Last week, a bipartisan group of 16 senators unveiled a draft bill to reform the Electoral Count Act of 1887. My colleague Andy Craig has already thoroughly summarized what the bill would do, explaining both its strengths and areas for improvement. I agree entirely with Andy’s bottom-line take on the bill: it is a solid effort and would represent a vast improvement over the current law. With that understanding, I will focus in this post on three aspects of the bill where I think the text could be further improved to remove ambiguity or solve additional problems. These three aspects of the bill are: the rules for a governor’s certification of the state results, the rules for expedited judicial review of that certification, and the rules for counting the electoral votes in Congress.

The Governor’s Certification of the Results

The draft bill, called the Electoral Count Reform Act (ECRA), sets a deadline for a state to certify the winner of its presidential election. Section 5 of the ECRA says that at least six days before the electoral college meets, the governor (or another state official designated in advance by state law) must issue a “certificate of ascertainment of appointment of electors.” For ease of discussion, I’ll refer to this certificate as simply the “governor’s certificate,” and I’ll refer to the person who issues it as simply “the governor,” even though a state may designate some other official to issue the certificate instead.

The ECRA would require that the governor’s certificate be issued “under and in pursuance of the laws of such State providing for such appointment and ascertainment enacted prior to election day.” In other words, a governor can’t ignore state law when he issues the governor’s certificate. If state law says the person certified by the state Election Board is the winner, the governor may not issue a governor’s certificate contradicting the Election Board. And the state can’t quickly pass a new law giving the governor the authority to ignore the Election Board, because the governor must follow the state law “enacted prior to election day.”

Unless a subsequent court order changes things (a possibility I’ll discuss below), this governor’s certificate carries heavy weight. The ECRA says that for the purposes of section 15 (which defines the procedure for counting the votes in Congress), the governor’s certificate “shall be treated as conclusive with respect to the determination of electors appointed by the state.” And later on, in section 15, the ECRA says that “only the votes of electors who have been appointed under a [governor’s certificate] issued pursuant to section 5” may be counted (with an exception, not relevant here, for electors appointed to fill a vacancy).

This raises the first ambiguity. What if a governor, through either malice or incompetence, fails to meet the statutory deadline to issue the governor’s certificate? If a governor is even one day late, the governor’s certificate has arguably not been issued “pursuant to section 5.” That is to say, the certificate has not been issued in a way that meets all the requirements of section 5, because it did not meet the deadline set in section 5. And section 15 seems categorical: if a governor’s certificate has not been “issued pursuant to section 5,” then electors from that state may not be counted.

This creates the possibility of intentional mischief by a “rogue governor.” If a state votes for a presidential candidate that the governor opposes, a rogue governor can seemingly disenfranchise his own state by slow-walking the certification and missing the deadline. And while this strategy wouldn’t succeed in switching a state’s electoral votes to the governor’s preferred candidate, it would still help that preferred candidate by erasing electoral votes that otherwise would have (rightly) gone to his opponent.

In my view, the ECRA could be improved by clarifying that a state’s electoral votes may be counted if confirmed by a judicial order issued after the governor’s deadline, even if the governor misses that deadline.

In addition, there is similar ambiguity over what happens if a governor issues a certificate by the deadline but does not follow state law. For example, what if the state law makes clear that a governor must follow the rulings of the state Election Board, but a governor ignores that requirement and certifies someone else as the winner instead? In other words, what happens if a governor’s certificate is not issued “under and in pursuance of the laws of such State,” as section 5 requires? To fully understand when this scenario might lead to uncertain outcomes, it is necessary to first move on to discuss the next phase of the bill, the judicial review procedures. I will return to discussing this problem after laying out the judicial review and counting rules in more detail.

Judicial Review

The ECRA clearly contemplates that a governor’s certificate would be subject to judicial review. The bill states that “any [governor’s certificate] as required to be revised by any subsequent State or Federal judicial relief granted prior to the date of the meeting of electors shall replace and supersede any other certificates submitted pursuant to this section.”

