Most of the attention for Electoral Count Act reform has focused on the bill introduced by a bipartisan group of senators led by Sen. Susan Collins (R‑ME) and Sen. Joe Manchin (D‑WV), the Electoral Count Reform Act (ECRA). Now, however, another ECA reform bill has been introduced in the House by Rep. Liz Cheney (R‑WY) and Rep. Zoe Lofgren (D‑CA), dubbed the Presidential Election Reform Act (PERA).

Lofgren is the chair of the Committee on House Administration, which has jurisdiction over election law matters, and both Lofgren and Cheney also serve on the January 6th Committee, which Cheney vice-chairs. The House apparently intends to move quickly on the Lofgren-Cheney bill, which will be taken up by the House Rules committee this afternoon, with a vote by the full House expected by the end of the week. In contrast, the Collins-Manchin bill is slated for a committee markup in the Senate Rules committee later this month, on September 27.

The two bills are broadly similar in most respects, but with some key differences that will have to be reconciled. It is likely some of these changes will be reflected in amendments made to the Senate’s bill, with amendments expected to bring the two closer together in at least some respects.

Broadly speaking, the new House bill reflects a more aggressive approach to constraining the various actors involved in the presidential election process: state legislatures, state governors and executive branch officials, the National Archives, the vice president, and Congress. In some respects, that makes it a more conservative and originalist bill, keen to limit the discretion of political actors in what is supposed to be a mandatory and fundamentally non-discretionary constitutional process. To do this, PERA provides more specific and concrete limits in the law itself, and it also provides a somewhat more expansive role for the federal courts in enforcing those provisions.

Here are the key provisions in the House’s PERA bill, how they differ from the current draft of the Senate’s ECRA bill, and some thoughts about which provisions are the likeliest to be incorporated into a reconciled bill.

State Law Enacted Prior to Election Day

One central purpose of Electoral Count Act reform is to repudiate the idea that state legislatures have any power to overturn their state’s presidential election results after the fact. This reflects the Constitution’s basic division of power as to presidential elections: states are free to choose whatever manner they want for choosing their members of the Electoral College (all states have chosen to do so by popular election), but Congress sets the time of choosing electors.

Both bills insert, in relevant places throughout Title 3, Chapter 1 USC (the existing Electoral Count Act), the clarification that state law as it stood on election day is determinative, and states are not permitted to change this after the fact. The House bill refers to “in accordance with State laws duly enacted prior to such day,” (meaning election day in early November), while the Senate version likewise refers to “in accordance with the laws of the State enacted prior to election day.” (On a minor stylistic note, the Senate bill explicitly defines this day with the label “election day,” making it easier to refer back to throughout the rest of the bill, though the substance of such references isn’t affected).

Most important of the various places this language is inserted is in 3 USC 1, which is where Congress exercises its time-setting power by specifying the first Monday after the first Tuesday in November, the same day as elections for Congress and most state-level offices. Since this is the deadline for states to actually choose their electors (aside from any time needed to count the votes already cast), it is also the inherent deadline for states to decide how they will do so.

This repudiates any notion that state legislatures can convene in late November or December to overturn their popular election results, a spurious idea advanced by Trump allies in their attempt to obstruct the 2020 election. While no legislature ultimately acted on that unconstitutional scheme, both ECA reform bills would provide a clear statement that any such attempt is null and void.

Extended Voting in Emergencies

Both bills also address 3 USC 2, the so-called “failed elections” provision. Here, the goal is to provide a narrow deadline extension for elections in case of a catastrophe, but limited only to severe natural disasters and comparable emergencies severely disrupting an election.

Unlike the Collins-Manchin bill, the Lofgren-Cheney proposal provides a stricter definition of when this emergency extension could be invoked, as well as much more specific procedures for how it can be invoked. As currently drafted, the Senate’s ECRA bill allows a state to extend voting “as necessitated by extraordinary and catastrophic events” with no further definition of that term or any details, simply providing that it must be done under state law as enacted prior to election day.

In contrast, the new House bill would have a more detailed definition of what counts as a sufficient catastrophe, including that it must affect a “substantial” and possibly outcome-determinative number of votes. The House bill also explicitly covers emergencies which affect not just the ability of people to cast votes, but also the ability of the votes to then be counted, an important addition. Consider, for example, the worst-case scenario of a fire destroying the building where a large number of ballots are being counted, rendering it impossible to count the destroyed votes.

