The Electoral Connection to Constraining Executive Power
Only Congress can constrain an autocratic President, and only electoral reform will enable Congress to perform this essential constitutional function.
This essay was written for a recent academic conference devoted to the topic of how to constrain the Executive Branch of the federal government so that it complies with the Constitution and the rule of law. I am posting it here because, based on all that I have read and heard (including for and at this conference), it is my belief that America risks an unconstrained Chief Executive—and thus the demise of the Constitution and the rule of law—unless it adopts whatever electoral reform is necessary to enable Congress to operate as an effective check on abuses of presidential power. If you are concerned about this risk and believe this essay offers a constructive analysis of how best to mitigate this risk, please share it with others.
It has dawned on most of us that the courts cannot constrain a lawless president without the backing of Congress. As is well-known by now, courts have no power to enforce their orders against Executive officials if the president refuses to cooperate. The marshals, upon whom the courts rely for the enforcement of their orders, are within the Executive branch, and the president can insist upon their noncompliance with judicial decrees. Holding Executive officials in criminal contempt of court, which two judges are currently contemplating, is a constitutionally dubious proposition if the Department of Justice refuses to prosecute (as surely this DOJ will).
Within our constitutional system, only Congress has the power to force a rogue president back in line. If Congress were prepared to use it, the impeachment power would be an effective tool. A conviction in the Senate upon impeachment by the House carries with the penalty of removal from office, which would immediately neutralize the presidential lawlessness. (In the case of Trump, it would not be necessary for the Senate to take a second vote to disqualify him from returning to the presidency, since he would already be ineligible for a third term because of the Twenty-Second Amendment. But a Senate willing to convict surely would be willing to take the second disqualification vote, which would remove any doubt about Trump’s eligibility for any federal office.) One must briefly contemplate the possibility of a president refusing to leave office after being convicted by the Senate, but surely the military has not become so corrupted that it would treat the convicted ex-president as the incumbent commander-in-chief. Instead, recognizing its loyalty to the Constitution itself, the military would begin to obey the new president. In any event, if the military sided with the convicted ex-president against the clear command of the Constitution, we no longer would be operating under our existing constitutional system, and instead would need to acknowledge its replacement with an entirely new autocratic regime.
Of course, however, the otherwise efficacious impeachment power is useless if Congress is unwilling to exercise it. Right now, there is no evidence that either chamber of Congress would be willing to employ this authority on the ground that the president defied a direct order of the Supreme Court. Although it takes only a majority in the House to impeach, and although the Republicans currently have only a seven-seat margin in the House, can one identify even four House members of that party who would be willing to vote with Democrats to impeach President Trump (even assuming they could employ a discharge petition to overcome the inevitable obstacles that would be imposed by super-Trump loyalist Speaker “MAGA-Mike” Johnson)? In the Senate, as we all know too well, it takes a two-thirds vote to convict. Given that not enough GOP Senators were willing to convict Trump because of his efforts to overturn his electoral defeat in 2020—only seven Republicans joined all 50 Democrats, falling ten votes short of the sixty-seven necessary—and because the Senate has become even more “MAGA-fied” since then (Richard Burr of North Carolina, for example, who voted to convict Trump was replaced by Tedd Budd, one of the leading election deniers in 2020), there is no reason to think it would be possible to summon sixty-seven votes to convict in the current Senate even if miraculously a majority of the House would impeach.
Here is where the electoral connection comes in. If procedures for elections to Congress worked as they should, both the House and the Senate would be willing to exercise their impeachment powers, and these powers would operate as an effective constraint against Trump. The American people do not support the idea of a lawless president who defies court orders, and members of Congress elected by a system in which the winners accurately represented what their constituents want would act in accordance with those preferences.
The problem, however, is that the current electoral system does not produce outcomes (officeholders) that match inputs (the collective preferences of voters). This is what election law scholar Nick Stephanopoulos calls “misalignment” and correctly identifies as the serious pathology afflicting America’s democracy. But electoral misalignment is more than just a deficit in translating the electorate’s wishes into public policy. As is becoming increasingly apparent every day, it also threatens the continuing existence of the constitutional Republic and the rule of law because it prevents Congress from performing its essential function of keeping the president in check.
The most obvious way that the current electoral system distorts the overall preferences of the people is the role of partisan primaries and, specifically, the fear that congressional incumbents have of being primaried. There is no doubt that Republican members of Congress change their behavior in office because they are afraid that Trump will endorse a primary opponent. As veteran political analyst Charlie Cook told the Wall Street Journal in the aftermath of the 2024 election, “You sit down with Republican senators and one thing that comes through—they are absolutely terrified of a primary challenge.”