For the most part, this provision is a great improvement on the current ECA. In the unlikely event that a “rogue governor” does ignore state law and certify the wrong winner, this provision makes clear that a court-ordered correction to that certificate would take precedence. In the event of such a blatant instance of ignoring the law, it is likely that a court order overruling the governor would indeed come quickly (whether in state or federal court).

Still, there are two potential weaknesses to this provision as currently drafted. First, as noted above, the provision contemplates an erroneous governor’s certificate being revised by court order, but it doesn’t explicitly contemplate a court remedy for the failure to issue a governor’s certificate altogether. This could be fixed by adding one more sentence to the end of this provision, to the effect of: “If no certificate of ascertainment of appointment of electors has been issued by the date that is 6 days before the time fixed for the meeting of the electors, then any certificate of ascertainment of appointment of electors as required to be issued by State or Federal judicial relief granted prior to the date of the meeting of electors shall be treated as the certificate of ascertainment of appointment of electors issued pursuant to this section.”

The second potential weakness of this provision is the precise wording of the “superseding” clause. It is not the court’s order itself that supersedes a governor’s erroneous certificate. Rather, it is a new governor’s certificate “as required to be revised” by a court order.

Why might this distinction matter? Because, in the unlikely event that a governor goes rogue, the law should consider what might happen if a governor attempts not to comply with a court order. If a governor simply ignores a court’s order to issue a “revised” certificate (even upon threat of contempt of court) until January 6, does that mean that there is no revised certificate to supersede the erroneous original one?

In my view, it would be preferable to not even give governors an opening to test the limits of this strategy. One option, which my colleague Andy Craig suggests in his comprehensive proposal for ECA reform, is to make explicit that a revised certificate issued by any state official upon court order would suffice, not just a revised certificate issued by the governor. The odds of every state official refusing to comply with a court order are even slimmer than the odds of the governor refusing to comply, and so this would likely eliminate state resistance to a court order as a feasible strategy.

A second option would be to simply skip the requirement of a state official’s compliance altogether, making explicit that a judicial order to revise a certificate shall be treated as if the order itself were the revised certificate. I would prefer such an amendment, as it would remove any lingering doubt as to whether refusal to comply with a court’s order might be a successful strategy.

One potential objection to such an amendment is that it would be unconstitutional to treat a court order as if it were a state’s certificate. The Constitution mandates that “[e]ach State shall appoint” its electors (emphasis added), and this could be interpreted to mean that some formal act of appointment must come from a state official. Ramesh Ponnuru has taken this view, praising the draft ECRA for not empowering the courts to directly certify electors.

I lean toward the opposite view—that so long as a court is merely applying state and federal law to determine who in fact won the state, the state has appointed its own electors in every meaningful sense. After all, when a governor acts on direct order of a court to certify a particular candidate as the winner, the court has for all intents and purposes made the final judgment as to who has won the state. It would be strange if the Constitution required the formality of a state official acceding to a judicial order, when that judicial order itself leaves no room for the state official to exercise any discretion.

The best solution, then, might be to add both of these options to the ECRA, so that courts can enjoin any state official to issue a revised certificate, but the order itself would be treated as the certificate in the extremely unlikely event that no state official is willing to comply.

The next potential issue lies in the provision for “expedited” judicial review. The draft ECRA states that “[a]ny action brought by an aggrieved candidate for President or Vice President that arises under the Constitution or laws of the United States with respect to the issuance of the [governor’s certificate]” shall be subject to special rules, most notably an initial hearing before a 3‑judge panel and then direct appeal to the Supreme Court. The ECRA stresses that in such an action it would be “the duty of the court to expedite to the greatest possible extent the disposition of the action[.]”

The clear intent of this provision is to create a speedy option in federal court to challenge a governor’s erroneous certificate. Since there could potentially be as few as six days between the issuance of the governor’s certificate and the day the electors meet, a legal challenge to the certificate would need to be handled with extraordinary speed.

