The Legitimate Role of Congress in the Electoral Count

The Electoral Count Act of 1887 (ECA) allows the House and Senate to discount a purported electoral vote by a majority vote in both houses.¹ The ECA is the statute under which Josh Hawley, Ted Cruz, and other Republicans in the House and Senate attempted to invalidate the electoral votes of multiple states on January 6, 2021. It’s also the statute under which senators and representatives have lodged challenges to electoral votes in previous years, including challenges to a “faithless elector” in 1969, to George W. Bush’s electoral votes in Ohio in 2005, and to Donald Trump’s electoral votes in multiple states in 2017.²

There is now growing support among scholars, members of Congress, and the public to amend the ECA so as to avoid a repeat of the 2020 election certification process.³ The details of these reform proposals are important, but they are not the focus of this briefing paper. Instead, I will examine an antecedent question that must be considered before any amendment to the ECA is enacted: Is the law even constitutional? How is it that Congress could have the power to invalidate an electoral vote in the first place? Where does that power come from in the Constitution? These questions must be addressed because if Congress does not have such power, any amendment to the ECA would only serve to further entrench an unconstitutional law.

Whenever the ECA has come under debate, so has the question of its constitutionality. After the 2000 election, scholar Vasan Kesavan published a lengthy law review article in which he concluded that the act is unconstitutional for several reasons.⁴ After the events of January 6, 2021, with fresh calls to amend or abolish the ECA, that piece of scholarship has once again become highly relevant. In addition, professors Jack Beermann and Gary Lawson have joined the fray with their own recent article, in which they reached a similar conclusion,⁵ as have former Fourth Circuit judge Michael Luttig and attorney David Rivkin.⁶ However, none of these arguments make a persuasive case that the ECA is unconstitutional, and none should dissuade Congress from enacting an updated and improved ECA.

Why is there so much long-standing skepticism and doubt as to the ECA’s constitutionality? The root cause is the unusually vague language of the Constitution itself. After laying out a detailed procedure for the electors of the electoral college to meet, vote, and make lists of their votes, the Constitution directs that the electors must transmit their lists of votes “sealed to the seat of the government of the United States, directed to the President of the Senate,” who “shall, in the presence of the Senate and House of Representatives, open all the certificates and the votes shall then be counted.”⁷

As any middle school English teacher will point out, the switch to the passive voice raises an obvious question: The votes shall then be counted … by whom? And just as important, does the power to count implicitly carry the power to judge whether a vote should be counted? As Justice Joseph Story observed, “no provision is made for the discussion or decision of any questions, which may arise, as to the regularity and authenticity of the returns of the electoral votes,” and it “seems to have been taken for granted, that no question could ever arise on the subject.”⁸

If that was indeed the assumption, it turned out to be inaccurate. Given that questions do arise, what level of discretion in counting the electoral votes does the Constitution permit? This question was debated in the 1880s as Congress considered the Electoral Count Act. Supporters of the ECA argued that the duty to count included the duty to ascertain whether a vote sent to Congress was “in fact the lawful vote of a State.”⁹ The act’s proponents believed that Congress’s “determination that [an] alleged return is the legal return is the counting of the vote of that State within the meaning of the Constitution.”¹⁰

Ultimately, those who argued for this more robust form of counting won out with the ECA’s passage. By statute, Congress designated itself as the final arbiter of the electoral count, denying any meaningful role to the president of the Senate (who is normally the vice president) and everyone else present at the counting. This final say includes the discretion to reject an electoral vote as illegal or inauthentic by the mutual decision of a majority of both houses.

But did Congress have the power to establish these rules by statute? Scholar Vasan Kesavan urges that it did not. In his extensive article, Kesavan argues that the Constitution denies Congress the power to enact a statute regulating the counting of electoral votes. Kesavan’s strongest argument for this point is one of “negative implication,” which is the lack of any grant of power when other comparable powers are expressly given. The Constitution explicitly grants Congress the power to pass statutes regulating the “Times, Places and Manner” of congressional elections.¹¹ It likewise grants Congress the power to establish the time period during which states must choose their electors and the day on which those electors must vote for president.¹² But there is no comparable language granting Congress the power to determine the manner of counting the votes. As Kesavan notes, the Constitution provides that “‘the Votes shall then be counted’—not, ‘the Votes shall then be counted as Congress may by Law have directed.’ The Framers could have so provided but they did not.”¹³