PERA would only allow an extended period of voting for up to five days after election day, whereas the ECRA draft provides no specific time limit. PERA would only allow extended voting to apply in the geographic area directly affected by the emergency, rather than statewide, and would specify that votes already received must still be counted in such a scenario. PERA’s definition of a sufficient emergency would also require showing that the number of votes affected is enough to potentially affect the outcome. For all of these rules, PERA adopts a “clear and convincing evidence” standard.

But the biggest difference is that PERA would not leave it up to the states to decide when such a force majeure emergency has occurred. Rather, an affected presidential candidate would have to bring an action against the state’s chief elections official in federal court, and a three-judge panel would provide an expedited ruling. Only then, with federal court order in hand, would a state be authorized to extend presidential voting.

There are pros and cons to PERA’s approach. On the plus side, PERA’s narrower and more specific definition of an emergency, as well as including a fixed deadline for extended voting, are worthwhile inclusions. These are both important points to prevent partisan abuse of this provision and to make sure that the process keeps moving along in a timely manner.

On the downside, this more detailed procedure and requirements might conflict with existing state laws and procedures for extended voting in an emergency. Such a catastrophe has never affected a presidential election, but we saw an example of how state laws address the possibility during the September 11th attacks, when New York City had to reschedule its municipal primaries which were originally being held on that date.

In particular, it’s possible this distinct set of presidential-only rules would split the presidential election from how the state is handling other offices on the same ballot in the same emergency. This would present states with a complicated logistical problem if, for example, an extended period of voting is authorized for five days for president but extends to ten days for all other offices on the ballot, such as for governor and other state offices. A similar complication could occur if the area subject to extended voting differs for president and for those other offices.

In considering possible amendments to ECRA, the Senate would do well to incorporate PERA’s stricter definition of emergencies authorizing an extended voting period. The inclusion of a deadline for extended voting is also worthwhile–states should not put it off too late with sensitive dates for the Electoral College process fast approaching. However, as a logistical matter, allowing the state’s own election emergency laws to be used would be preferable, within some guardrails such as a reasonable timeframe (I have previously suggested the end of November at the latest), and the narrower definition of when the provision may be used.

There are potential complications from only allowing this kind of emergency to be declared via an adversarial judicial process. In most cases where there has been such an extreme emergency, the state might not be inclined to oppose allowing an extended period of voting. This presents an awkward situation for the courts, especially if no candidate intervenes to oppose the request. It could even raise complicated Case or Controversy Clause questions, insofar as a federal court is being asked to order a state to do something the state already wants to do and does not oppose.

In terms of process, it would probably be preferable to let state election authorities initiate the process of declaring an emergency extension, and then provide a judicial mechanism to litigate that decision if it is contested by any of the affected candidates. With that one caveat, the stricter guidelines in PERA are on the whole preferable to the more open-ended provision in ECRA.

Safe Harbor

As I previously discussed here, PERA would completely eliminate the “safe harbor” provision. This is the muddled rule that currently attempts, but largely fails, to give conclusive effect to a state’s appointment of its electors. In PERA, this function is made redundant because the same principle is addressed in the section outlining how the state’s appointment certification process (including any relevant litigation) is to be conducted, as well as in the section covering the rules for the joint session of Congress.

ECRA gets to much the same result, but with a slightly different organizational approach. The existing safe harbor section is instead amended and much expanded to cover the entire certification and related litigation process. One distinction, however, is that this leaves the “conclusive” nature of the resulting certificate slightly more ambiguous with respect to the rules of the joint session. As I’ll cover below, PERA is much more extensive in its rewriting of the joint session rules, including on this point.

In their broad strokes, however, both bills would handle this process very similarly. The governor of each state (or, under ECRA, another official if the state so chooses in advance) would have an affirmative duty to issue the correct certificates to the duly chosen electors. If a governor refuses to do so, another three-judge expedited panel would be able to order them to comply. Both bills would also generally provide that the court’s ruling (with possible appeal to the Supreme Court) is a conclusive, final determination of who the state has appointed as its electors. While PERA contains other judicial review mechanisms, this is the core opportunity for the courts to intervene that is common to both bills, ensuring that prior-enacted state law as adjudicated by the courts gets the final say on the matter, not Congress.