But there are two potential issues that could cause this provision not to function as intended. First, it seems clear that this provision does not establish a new “cause of action.” That means this provision does not create any new right to bring a lawsuit that would not otherwise exist without the provision. As the ECRA itself emphasizes, the provision “shall be construed solely to establish venue and expedited procedures” in any action that meets the criteria laid out. The clear implication is that the provision does not create the right to bring such an action in the first place (an interpretation supported by statements from some of the bill’s sponsors in the Senate). As Bob Bauer and Jack Goldsmith rightly put it, such a lawsuit must be “brought under extant law.”

Now, I believe it is likely that some cause of action already exists under federal law to challenge a governor’s failure to issue an accurate certificate. That cause of action might lie under the Fourteenth Amendment’s guarantees of due process or equal protection, for example. And so long as such a federal cause of action does indeed exist, then the expedited procedures may be used as they are clearly intended.

Nonetheless, it is odd that the ECRA does not remove all doubt and explicitly establish a limited federal cause of action to challenge a governor’s erroneous certificate (or failure to issue a certificate). If such a cause of action already exists, then explicitly establishing it in the ECRA would be nothing more than a “belt and suspenders” guarantee, a superfluous assurance removing all doubt.

But in the event that such a cause of action does not already exist, then establishing such a cause of action is necessary for the expedited procedures to ever be used as intended. In other words, if in fact an aggrieved candidate does not have a cause of action to challenge an erroneous certificate in federal court under current law, then the expedited procedure cannot serve its intended purpose of providing a speedy resolution for such a challenge. Aggrieved candidates would simply have to challenge erroneous certificates in state court, relying on whatever state court procedures are available. If that is the case, the statutory solution is the same: establishing a narrow cause of action in the ECRA itself.

The second potential flaw in the expedited judicial procedure provision is in some ways the opposite of the first—that it may unintentionally apply to more actions than just post-issuance challenges. The draft ECRA limits the special expedited procedures to actions brought by aggrieved candidates “with respect to the issuance of the [governor’s certificate]” (emphasis added). Based on the structure of the ECRA, it seems clear that the intent is to limit the expedited procedures to only the situation when time is most of the essence: the period between a potentially erroneous certification and the meeting of the electors. But if that is indeed the intent, the choice of the term “with respect to” is not the best language to implement that intent.

The Supreme Court has explained that “[u]se of the word ‘respecting’ in a legal context generally has a broadening effect, ensuring that the scope of a provision covers not only its subject but also matters relating to that subject.” For example, the Court has held that

a statement is “respecting” a debtor’s financial condition if it has a direct relation to or impact on the debtor’s overall financial status. A single asset has a direct relation to and impact on aggregate financial condition, so a statement about a single asset bears on a debtor’s overall financial condition and can help indicate whether a debtor is solvent or insolvent, able to repay a given debt or not. Naturally, then, a statement about a single asset can be a “statement respecting the debtor’s financial condition.”

If courts were to apply this same canon of interpretation to the text of the ECRA as currently drafted, they would likely find that nearly any election-law challenge brought by an “aggrieved” presidential candidate is “with respect to” the issuance of a governor’s certificate, since nearly any legal dispute related to the election could affect which candidate is ultimately certified as the winner. In other words, the use of the broad phrase “with respect to” may unintentionally lead to the expedited procedures being invoked not just for legal challenges after a certificate is issued, but also for the mine run of presidential election litigation that occurs before the certificate is issued.

Assuming this is not Congress’s intent, there is a straightforward fix. As my colleague Andy Craig suggests, the text could further limit the expedited procedures to apply only to actions brought either after the governor has issued a certificate or after the 6‑day deadline has passed, whichever comes first. This addition would properly limit the expedited procedure to the period when it is actually necessary.