This basic truth, which has been documented by political scientists as well as journalists, was vividly displayed during the confirmation process for Trump’s second-term nominees. We know that Senator Joni Ernst did not want to confirm Pete Hegseth for Secretary of Defense, but she could not resist the intense pressure of a threatened primary challenge next year. Likewise, Senator Bill Cassidy of Louisiana, who previously had the courage to vote to convict Trump for attempting to subvert the 2020 election, was unable to oppose Trump’s choice of RFK, Jr. as HHS Secretary despite Cassidy’s obvious misgivings because he is up for reelection in 2024, and he risked being primaried for continued disloyalty to Trump.
Indeed, there is good reason to think that the Senate’s failure to convict Trump for his role in the January 6 insurrection was because too many Republican Senators feared that Trump’s supporters might vote against them in a primary. From the somersault that Senator Lindsay Graham did in the days after the insurrection, we know that his initial instinct was to condemn Trump. But once he felt the backlash of Trump’s base, he did his flip. Likewise, according to available reports, Mitch McConnell, then the leader of Senate Republicans, wanted to convict Trump in the impeachment trial, but he abandoned any effort at whipping the votes of colleagues once it became clear to him that too many of them were unwilling to vote against Trump. Mitt Romney has also said that some Senate Republicans who otherwise wanted Trump convicted refused to vote that way out of fear.
Trump’s supporters are voters—and indeed are specifically constituents of the Senators who “are absolutely terrified” of being primaried by them—and so one might think that, far from contradicting the wishes of their voters, these Senators are doing exactly what their voters want. But that thought would be mistaken. Sure, Trump’s supporters are some of these Senators’ voters, but they are hardly all of them. Trump’s base is now a majority of the voters in the Republican primaries, which is why the Senators are so fearful of them. But Trump’s MAGA base is not the majority of the voters in the general election, even in most states where Republican candidates routinely win a majority of general election voters.
One recent analysis of survey data puts Trump’s solid MAGA supporters at roughly 35% nationally, with 45% solidly anti-Trump, and the remaining 20% up-for-grabs, depending on other factors. This is why Biden was able to win in 2020, while Trump won both before and after. Obviously, the composition of most states is somewhat different than the national average. Trump’s MAGA supporters are a majority of voters in deepest-red Wyoming, for example, which is why Liz Cheney couldn’t win there under any electoral system. But in most states, even bright red ones like Ohio is now, Trump’s MAGA voters are under 50%, and Republicans win statewide offices—including Senate seats—only by picking up the votes of non-MAGA Republicans (and independents), who are not on board with presidential defiance of the judiciary. Thus, insofar as the structure of the electoral system with partisan primaries forces Senators in these states (like John Husted, newly appointed by Ohio’s governor Mike DeWine, who like DeWine is certainly not a MAGA Republican at heart) to kowtow to Trump, this structure definitely distorts the Senators’ conduct in office so that it contradicts what a majority of all their voters want (as distinct from a majority of their primary voters).
The same point applies to House as well as Senate primaries. House elections of course are complicated by the additional problem of gerrymandering, which forces Republicans to be even more beholden to Trump’s base because gerrymandering reduces the competitiveness of House seats in the general election. But even if gerrymandering were eliminated, primaries would still have the same distorting effect in the House as in the Senate, which is not subject to districting and therefore cannot be gerrymandered.
The existing electoral system, with its partisan primaries, not only distorts the behavior of congressional incumbents. It also affects who gets elected to Congress—and does so in a way that contravenes the preferences of the majority of the whole electorate. The easiest way to understand this point is to look one of the few states that doesn’t have partisan primaries and then consider how things would be different if it did. Take Alaska, for example. Its new electoral system has a nonpartisan primary, in which all candidates regardless of party affiliation run against each other, and the four candidates with the most votes in this primary advance to the general election, where a version of Ranked Choice Voting is used to determine the winner. Lisa Murkowski, arguably the only openly anti-MAGA Republicans left in the Senate (unless one also counts Susan Collins), won reelection under this new system despite Trump endorsing another candidate in the primary. But Murkowski could not have won if Alaska had retained the conventional electoral system with partisan primaries that it previously had. The Trump-endorsed candidate almost certainly would have won the primary, blocking Murkowski from appearing on the general election ballot. Although Murkowski managed to win the general election in 2010 with a write-in campaign when she was successfully primaried by a Tea Party challenger and thus blocked from the general election ballot that year, the politics of the state had shifted enough during the next dozen years that it would have been much more difficult for her to win a write-in campaign in 2022.