Unlike ECRA, PERA would expressly allow the court to order a back-up official in the state, such as the secretary of state, to comply and issue the certificate. While this is something that would already be possible under generally applicable rules of civil procedure, expressly including it in this part of the law is a good call. This also avoids some ambiguity in ECRA that could be read to suggest the court’s ruling in and of itself is to be treated as the necessary certificate by Congress, even if no state officer ever actually provides it. While federal courts can compel a state to abide by its mandatory constitutional duties, the appointment of electors must ultimately be carried out by the state in some form or another, and the federal court is not a state. Explicitly ensuring that the court can go around a defiant, contemptuous governor to secure the necessary certificate from an alternative state officer is a good improvement.

Transmitting the Appointment Certificates

The next section concerns the role played by the Archivist (the head of the National Archives and Records Administration), and how a state’s appointment certificate is to be transmitted to the Archivist. This is important because it is the Archivist who ultimately provides these certificates to Congress for its use during the joint session.

One problematic logistical element of the existing ECA has been its requirement to use “registered mail” for this purpose. As anybody familiar with the United States Postal Service is aware, this is less than ideal for time-sensitive documents, and in fact the requirement has been neglected in practice at times.

Both ECRA and PERA would instead allow states to transmit their certificates by the most expeditious method available. In practice, this amounts to authorizing the use of couriers, which is a desirable best practice for such constitutionally important pieces of paper. PERA goes further and also explicitly provides for simultaneous electronic transmission–in other words, by email, fax, or the like. In a worst case scenario interrupting physical travel between the states and Washington, DC, this electronic copy would suffice.

This section in PERA also contains the judicial review mechanism for addressing any failure or refusal by the state to provide the certificate naming the electors, as discussed above. Unlike ECRA, this mechanism would also allow seeking an injunction against the Archivist in the unlikely scenario this official goes rogue and refuses to accept the certificate. While such a scenario is unlikely, this is an important inclusion because it speaks to the only official in the process who is within the federal executive branch and thus, ultimately, under the control of the incumbent president.

PERA is also more definitive than ECRA in spelling out what, exactly, it means for the state’s appointment certificate (as issued in compliance with any court rulings) to be conclusive. This tracks an important distinction running throughout the ECA that, in general, PERA is clearer on than ECRA: the distinction between who a state has appointed as its electors, as opposed to other questions regarding how and for whom the electors have voted. On the former, Congress is properly bound to accept the outcome as a matter of fact, reserving its judgment on possible objections to only the latter category.

Both PERA and ECRA also contain language, in their relevant judicial review procedures, making clear that these avenues for expedited litigation do not supplant other avenues for possible lawsuits over voting rights issues, Fourteenth Amendment claims, and other matters where individual voters might be plaintiffs instead of a directly affected presidential candidate. PERA, however, drafts this disclaimer in somewhat tighter terms, and avoids ECRA’s potentially unintentionally broad authorization of expedited litigation “with respect to” a state’s appointment of electors. On the other hand, PERA also adds an additional very broad general cause of action discussed below, which undercuts the intention for ECA reform’s judicial mechanisms to be very narrow in scope.

Calendar Changes

ECRA is very minimalist in its changes to the timeline, with only one small change: moving the date of the Electoral College meeting back by a single day. PERA moves that date further back and also provides a number of other key deadlines leading up to it, such as requiring any Supreme Court appeals to be settled by the day before the Electoral College meets.

Under current law, the Electoral College meeting day is defined as “the first Monday after the second Wednesday in December.” In 2020, for example, this fell on December 14. PERA would move that later, to December 23, in order to provide a bit more breathing room for any litigation to be resolved.

Tweaks to the calendar aren’t the most important question for ECA reform, but providing more time for the courts is desirable, and would help avoid a scenario like Bush v. Gore, where the looming statutory deadlines provided the impetus for the Supreme Court to declare that a constitutionally compliant recount in Florida simply could not be completed in time. Though this probably did not affect the ultimate outcome, it’s not a good look for courts to give the impression that they weren’t able to properly adjudicate an election dispute in the time allotted. In this respect, PERA’s later date for the Electoral College meeting makes sense.