Counting the Votes in Congress

Section 15 of the proposed ECRA lays out the procedures for counting the electoral votes once they are sent to Congress. One of the ECRA’s major improvements of this process over the current ECA is that the ECRA identifies one single valid slate of electors. As noted above, that is the slate backed by the governor’s certificate (as revised by court order, if necessary). For the purposes of section 15, a governor’s certificate “issued pursuant to [section 5] shall be treated as conclusive with respect to the determination of electors appointed by the State[.]”

Yet puzzlingly, section 15 includes language that seems much less definitive. In laying out the step-by-step process for counting the votes, this section requires that the president of the Senate (who is normally the vice president) must “open the certificates and papers purporting to be certificates of the votes of electors appointed pursuant to a [governor’s certificate] issued pursuant to section 5” (emphasis added). All such papers are then read and presented to Congress for potential objections.

The word purporting is borrowed directly from the current version of the ECA, which contemplates the possibility of Congress receiving multiple competing slates of electors backed by different state officials, each purporting to cast the state’s true electoral votes. The ECRA rightly discards the paradigm of competing slates from a single state. So why does the word purporting remain?

It’s genuinely unclear to me whether this language means, for example, that the vice president must present: 1. Any paper claiming to be the true votes of a state, even if its certificate does not match the single correct governor’s certificate; 2. Any paper attached to a certificate that appears on its face to be the true governor’s certificate, even if more than one is received and all but one is a forgery, or 3. Only the single paper that appears most likely to be the true list of votes accompanied by the true governor’s certificate. The bill’s language could arguably direct the vice president to take any one of these three courses, or perhaps more.

But ultimately, even if multiple purported returns from a single state are presented to Congress, this should not matter because of the ECRA’s counting rules. After all of the purported votes have been read and presented to Congress, the ECRA mandates (as noted above) that “only the votes of electors who have been appointed under a [governor’s certificate] issued pursuant to section 5” may be counted. Clearly, to perform this tabulation, a judgment must be made as to which electors have actually been appointed under a governor’s certificate (not just which have purported to be so appointed).

If this judgment identifying the single correct governor’s certificate can be made at the end of the counting process, I see no reason why it can’t be made at the beginning of the process as well. Rather than opening and reading every paper purporting to be the true return, the ECRA should make clear that the vice president should only read those papers that actually are accompanied by the true and operable governor’s certificate. This would have the added benefit of avoiding an embarrassing scenario where fake votes are read out to Congress out of mere formality, with Congress knowing that it need not do anything because the votes will be discounted at the final tabulation.

And this leads to a second ambiguity in the counting provisions. To properly apply the counting rules at the end of the process, some judgment must be made as to which paper is in fact accompanied by the true governor’s certificate, if more than one paper is received from a state. But who makes that judgment? It seems clearly not the vice president. The ECRA makes explicit (in a clarification that I wholeheartedly support) that the president of the Senate “shall have no power to solely determine, accept, reject, or otherwise adjudicate or resolve disputes over the proper lists of electors, the validity of electors, or the votes of electors.”

But if not the vice president, then who? The most likely candidates, under my reading of the bill, are the four “tellers,” two of whom are appointed by the presiding officer of the House and two by the presiding officer of the Senate. The ECRA states that at the end of the counting process the tellers “shall make a list of the votes …; and the votes having been ascertained and counted according to the rules in this subchapter provided, the result of the same shall be delivered to the President of the Senate[.]”

If it is the teller’s responsibility to count the votes, then it is also presumably the teller’s responsibility to adjudicate how to apply the rules to the votes. Rather than using the passive voice, it would be preferable if the ECRA made this explicit. And the bill should also lay out how to deal with the contingency of a disagreement among the four tellers.