One can imagine then Murkowski-like candidates who would be in the Senate instead of Trump-supporting Republicans if other red states had Alaska’s type of electoral system. For example, Mitt Romney maybe wouldn’t have retired if he had been able to run again without worrying about a Trump-endorsed primary opponent. Senator Richard Burr also might have decided to run again in 2022 if he had known he could compete against a Trump-endorsed election denier from the House in the general election to win support from all of North Carolina voters, instead of having to beat that opponent among just MAGA-dominated primary voters. A Senate that has more non-MAGA Republicans like Burr, and fewer MAGA ones like his replacement, would be more likely to stand up to Trump in defense of the Constitution and the rule of law.
But it is important to realize that Alaska’s system in its exact details is not a perfect panacea. The particular form of Ranked Choice Voting that Alaska uses to determine the general election winner can fail to elect a non-MAGA Republican whom the majority of general election voters preferred compared to each opponent, and instead will elect a MAGA Republican despite being disfavored by a majority of voters compared to the MAGA opponent. How can this defeat of the majority’s preference happen? It is because of the specific way Alaska’s version of Ranked Choice Voting tabulates the ballots.
The Alaska method is to eliminate the candidate with the fewest first-choice votes and then redistribute the ballots that ranked the eliminated candidate first to whichever other candidate is ranked second on those ballots. This elimination-and-redistribution procedure is repeated until a candidate has accumulated a majority of ballots, or only one candidate remains, who in either case is the winner. This procedure, as sensible as it might seem to some, can eliminate a candidate whom a majority of voters prefer to each opponent.
By contrast, there is an entire family of alternative tabulation methods for Ranked Choice Voting ballots that will always elect a candidate whom a majority of voters prefer to each opponent. These alternative methods are called “Condorcet Voting” after the French mathematician who analyzed this electoral property, and a candidate whom a majority of voters prefer to each opponent is called a “Condorcet Winner.” One of the simplest Condorcet-based versions of Ranked Choice Voting pairs the two candidates with the fewest first-place votes and eliminates whichever of these is preferred by fewer voters. This Condorcet-based method, which some call Bottom-Two Runoff, repeats this alternative elimination procedure until one candidate remains, who wins the election. A Condorcet Winner cannot lose an election that uses the Bottom-Two Runoff method, no matter how many or few first-choice votes the Condorcet Winner has, because the Condorcet Winner by definition will prevail when paired against each opponent to see which of the two is preferred by a majority of voters. The key point to recognize here is being majority-preferred is not determined solely by the number of first-choice votes a candidate has but instead by the relative strength of the candidate against the entire opposition.
To illustrate with examples from Ohio, in the 2022 and 2024 U.S. Senate election a non-MAGA Republican who ran both times, Matt Dolan, would have been the Condorcet Winner each time. He would have been preferred by a majority of Ohio’s general election voters against the Democrat: Tim Ryan in 2022 and Sherrod Brown in 2024. (A majority of Ohio’s general election voters would have preferred any of the top three Republican candidates against the Democrat either time.) Dolan also would have been preferred by a majority of Ohio’s general election voters against either of the two MAGA candidates running each year: JD Vance and Josh Mandel in 2022, Bernie Moreno and Frank LaRose in 2024. (The reason is that Ohio’s Democrats would have preferred Matt Dolan as the least objectionable Republican, and there would have enough of them combined with non-MAGA Republicans and independents to outvote MAGA supporters.)
Despite being a Condorcet Winner, preferred by a majority of voters against each opponent, Matt Dolan couldn’t win in 2022 or 2024 under Ohio’s existing electoral system. He couldn’t survive a MAGA-dominated Republican primary. But crucially, he also couldn’t win if Ohio used the exact version of Alaska’s new electoral system. Dolan would have received fewer first-choice votes than both the leading MAGA opponent and the Democrat, and thus he would have been eliminated by Alaska’s specific tabulation procedure. But he would have won either time with the Bottom Two Runoff tabulation method.
These Ohio examples can be supplemented with many others, from various states, in both Senate and House elections. Given the nature of contemporary polarization in the United States, there will be numerous instances where the two candidates with the most first-choice votes will be a MAGA Trump supporter on the right and a Democrat on the left, while a non-MAGA Republican will come in third in terms of first-choice votes. The Alaska tabulation method will eliminate this non-MAGA Republican, leaving the election to end up a contest between the MAGA candidate and the Democrat, and in a red state the MAGA candidate will win. But the Ranked Choice Voting ballots would show the non-MAGA Republican to be the Condorcet Winner, preferred by a majority over either the MAGA candidate or the Democrat.