Kesavan thus believes that Congress may not determine the manner of counting the votes. Kesavan argues that instead there is only one permissible manner of counting the votes. In Kesavan’s view, this manner must be discerned from the Constitution’s text itself, not established by statute.¹⁴

Nonetheless, it turns out there is significant overlap between the manner that Kesavan believes the Constitution prescribes and the manner Congress has, in fact, chosen with the ECA. Kesavan agrees with those who enacted the ECA in one respect—that the constitutional command that the votes be “counted” necessarily requires discerning what is and is not an electoral vote. Kesavan thus believes that lists of electoral votes must be discounted if they are either hoaxes or not signed, certified, and sealed as the Constitution prescribes. Similarly, Kesavan agrees that votes not given on the date established by law must be tossed out, as must lists containing more votes than a state is entitled to or lists sent by territories that are not states. In all of these instances, Kesavan accepts the ECA’s premise that the lists received are not truly votes and thus should not be counted.¹⁵

But the ECA departs from the procedure Kesavan discerns within the Constitution in two key respects. First, Kesavan argues that the only acceptable decisionmaking body is the two houses of Congress combined into a single 535-member voting body, not the two houses voting separately. Kesavan finds this “unicameralism principle” in several sources, including the command that the votes be counted “in the presence of the Senate and House of Representatives,” which Kesavan takes to imply that the members may not briefly separate and go back to their distinct houses in the middle of the count.¹⁶

And second, the ECA permits Congress to discount a purported electoral vote for a broader range of reasons than Kesavan believes is permissible. What impermissible reasons are these? The text of the ECA states that Congress may discount votes for not being “regularly given by electors whose appointment has been lawfully certified.”¹⁷ And debates during the ECA’s passage show that these two terms—“regularly given” and “lawfully certified”—were generally understood to encompass reasons extending further than those Kesavan would accept.

For example, one legislator believed a vote could be rejected if it were cast by an elector who held some other public office at the same time—a dual role that the Constitution expressly prohibits.¹⁸ Another legislator similarly argued that a vote could be discounted if it violated one of the Constitution’s more obscure rules: an elector may not cast votes for president and vice president for two people both from the same state as that elector.¹⁹

Kesavan believes this goes too far, because the Constitution draws a line such that neither Congress nor anyone else present at the counting may “judge the manner of appointment or qualifications of electors. Once the vote of a constitutionally ineligible elector is transmitted in the electoral certificate, that vote is final and must be counted.”²⁰ Likewise, Kesavan believes the electoral vote counters “may not judge the acts of electors—period.”²¹

To Kesavan, a problem with a vote list itself can justify rejecting it as not truly a list of countable votes, but problems with who the electors are or who they voted for go beyond that counting function. In Kesavan’s view, the electors themselves must police their own qualifications and votes, not Congress. Kesavan argues that “the Constitution trusts electors with the last word on the persons receiving votes,” or, at a minimum, “trusts electors more than Members of Congress,” and that it is “thus unconstitutional for the joint convention to reject electoral votes contained in authentic electoral certificates—even when those electoral votes are unconstitutional.”²²

Kesavan finds this principle in several sources. First, “the electoral colleges constitute a separate and coordinate branch of the Government of the United States” that is not subordinate to Congress.²³ He therefore argues that “the electors should have interpretive authority of the Constitution with respect to the powers committed to them.”²⁴

Second, the Constitution explicitly assigns one such quasi-judicial role to Congress, making each house “the judge of the elections, returns and qualifications of its own members.”²⁵ Kesavan reads from this grant of limited judicial power a negative implication that Congress cannot also be the judge of the qualifications of the electoral college members. And by extended analogy, this clause also suggests, in Kesavan’s view, that the electoral college itself is the most natural judge of its own members’ qualifications and actions.²⁶

Is Kesavan right? Is the Electoral Count Act unconstitutional? Given all of the structural analogies that Kesavan identifies, the line he draws between judging the validity of the lists and judging the acts of the electors is a reasonable one. But the question is whether this is the only line that the Constitution permits, such that the choice to draw a different line is unconstitutional. It is on that question that Kesavan’s arguments ultimately fail to persuade.

