Real Election Reforms Are Possible, But Not With Election Deniers Running the Legislature

There have been A LOT of terrible bills dropped by AZGOP state legislators this session. Probably none of them will make it past COW in the Senate: State Senator Paul Boyer has made it very clear to his colleagues that he is not on board with any nonsense stemming from conspiracy theories nor the misinformation campaign that is the #AZFraudit by the Senate and Cyber Ninjas. So, despite much sound and fury (and rightfully so, as many of these bills are truly stupid and/or dangerous), probably none will actually make it to Ducey’s desk this session.

But there is room for improvements in Arizona’s elections procedures that could help head off future misinformation campaigns and conspiracy theories. Reforms and transparency measures that election reform advocates have been making for years just might have a chance to make it over the top with bi-partisan support if common sense and expertise can overcome over-heated rhetoric, blatant election lies, and conspiracy theories. That’s a big “IF”.

The former mouthpiece of the CyberNinja #Fraudit, former AZ SecState and AZ Senator Ken Bennett, has an uphill climb to gain any credibility in suggesting any real reforms, especially among Democrats. But he is suggesting some reforms that could be relatively benign if properly designed.

AZCentral reports: “Bennett wants a quick way for residents to verify results on their own. He and John Brakey, an activist for more transparent elections who founded Audit USA, met with many state lawmakers in December to pitch their ideas.”

Their plan is pretty simple and has been circulating in election integrity and transparency circles for years, long before the 2020 election:

“Before an election, counties in Arizona would publicly release the name and address of every registered voter. This is public information, but currently people, companies and political groups must request and pay for the information.

After the election, and within a certain time frame before the votes were finalized by the county and state, counties would be required to publicly release the name and address of every voter who voted. Separately, and not associating the data with a particular voter, they would release a record of how votes were tabulated on each ballot and an image of each ballot.

To try to verify that only registered voters cast ballots, people could compare the two voter lists.

People also could match the image of a ballot with a record of how the ballot was tabulated, to see whether the machine counted the votes accurately.”

These are, on the whole, pretty reasonable reforms.

I do have some concerns that are widely shared in the election integrity community. The first problem is that it is impossible for some ballots (those cast by certain electronic means for the disabled and overseas military, for instance, but including other types) to be scanned and released in this manner, so such a records release of scanned ballots would never be a complete record of all the votes validly cast, potentially simply giving more ammunition to election deniers. While it is certainly unlikely that those records that could not be included in such a release would make an appreciable difference to most elections, it is conceivable they could do so in a very tight race.

Of course, such records don’t by themselves, nor in combination, actually enable the public to fully audit an election – that is an impossible and likely a futile goal, as it requires more access, expertise, and experience to fully audit any election than can be given safely to the entire public.

I cautiously support, with many safeguards and caveats, the release of the Cast Vote Record (CVR, the database of ballots cast) and the actual ballot images. Neither data set has a significant possibility of revealing a voter’s data nor how they voted (but see “Preserving Anonymity of Cast Vote Record” a report prepared in conjunction with Colorado’s risk-limiting audit process, for examples of how a CVR might expose voter preferences). While these records would, as previously stated, not enable the general public to fully audit an election, they can be compared to voter registration totals and turnout estimates to ensure that no major monkeying about took place, i.e. a risk-limiting audit. These measures do make it harder for groups with malign intent to allege major election fraud has taken place. Because of Trump’s #BigLie of 2020, fighting misinformation efforts about our elections, whether domestic or foreign in origin, is now a valid purpose of reforms.

There are bills pending that would make even this step hazardous to our well-founded tradition of a secret ballot, however. The hideously stupid SB1028, introduced by the hideously insane AZ Sen. #WackoWendy Rogers, for instance, if combined with the release of ballot images, could allow correlation of individual voters and their cast ballot by requiring every ballot to have a unique identifying mark: a terrible and useless idea that can only result in dangerous outcomes, adding nothing to election integrity or security.

I do have concerns about the release of personally identifying information about Arizona voters as well. Names and addresses of voters may be public records already, but actually obtaining them requires an investment of resources that your average crackpot is not going to make.

Giving that information out for free to the public poses a security risk that I don’t believe is justified by any additional quanta of election integrity. In fact, I can certainly foresee the possibility of such information, freely available, being misused to harass and intimidate voters before, during, and after an election. Certain groups might easily use that information to target certain areas, particular voters, and voting locations. It’s not as if that can’t be done now, but making the information freely available has the side-effect of enabling the crowdsourcing of such tactics, whereas now it would require a great deal more planning, resources, and organization; which is why you’ve really only seen organizations such as the Republican Party and the KKK doing it in the past.

