Pursuant to its Notice of Public Hearing, I write in opposition to Rule Set 2021-61 ST. Please acknowledge receipt of this letter, submitted for the record.
Rule Set 2021-61 ST, Signature Matching for Absent Voter Ballot Applications and Absent Voter Ballot Envelopes (the “Signature Match Rule”), is a transparent attempt to render meaningless and effectively nullify the signature match requirement described in MCL 168.759, 168.767, and 168.932a. As such, it breaches the authority cited in the Draft Rule Language.
Furthermore, the limited application of the rule only to absentee voting – as opposed to all electoral processes where a signature is required, such as a ballot proposal or petition for candidacy – illustrates the bad faith in which the rule set was proposed.
The signature match requirement described in statute exists for exactly one purpose: to discern whether the ballot was lawfully requested or returned by the qualified voter to whom it issued, or was criminally requested or returned by another individual: a felony. Impersonating a voter to cast a ballot is a crime for which there is no plausible mitigating factor: One cannot, by accident or neglect, sign the name of another person.
The signature match requirement exists solely to prevent intentional election fraud.
Yet the proposed Signature Match Rule not only removes any meaningful safeguard, it actually provides criminals a road map: a fool-proof formula for forgery.
According the proposed rule, signature verification begins “with the presumption that the voter’s signature is his or her genuine, valid signature.” Notably, this introductory presumption itself presumes all signatures’ validity by referring to them as “the voter’s signature” (rather than “the submitted signature”, “the signature on the envelope”, etc.)
The rule continues to say “A [presumed] voter’s signature should be considered invalid only if it differs in multiple, significant, and obvious respects from the signature on file. Slight dissimilarities should be resolved in favor of the voter. Exact matches are not required to determine that a signature agrees sufficiently with the signature on file.”
The first two sentences establish an extraordinarily lenient standard as the general rule. “Multiple”, “significant”, and “obvious” are undefined: If a voter’s signature on file is cursive, but the signature submitted is printed, is that “multiple” obvious differences, or just one?
The third sentence is nonsensical because “exact matches” is undefined and nobody expects anyone to be able to create an “exact match” (perfectly identical) signature to the signature on file.
Next, the rule would establish that if there are ANY “redeeming qualities” in the submitted signature, the signature MUST be treated as valid. In other words, even if a signature differs from the signature on file in “multiple”, “significant”, and “obvious” ways, it MUST be accepted if ANY redeeming quality exists – regardless of whether the one redeeming quality meaningfully explains and excuses the “multiple”, “significant”, and “obvious” deficiencies in the signature.
Notably, examples of “redeeming qualities” are provided, but qualities are “not limited to” those examples. In other words, any adjudicator could invent any “redeeming quality” as a pretext to accept an otherwise invalid signature.
Among the enumerated “redeeming qualities” is if signature features “do not match because it appears as if the voter’s hand is trembling or shaking.” This overly broad catch-all provides a road map for perpetrators of election fraud.
All one must do to forge a signature is simulate trembling.
If the language instead said “do not match ONLY because it appears the voter’s hand is trembling or shaking” – i.e. the general appearance and style of the signature, after trembling is accounted for, is still similar – it would at least allow for discretion in a complexity of factors. Even then, it is improper for a single “redeeming quality” to nullify other multiple, obvious, and significant deficiencies. But because the language is so broad as to allow an appearance of trembling or shaking to overrule any and all factors to the contrary, it effectively nullifies and renders meaningless any safeguards on signature validity. It provides a road map for election fraud.
The expressly accepted “redeeming features” also include “similar distinctive flourishes.” Again, like all other enumerated “redeeming features”, the proposed rule fails to consider this in context with discrepancies and would simply allow it to overrule multiple, significant, and obvious discrepancies.
These obvious and severe security hazards were apparently not enough for the Michigan Department of State, which continued to provide EVEN MORE options for a signature with multiple, significant, and obvious deficiencies to be accepted.
The most absurd: the election official should accept a REVERSAL of first and last name. This is an obvious hallmark of fraud. No person ordinarily reverses his or her first and last name.
The rule also instructs election officials to consider that the signature “may have been written in haste.” In other words, any scribble is acceptable.
The rule provides an additional avenue for forging signatures: election officials should consider “the surface of the location where the signature was made may have been rough, soft, uneven, or unstable.” This is an additional excuse for processing patently invalid signatures.
Finally, the rule allows for consideration of “factors applicable to a particular voter, such as the age of the voter, the age of the signature or signatures contained in the voter’s record, the possibility that the voter is disabled, the voter’s primary language, and the quality of any digitized signature or signatures contained in the voter’s record, and any other plausible reason given by the voter that satisfies the clerk when following up on a questionable signature.“
Nowhere in the rule is it explained how these factors would explain a signature mismatch, but the inclusion of “the age of the signature or signatures contained in the voter’s record” constitutes yet another catch-all that would render the signature verification process meaningless for an unspecified number of voters. Consideration of a voter’s age provides another open-ended pretext for accepting a signature, no matter how numerous, significant, and obvious the submitted signature differs from the voter’s actual signature.
-All signatures are presumed valid.
-Signatures with differences that aren’t “multiple”, “significant”, AND “obvious” are valid.
-Signatures with all of those deficiencies may still be accepted if ONE “redeeming factor” is determined, whether or not it negates the multiple, significant, and obvious deficiencies.
-The one redeeming factor can be anything – the definition is open-ended.
-Trembling and shaking is a specific excuse, as is “similar distinctive flourishes.” Any single “redeeming” factor overrules all deficiencies.
-Obvious evidence of fraud, such as reversal of first and last name, should be ignored.
-Any discrepancy can be explained by any other explanation. The rule even allows for ignoring the signature contained in the voter’s record based on an unspecified “age” of the signature.
The inescapable conclusion is that the proposed Signature Match Rule provides no safeguards whatsoever against forgery and election fraud.
This stands in stark contrast to the standard accepted practice at banks, which employ signature matching technology on a daily basis to prevent against fraud. These signature matching systems use mature technology to assess static and dynamic characteristics of a signature to verify its likelihood of authenticity: the start and end of strokes, placement of crossings, curves, and loops, even signals of time and pressure. These standards are highly effective at discerning fraud.
There is no legitimate reason why the same standard employed regularly across America cannot be employed in Michigan elections – unless the intention is to facilitate absentee ballot fraud.
And this is clearly intentional, because the rule was proposed to apply only to apply to absentee ballots and their applications, but not any other document where a voter affixes his or her signatures. These include nominating petitions, petitions for ballot proposals, and more.
If the proposed rule was made in good faith, the signature standard would be uniform across all election-related documents, not just absentee ballots.
For all of the reasons above, the proposed Signature Match Rule should be abandoned. Its deficiencies are incurable. It intentionally sabotages the only protection against election fraud in statute that ensures the person requesting and returning an absentee ballot is the actual voter to whom it is issued. It does this in violation of Michigan Election Law.