Rules for the Joint Session of Congress

Now we come to the biggest difference between ECRA and PERA: the rules for the joint session of Congress, when both chambers meet as required by the Constitution to count the electoral votes. Under both bills, this proceeding would remain on January 6, a date which is both practical but has also now acquired heavy symbolic importance. It is this provision, 3 USC 15, which is the most notoriously confusing and uncertain part of the existing Electoral Count Act.

ECRA adopts a least-change approach to these rules, largely retaining the relevant language, with some stylistic restructuring for clarity. PERA, in contrast, largely junks the existing Section 15 and goes in for a more comprehensive re-write of the procedures. Given the centrality of this issue to ECA reform, PERA’s approach is, on the whole, better.

Both bills would adopt one of the most widely agreed upon changes: raising the number of cosponsors needed to trigger a debate and a vote on objections to a state’s electoral votes. Under the current ECA, this process can be invoked so long as at least one Representative together with at least one Senator both co-sign the objection. This is an untenably low requirement, and has proven to be an invitation for ill-founded and constitutionally dubious grandstanding, not just in 2021 but also in several prior elections.

ECRA would instead require one-fifth of both chambers, whereas PERA would go further and require one-third of both chambers. The objections to votes in the 2020 election would have failed under either standard, falling far short of the number of senators needed in either case. It’s not the most important issue, but there is some constitutional reason to favor one-fifth, as this is the level the Constitution mandates to request “the yeas and nays” (that is, a recorded vote) on any matter in both houses. On the other hand, one-third would probably be a constitutionally permissible rule to adopt in this context, and would go further to ensure that Congress only considers objections with a realistic chance of being sustained.

PERA would provide some definitive answers for the workings of parliamentary procedures in a joint session, something Congress has never quite sorted out in detail. It would accomplish this by providing an exhaustive list of motions that are permitted, as well as the necessary number of cosponsors for each. This would most substantively cover motions challenging electoral appointments or electoral votes, but also provide a clear procedure and limits for when the body may vote to recess.

These rules relating to possible recesses would serve to impose strict limits on the ability to delay the proceedings, by only allowing a recess until the following day, and not allowing any motions to recess at all if the joint session has already gone on for three days past January 6. In this respect, PERA would repudiate any obstructionist strategy of trying to run out the clock past Inauguration Day, fixed by the Constitution as January 20. Similarly, PERA also contains a detailed list of time limits for debate on the various kinds of motions, with a provision to shorten those time limits (from two hours to a half-hour) if the process has already dragged on for several days.

In handling the counting of votes from each state, PERA would also clarify the two-step procedure: first Congress acknowledges (and hears any possible objections to) the list naming who has been appointed as a state’s electors. Only after this step, and after the opportunity to raise objections to any elector appointments, would Congress move on to a separate announcement of the actual votes cast and the opportunity to raise objections to those votes. This bifurcation would eliminate the bad precedents Congress has established at times in confusing these two steps, by improperly allowing objections to electors under the guise of objections to their votes, thereby evading the intended limits on the former. Such confusion would not be possible under PERA’s rules, since objections to particular votes cast would only be allowed after the opportunity to object to the identity of the electors has passed.

PERA also provides a clearer and more absolute disclaimer of unilateral authority for the vice president as presiding officer, and a mechanism for parliamentary appeals to enforce the rules of order. This gives teeth to the limit on the vice president’s authority by making sure Congress has a clear path to overrule a rogue vice president, something missing from both the current ECA and the ECRA draft.

Valid Grounds for Objections

The biggest rules change, and the biggest difference between ECRA and PERA, concerns the valid grounds for members of Congress to raise an objection to counting a state’s electoral votes.

The most curious choice of ECRA’s drafters was to leave this provision largely untouched, retaining two of the status quo ECA’s most confusing terms: that objections may be made only on the basis that electors were not “lawfully certified” or that their votes were not “regularly given.” It is the vagueness of these terms which has provided the basis for the dangerous and erroneous claims that Congress has essentially unlimited authority to reject electors and electoral votes for any reason it chooses.