James Madison late in life recognized that a candidate who is third in first-choice votes can still be, in his words, “the real preference of the Voters” because, when pitted against “either of the first two” head-to-head, that candidate would “outvote him.” Madison, however, did not realize in time that conducting congressional elections with Condorcet Voting procedures is necessary to populate Congress with enough members willing to constrain an authoritarian president. If the United States used a Condorcet Voting version of Alaska’s electoral system for both its Senate and House elections, the composition of Congress would consist of Democrats and non-MAGA Republicans who would not tolerate Trump’s despotic behavior.
Perhaps the Republic will be fortunate to survive Trump’s onslaught without having a Condorcet Voting electoral system for its congressional elections. If enough non-MAGA Republicans and independents (the roughly twenty percent in the middle) are willing to vote for Democrats in the midterms—or, alternatively, a new party forms to replace the Democrats in order to be competitive in all the red states where candidates branded as Democrat have become unelectable—then assuming the country continues to hold elections under its existing system, it will be able to save itself from Trump’s despotism without electoral reform.
The United States, however, would be much better positioned to protect its constitutional democracy from a tyrant’s attack if either Congress or the states required some form of Condorcet-consistent electoral procedures for congressional elections.
Thanks for this essay. To put your criticism in different words, Instant Runoff Voting pays much more attention to the top of your ranking than to the bottom, and in this sense is only a slight improvement over single-vote plurality. Both systems favor candidates with passionate supporters, ignoring their equally-passionate opponents. Both systems aim to maximize the number of voters extremely happy with the election results, with no concern for minimizing the number of voters extremely unhappy with the election results.
That last point matters, because people extremely unhappy with election results are the ones who turn to violence, or (less extremely) give up on voting. A divisive candidate whom a substantial fraction of the people despise or fear will have a more difficult time governing (and is more likely to do it oppressively) than a consensus, compromise candidate whom most people find tolerable. Indeed, a candidate who was elected through compromise is probably more likely to govern that way, rather than seeking out performative conflict and saying "my way or the highway".
One could go to the opposite extreme, to "single-vote anti-plurality voting", where voters cast a single vote against the candidate they most dislike, and whoever gets the fewest anti-votes wins. I don't think anybody seriously proposes that approach, because in an election with more than a few candidates, it can easily elect a candidate nobody knows anything about, one with no qualifications except the lack of a record.
My standard example to distinguish these is the 4-candidate scenario in which A, B, and C split the first-place votes while D gets all the second-place votes. In single-vote plurality, one of A, B, or C wins by the luck of the draw. In IRV, D is immediately eliminated for getting no first-place votes, and one of A, B, or C wins narrowly depending on which one is the other two's supporters' third-place choice. In Condorcet, D wins because 2/3 of voters prefer D to each of A, B, and C. Anti-plurality will probably elect D, but D could be tied with any of the others.
Another commonly-studied voting system, as you probably know, is the Borda count, in which your first choice in a 4-candidate race gets 3 points, your second choice 2, your third choice 1, and the candidate with the most points wins. In the above scenario, each of A, B, and C gets at most 1-2/3 points per voter while D gets 2 points per voter, so again D wins, as in Condorcet.
Both Borda and Condorcet systems are symmetrical, in that they pay exactly as much attention to upvotes as to downvotes. But where Condorcet only cares how many voters prefer Anne over Bob at all, Borda also cares _how strongly_ they prefer Anne over Bob, as measured by _how far_ Anne is above Bob in a ranking. Where single-vote plurality (and anti-plurality) only ask very little information from the voters, and IRV asks for more but intentionally throws away as much of it as possible, Borda and Condorcet both ask for full rankings but Borda uses slightly more of that information than Condorcet does.
In rare cases, Borda goes farther than Condorcet in favoring consensus candidates over divisive ones. Consider the 3-candidate scenario in which 51% of voters rank A > B > C, while 49% rank B > C > A. A is the Condorcet winner, but 49% of voters really hate A and really like B, while nobody hates B, so the Borda election goes to B.
I'd be interested to hear your thoughts on Condorcet vs. Borda. I suspect we're in agreement that either one would be a huge improvement over single-vote plurality, and a substantial improvement over IRV.