registration reform groups.  Now, however, the United States Department of Justice is jumping on this bandwagon as detailed in this undated and unsigned memo that appears on their website (to which I was pointed by the electionline newsletter in early June).  Adherence to the shaky legal interpretations of the memo would eviscerate any fraud protection passed by Congress in 2002.

The first thing that should be mentioned about the latest DOJ memo is the last item in the memo.

46. How can I contact the Department of Justice about the NVRA’s voter registration requirements?
As a general matter, the Department of Justice does not issue advisory opinions concerning the statutes that it enforces.

The 45 items before this, however, are nothing but an advisory opinion.  As an advisory opinion though, it is short on analysis and hardly fit for any legal department, especially the chief law enforcement agency in our country.

It is hard to avoid the notion that DOJ is offering up a shallow legal opinion to prop up the overall agenda of many people, which is to worry less about the integrity of our voter registration rolls and more about making sure that everyone is on them, no matter how many times.

The list maintenance section of the DOJ memo starts at number 26.  Sections 30, 31, 34, 35 and 36 are all flawed and largely ignore the list maintenance provisions in the Help America to Vote Act.  For example, item 31 of the DOJ memo states that new applications in another jurisdiction only trigger a removal in the former jurisdiction if the registrant provides information about the fact of registration in the former jurisdiction.  That’s not what happens in practice, nor was it what was intended with HAVA.  In fact, the statewide voter registration file was established precisely to eliminate the problem of voters not informing the former jurisdiction of their address change.  The 18 page DOJ memo mentions the HAVA required statewide voter file just once and that is only regarding deaths and felony convictions.  It is as if Congress had done nothing.

However, over the last five years, DOJ has entered into a number of memoranda of understanding regarding list maintenance.  I detailed some of them in a post in December.  As an example, here is the procedure for removing duplicate registrations in California.

§20108.60. Duplicate Registration Records. (a) Beginning January 1, 2006, the Secretary of State shall conduct weekly checks within the Calvoter statewide registration list to identify potential duplicate registrations for the same voter within that list, based on established rotating criteria. Upon identification of potential duplicate registration records, the Secretary of State shall automatically send an electronic notice to the county with the record that has the oldest date of registration.
(b) Within five (5) business days of receipt of a notice of potential duplicate registration the elections official shall take all necessary steps to determine whether or not the registration record is a duplicate of an existing newer registration, and if a duplicate registration is confirmed, shall cancel the older duplicate registration and submit a registration update file or full load file to Calvoter in accordance with Section 20108.15 and Section 20108.40.

If list maintenance procedures for election officials are to be based solely on NVRA, there was essentially no reason for Congress to mandate the statewide voter file.  I doubt that a single state administers their statewide voter file in a way that meets the standards laid out by DOJ in this recent memo.  Following DOJ’s guidelines would raise costs dramatically and probably inflate the voter rolls by millions.  In Illinois, over 200,000 voters have been removed through the statewide voter file.  Without the statewide voter file, those nonvoters would still be on the voter rolls at a cost to taxpayers and with a potential risk for fraud.