PERA would instead scrap “lawfully certified” and “regularly given” altogether. The former is reduced to a very narrow scenario by the conclusive authority given to the state’s appointment certificates as ruled on by the courts. Thus, any potential objections to the appointment of electors would be limited to a kind of fail-safe, allowing Congress to enforce that finding in the event that, somehow, votes not comporting with the state’s certified list of named electors are presented to the joint session.

It is under the category of “regularly given” that most of Congress’s proper constitutional authority falls, to reject votes otherwise cast by the correct electors but which fail to comply with a number of specific constitutional rules. PERA enumerates those specific rules, similar to the case I made for doing so in my policy analysis on ECA reform.

These rules would include, first and foremost, that only states (and the District of Columbia under the Twenty-third Amendment) may cast electoral votes. Though this might seem obvious, it’s come up before, with a dispute over if Missouri had become a state in time to participate in the 1820 election. Additionally, a state may only submit the number of votes to which it is entitled, which again seems obvious but must be included to ensure this list exhausts all possibly legitimate objections. So, if Delaware attempts to cast fifty votes, Congress obviously should not count more than the three votes Delaware actually gets under the Constitution.

Next up is the rule for elector eligibility: members of the Electoral College are forbidden from holding any federal office, and are also potentially disqualified under Section 3 of the Fourteenth Amendment, a provision written to target ex-Confederates in the aftermath of the Civil War. An electoral vote cast by a constitutionally ineligible elector, even if that person is otherwise the person duly chosen by the state, should properly be discarded.

After elector eligibility comes presidential eligibility. Here, PERA perhaps errs on the side of getting too specific. Rather than simply allowing objections on the grounds that a person voted for is not eligible to be president or vice president, PERA’s list breaks this category down into a separate sub-list, enumerating each of the applicable constitutional clauses. Uncontroversially, this includes the provisions saying presidents must be thirty-five years old, a natural-born citizen, etc. But PERA would also weigh in on two uncertain questions, by explicitly providing for eligibility objections under Section 3 of the Fourteenth Amendment, as well as allowing objections to vice presidential candidates under the Twenty-second Amendment’s establishment of presidential term limits.

These are both open questions of constitutional law, perhaps best left for Congress to address if and when they actually arise. It is unclear and actively debated if the Fourteenth Amendment’s prohibition actually covers the presidency, which is not enumerated alongside the various other offices (such as senator and representative) which are specifically named. The question hinges on if the presidency falls under the catch-all term of art “office under the United States,” and there are reasonable arguments to be made that it does not, and that this omission was deliberate. Obviously, this is a particularly contentious issue given the attempts to use Section 3 to disqualify participants in the January 6th attack on the Capitol, and the desire of some to ultimately apply the provision to Trump himself in the event that he runs in 2024. While a Section 3 objection would already be allowed under the more general terms of both the existing ECA and the ECRA draft, specifically enumerating it in this manner would mean Congress taking a clear position now that Section 3 does cover the presidency.

Less contentiously, PERA’s list of eligibility objections also includes the Twenty-second Amendment’s term limits provision, and it would apply that as a possible objection to both candidates for president and candidates for vice president. The latter is a similarly untested and debatable point: it’s not clear if the amendment’s ban on any person being elected president more than twice would also forbid a former two-term president from being elected vice president. This discussion is likely to remain academic, given how unlikely it is that any ex-president would attempt a run for the number two seat. But this uncertainty likewise weighs in favor of a more general, categorical provision for candidate eligibility objections, leaving the matter for Congress to decide if and when it ever arises.

Lastly, objections would be permitted to enforce the procedural requirements of the Twelfth Amendment regarding when and how electors must vote: that they must meet on a certain day, which the Constitution requires to be the same throughout the United States, that they must vote by ballot, and that they must sign and certify the list of votes in a certain manner before transmitting the results to Congress, addressed to the president of the Senate. Like with the rules limiting votes to states and limiting states to the proper number of votes, these rules are unlikely to ever be violated, but including them serves to demonstrate the exhaustive nature of the list. This is important, because it shows that the law allows Congress to handle any theoretically possible invalid vote, and thus there is never any need for Congress to go around this list and consider other kinds of objections.