To be sure, Kesavan is correct that the Constitution does not allow Congress to “second-guess the electors’ judgments.”²⁷ But finding that the electors have violated an explicit constitutional rule is not overruling their judgment as to who would make a good president. It is rather more akin to taking a quasi-judicial role of examining whether the electors complied with the law, which in this case is the Constitution. And there is nothing inherently unconstitutional in a branch separate and coordinate to the electoral college taking such a role. After all, the executive and legislative branches are separate and coordinate branches with respect to the judiciary, yet the judiciary nonetheless frequently sits in judgment of the other two branches and of their exercises of power.

Further, a plausible argument could be made that a vote cast by a disqualified elector is not truly a vote in just the same way that an improperly certified vote is not truly a vote. The command that only the votes be counted can justify discarding both types of violations. And while it is true that the Constitution does not explicitly give that judicial role to Congress, it does not explicitly give it to the electoral college, either.

What is the best argument that the manner of counting votes chosen by the ECA is permissible? Given the sparseness of the Constitution’s language, the ECA can best be understood as “gap filling.” The Constitution itself demands that the votes shall “be counted,” but it gives no explicit guidance as to how to conduct that count. The ECA adds a specific procedure onto this barebones framework for determining how the count proceeds and how a decision should be made as to whether to include any purported vote.²⁸ Both the “regularly given” and “lawfully certified” categories, properly understood by their meaning at the ECA’s passage, limit Congress to rejecting purported votes that can reasonably be described as not truly being votes at all.

In addition, assigning the two houses of Congress to make this determination is a plausible choice, though not necessarily the only acceptable one. In their recent article, Beermann and Lawson propose that under the Constitution it is the vice president who must make this determination, since the Constitution places the votes in the vice president’s hands as the opener of the certificates.²⁹ That choice might be acceptable, too, as might Kesavan’s proposal of a single 535-member unicameral Congress.³⁰ Indeed, even assigning the decision to a disinterested arbitrator, akin to the Senate Parliamentarian, would not clearly violate constitutional text or structure. Nor would imposing a two-thirds threshold in each chamber to reject a vote, rather than a mere majority. The fact that scholars have reached different conclusions as to what the Constitution requires suggests that the Constitution truly has left a gap—one that can be filled in more than one way.

But looking beyond the particularities of the ECA’s rules, Kesavan also argues that the process by which the ECA became law was itself unconstitutional. The ECA was enacted as a statute, passed by the House and Senate, and signed by the president. Among other provisions, the ECA establishes rules for how Congress must conduct the electoral count. But as Kesavan notes, the Constitution allows each house to “determine the Rules of its Proceedings” by a vote of that house alone.³¹ Proceedings before both houses, like the State of the Union Address, are determined by a vote of the two houses concurrently without the signature of the president. Thus, Kesavan argues that rules regulating how Congress conducts the electoral count must be enacted by the two houses (without the president’s involvement) every four years, as a concurrent resolution in effect for that count only.³²

Kesavan makes a strong argument that the ECA’s provisions regulating how Congress conducts the electoral count are indeed rules of proceedings that may be established by concurrent resolution rather than by statute. But that does not mean that the ECA is unenforceable or unconstitutional. Rather, as other scholars have suggested, the Constitution most likely gives the two houses the authority to change the ECA’s rules for conducting the count by concurrent resolution without the need for a presidential signature.³³ But until the two houses actually exercise that option, the ECA stands as a valid exercise of their rulemaking power (since it was, of course, passed by both houses in 1887). And so long as both houses retain the power to change a rule by simple majority vote, there is no constitutional requirement that the rule be reenacted every term.³⁴

Even if the ECA’s procedures and standards for counting the electoral votes are constitutional, one question remains as Congress considers potential reforms. No matter who makes the judgment call as to which votes are valid, that judgment is undeniably of a judicial nature. None of the potential counters present at the counting session seem as natural a fit for this role as the judicial branch. Indeed, much of the controversy over who does the counting would be less consequential if the judicial branch had the power to review the decisions made during the count. What role, if any, can the judicial branch play after the electoral votes arrive at Congress?