There is a bill pending in the AZ Lege now that would enable half of this system: SB 1119, sponsored by Senator Borrelli (who is a notorious liar that the 2020 election was stolen). Its provenance is not, in itself, a reason to reject the bill, but it is a good reason to closely examine it. It has some major problems: it does not specify any deadline for the release of the records, which could be used to delay the release for any amount of time, and; it doesn’t allow for redactions to protect personally identifying information on the ballots (some ballots can by voter action or the nature of the races in a precinct reveal voter information).

I’m not aware of any bills now pending that would require publication of the Cast Vote Record. Of course, I might just be ignorant of such a pending bill, if so, please leave a comment with the bill number if you are aware of one.

Maricopa County Recorder Stephen Richer has also made some very reasonable recommendations for elections procedure reforms. Of course, his integrity and honesty in defense of the legitimacy and accuracy of the 2020 election have put him in bad odor with the current crop of muppets and stooges running the AZGOP, so his suggestions may fall on deaf ears. AZCentral reports them thusly:

• Make ballot images public, or make them available to a limited group of “government actors,” such as the majority and minority leaders in the state Legislature. What happens now: The county currently has ballot images, but they are not public records.
• Require the Secretary of State’s Office to enlist an “EAC-certified, professional, election auditing company” to perform a post-election audit of Maricopa County and one other county selected at random and announced at least six months before the election, to be complete before the official canvass of the election. The audit would check the accuracy of the vote-counting machines and results. The U.S. Election Assistance Commission, or EAC, accredits companies to certify voting machines before they are used but does not certify auditors. What happens now: State law currently requires counties to perform logic and accuracy tests on machines before and after elections to test their effectiveness.
• Require a post-election audit of duplicated ballots similar to the hand-count audit of ballots already required in state law. Counties duplicate ballots when the original ballot cannot be read by a vote-counting machine, such as when it is a Braille ballot or when the ballot has been damaged. What happens now:  State law currently requires counties to have political parties conduct a hand-count of a statistically significant random sample of ballots after each election
• Authorize and fund the Secretary of State’s Office to enlist a “reputable, professional, EAC-certified external auditor” to perform an annual review of a small percentage of the statewide voter registration database. The information would be forwarded to counties, and the third party wouldn’t have the ability to modify voter records. What happens now: Maricopa County uses multiple federal and state databases to verify eligible voters on a regular basis.
• Require all Arizona counties to participate in the Electronic Registration Information Center, a national database that allows counties and states to compare voter registration records. What happens now: Maricopa County voluntarily participates in this, but some counties do not.

The first suggestion, making the ballot images public, has already been discussed. I tend to think that making them available to all the political parties and/or the candidates would be more useful, on balance, than either making them fully public or limiting them to certain officials. But I think this sort of suggestion is the basis for further productive discussions.

Richer’s second suggestion of a qualified post-election audit of the equipment is reasonable and measured.

I don’t fully understand Richer’s third suggestion, I must admit. The idea of auditing duplicated ballots has some merit, although multi-party accuracy checks could simply be integrated into the duplication process. But the mention of the statutory required statistical, risk-limiting hand count audit seems to be a non-sequitur, as that has nothing to do with the duplicate ballots, but is merely a means of ensuring that a random statistical sample of ballots does not vary from the overall election results outside a specified margin of variance.

The fourth suggestion of a periodic statistical audit of the voter registration database seems to overlap with the fifth suggestion to require all counties to participate in the ERIC system. Both are fine suggestions that would ensure the integrity of our registration databases, and flag the most common source of actual voter fraud: voting in multiple states. There have been a handful of such cases, almost invariably Republicans responding to Trump’s incitement to vote multiple times. The latter suggestion would help investigate and prosecute such crimes.

In short, all of these suggestions by Bennett, Brakey, and Richer are productive and useful bases for starting a discussion of how to actually make our elections systems more secure, reliable, and transparent.

If the GOP ever shakes off their obsession with the lie that the 2020 election was fraudulent and admits that the election was free of any substantial problems that could cast doubt on the outcome, we might have in these proposals a basis for a bipartisan agreement about how to proceed in enhancing our elections’ security. Not because there is a major problem with our elections now, but because we genuinely now have a need to combat the ability of malign actors to spread misinformation and lies about our elections.