Enumerating the list of valid grounds for objections is the desirable best practice here, and my first preference. There is, however, understandable concern about getting this list exactly right. The possible issues with getting too specific on uncertain questions of presidential eligibility is a good example. There is, however, a possible middle ground between ECRA’s excessive vagueness and PERA’s excessive specificity. That would be to provide a tighter categorical definition of the terms “lawfully certified” and “regularly given,” a possibility I outlined in more detail here.

PERA also makes an important change that was overlooked in the current ECRA draft, which preserves the current ECA’s rule that the vice president must acknowledge any papers “purporting to” be electoral votes. Since the whole point of the process is to present Congress with only a single set of definitive electors and electoral votes from each state, this provision is superfluous and problematic. In fact, on the face of it, Vice President Pence neglected to follow this rule when he refused to acknowledge the fake certificates submitted by Trump supporters from various states in the 2020 election. Even worse, this would seem to require acknowledging the totally frivolous joke submissions which have sometimes been mailed in by pranksters.

Though the current ECA and the ECRA would both have such fake votes discarded since they lack the necessary certification by a state, there is no reason such invalid submissions should be acknowledged at all. PERA correctly strikes this language, eliminating any possible confusion.

Lastly, PERA addressed an important issue my colleague Thomas Berry has written about: how possible objections to presidential candidate eligibility must be handled in compliance with the Twentieth Amendment. In short, the votes must still be counted, to ensure that the presidency then passes to the ineligible candidate’s running mate as the Constitution requires, rather than possibly going to the ineligible candidate’s defeated opponent. By enumerating the valid grounds for objections, PERA is able to identify this particular kind of objection as subject to this special rule.

General Cause of Action

In addition to all of these procedures, PERA would create a more general cause of action, allowing candidates to invoke the expedited three-judge procedure to address possible obstruction by state-level officials below the level of the governor’s final certification of the appointment of the state’s electors.

Think of the fights that have come up over county-level canvassing boards refusing to certify the popular vote totals in their respective counties, for instance. PERA would allow the three-judge procedure to be invoked against any unlawful obstruction of a state’s process for counting and certifying the results of its election.

However, it is unclear that existing judicial mechanisms at both the state and federal level are insufficient to address these possibilities. Unlike the governor’s appointment of electors, these lower-level electoral mechanics do not represent a single point of failure. States are already well-equipped to handle such attempts, and federal courts already have jurisdiction over any possible federal claims (such as voting rights) that might arise from such a scenario.

In general, the avenues for new federal causes of action and special expedited judicial procedures should be kept narrow, to only those steps in the process which otherwise might not be addressable in a sufficient and timely manner. While a rogue local canvasser is a serious problem, it is unclear that any new procedures are necessary to handle the matter, nor that ECA reform should reach that far into a state’s internal election administration processes. Given the reticence on the Senate side to impose any substantive expansion of federal court jurisdiction over such matters, it is unlikely this provision of PERA will make it into any bill ultimately passed by the Senate.

Conclusion

On the whole, both versions of ECA reform represent thoughtful, carefully considered attempts to answer some very complex constitutional and practical questions.

While my purpose here is to analyze the policy merits and not the political path forward, the widely shared presumption is that ECRA, with its broader bipartisan support in the Senate, is likely to be the vehicle that ultimately moves towards passage. But the timing of PERA’s introduction and possible passage by the House presents a good opportunity for the Senate to adopt some of its most important contributions when ECRA receives its committee markup later this month.

To summarize, the most important differences worth incorporating from PERA include a stricter set of rules of extended voting during a qualifying emergency, a stricter set of rules limiting the valid grounds for objections, and a more detailed and comprehensive overhaul of the rules of procedure for the joint session of Congress. A bill which adopted those points would address the most important concerns expressed in PERA, resulting in a much-improved final product.

The reformed Electoral Count Act is a law that must stand for decades to come, potentially a century or more. As we saw in the 2020 election crisis, once-obscure uncertainties and academic disputes about these procedures can suddenly become all too real, even leading to deadly violence. The rules for deciding who will hold our nation’s highest office must be clear, definitive, set in advance, and easy to follow based on the plain text of the law. In introducing their own well-considered version of what that looks like, Reps. Lofgren and Cheney have provided an important contribution to that goal.