To Kesavan, the answer is “none.” He notes that after the votes are opened, they must “then” be counted, suggesting an immediacy that does not allow any pause for judicial review.³⁵ Further bolstering this “immediacy principle” is the Constitution’s rule for what must happen if no presidential candidate receives a majority of the electoral votes: “the House of Representatives shall choose immediately, by ballot, the President.”³⁶ Shifting the final resolution of the count to the judicial branch would not allow for such immediacy. “After all,” Kesavan observes, “judicial determinations take time.”³⁷

Kesavan argues that this immediacy principle means there is no time for federal courts to investigate “the validity of an elector’s appointment once the electoral votes are being counted.”³⁸ Beermann and Lawson disagree. Given their view that the vice president is responsible for the initial count, they argue that making such unilateral authority unreviewable would be “inconsistent with the theory of separation of powers and with checks and balances.”³⁹ And Luttig and Rivkin would go even further. They note that “whether electors are validly chosen is a quintessentially legal determination, not a political one.” For that reason, they believe that only the judicial branch has the authority to decide such disputes, not any of the political actors present at the count.

In my view, the immediacy principle is not so ironclad as to override the Constitution’s default grant of the judicial power to the judicial branch. As Kesavan notes, the Twelfth Amendment contemplates that the House might fail to select a president by Inauguration Day, which suggests that the House’s duty to “choose immediately” need not be completed on the same day as the votes are opened.⁴⁰ Although the count must occur “then” when the certificates are opened, many events that must occur on a particular day are nonetheless subject to judicial review and confirmation at a later date (including the general election of the electors on a Tuesday in early November). And although the House must vote “immediately” if no candidate has a majority, this vote could be interpreted to occur immediately following when the courts make a final determination that no candidate has, in fact, won a majority.

As this discussion has shown, the determination of who may count, how they may conduct the count, and whether that count is subject to judicial review raises difficult questions of constitutional interpretation. But these difficult questions should not obscure a more basic point. The objections raised on January 6, 2021, like most of the objections raised under the ECA during its history, were not valid under the text of the ECA.

Properly understood according to the ECA’s original meaning, challenges to an electoral vote as not “regularly given” are limited to problems that occur on the day the electors vote and mail their sealed lists to Congress—problems that a court would not be able to resolve before that day.⁴¹ This means that “regularly given” challenges do not encompass objections to the general election in early November. The ECA’s drafters wanted such disputes resolved by the courts, not by Congress or by ad hoc commissions, as in 1876.⁴² The ECA, by pushing the date the electors meet and vote to later in the year, was intended to allow courts enough time to resolve all electoral disputes before that day.⁴³

And challenges to an electoral vote as being not “lawfully certified” are similarly cabined. When a state’s governor has, in fact, certified a single slate of electors and the state’s court system has resolved any election disputes, “lawfully certified” challenges are limited to problems that are evident on the face of the state’s certification, such as a list certifying more electors than a state is entitled to.⁴⁴ Thus, when the courts resolve disputes over the general election, Congress can’t use a “lawfully certified” challenge to relitigate them.

This dividing line makes sense. Before the electoral votes are sealed and mailed to Congress, there is no logistical reason why the courts cannot resolve any dispute. A hypothetical ECA that allowed Congress to routinely second-guess the election of the electors or to decide other disputes arising before the electors cast their ballots would be on more uncertain constitutional ground than the ECA we have. Since such questions can be resolved before the electoral votes are cast, they are not inherently part of the task of counting the votes mailed to Congress.

Fortunately, that is not the ECA we have. But unfortunately, that is the ECA that many members of Congress over the years have acted as if we have. What the events of 2021 and other recent elections have shown is that the question “Is the ECA constitutional?” might be framed too narrowly. In one respect, the question refers to the text actually passed by Congress in 1887. That text, in most respects, is a reasonable attempt to allow Congress to quickly count the electoral votes and resolve the legitimacy of those votes (although there are certainly many areas where the ECA could be improved by amendments that added more clarity).⁴⁵

But the second question is whether the ECA, as it has been used, is constitutional. Here, if the answer is no, the fault is not with anyone in 1887 but instead with those in 2021, 2017, and 2005, among other years. As these examples have demonstrated, a majority of both houses has the raw power to reject an electoral vote, not only because it was not properly cast as an electoral vote, but also for other improper reasons. Ultimately, it is only the failure to obtain a majority, not the original meaning of “regularly given” or “lawfully certified,” that has prevented Congress from doing so.

Instituting judicial review, if constitutionally permissible, would be one way to curtail this threat. But in the end, no matter who is responsible for counting the electoral votes, the system depends on those people doing so in good faith and with humility. Any given election cycle, the constitutionality of the electoral count depends not just on the rules we set in advance, but on the people we entrust to follow them.