Relatedly, the ECRA is also ambiguous as to just how much discretion the tellers (assuming they are indeed the ones to apply the counting rules) may have in applying the counting rules. Return to the hypothetical with which I ended my discussion of the governor’s certification. Suppose, in a highly unlikely event, that the governor fails to follow proper state law in issuing the certificate, yet a court fails to order a revised certificate. If that erroneous certificate reaches Congress, and it is not rejected by a majority vote of both houses, may the tellers themselves decide that the certificate was not “issued pursuant to section 5” because it did not follow section 5’s requirement to follow state law? And may the tellers on that basis decline to count the votes in their tabulation?

The text of the bill seems genuinely ambiguous as to just how much substantive power the tellers should have. But in my view, it would be a mistake to allow the tellers this level of discretion; the better course is to rely on the courts to correct any erroneous application of state law. It would be better to make explicit that in applying this counting rule, all that matters is that a certificate facially complies with section 5.

Finally, there is the procedure for Congress to object to the electoral votes. Closely echoing the current ECA, the ECRA would provide two grounds for a valid objection. The first is that the “electors of the State were not lawfully certified under a [governor’s certificate] according to” section 5.

Given the counting rules discussed above, the choice to retain this ground for objection from the current ECA is surprising. Seemingly, any such electors should not be counted under the final counting rules anyway, whether objected to or not, because such electors would presumably not be “appointed under a [governor’s certificate] issued pursuant to section 5.” In other words, it’s not obvious from the text of the ECRA when there could be an electoral vote that should rightfully be objected to under this standard but that would otherwise be counted at the final tabulation under the counting rules. But some such scenario must exist, or else this ground for objection would be entirely superfluous.

One potential answer is that Congress may look into whether a governor followed state law and toss out electoral votes if the governor did not. If that is so, then this could potentially distinguish Congress from the tellers, who (perhaps) may not exercise that level of discretion. In other words, the ECRA may contemplate Congress as an emergency backstop, should the courts somehow fail to catch and supersede a governor’s violation of state law. I am not sure that this is the best interpretation of the text, but I cannot think of another possibility for when this objection might need to be implemented.

If this is indeed the power of this objection, I have mixed feelings about it. I lean toward the view that the job of evaluating a governor’s compliance with the law should be left with the courts, and is not within the expertise of Congress. On the other hand, a plausible argument could be made that a majority of both houses could only be mustered to overturn a gubernatorial action that is blatantly lawless, and that such a power would serve as a necessary backstop in case the courts are not fast enough. In either case, this category of objection should at the very least be clarified with added detail to explain how, if it all, it differs from the final counting rule for tabulating the votes.

The second ground for objection is that an electoral vote has not been “regularly given.” This phrase is borrowed directly from the 1887 ECA, and no clarity is added. In my view, it would be far preferable to spell out in detail the precise valid grounds for objection. Andy Craig has shown that this is possible. But if Congress is not willing to go this far, it should at least discard a term of art that is now more than a century old.

It is a longstanding rule of statutory interpretation that “when Congress employs a term of art obviously transplanted from another legal source, it ‘brings the old soil with it.’” That means that the choice to retain the term of art “regularly given” from 1887 will lock in its meaning from 1887, making the quest to understand its precise bounds needlessly difficult. Although Professor Derek Muller has done excellent work in explicating the original meaning of this term, it should not be necessary to refer to complex historical research to understand the meaning of a law enacted in 2022. And as I have explained, the legal landscape has significantly shifted since 1887 (including the enactment of the presidential succession rules of the Twentieth Amendment), meaning that the original scope of the term “regularly given” isn’t necessarily the best fit for the law today.

If Congress is unwilling to spell out every ground for objection in detail, then something to the effect of “given in violation of the Constitution’s rules for the casting of electoral votes” would capture a similar meaning but with greater clarity.

Conclusion

By pointing out these areas for improvement, I once again do not mean to give the impression that the ECRA is not a worthwhile bill. If enacted tomorrow, it would be a significant improvement over the status quo. But that does not mean that we shouldn’t consider potential improvements to tighten and clarify the bill’s text. If a potential pitfall or loophole can be avoided, no matter how unlikely it may be, now is the time to consider that risk and address it.