(Unpersuasive) Challenges to the National Popular Vote Plan: Part One in a Series of Columns

Now that the 2012 election is in the rear-view mirror and the 2016 election is still somewhat distant on the horizon, this is an appropriate time to return to the question of presidential election reform.  As I have written about many times (including here) on this and other websites, and in academic journals, one important and prominent reform effort, known as the National Popular Vote (NPV) Compact, seeks to move the country in the direction of making it ever more likely that the President who is elected is the candidate who obtains the most voter support nationwide.

Some Key Background of the NPV Concept

The essential idea (elaborated by me, my brother Akhil Amar and, independently also by Professor Robert Bennett over a decade ago) is to get various states to sign onto an agreement that would require each signatory state to cast its electoral college votes not for the candidate who garners a plurality of popular votes in that state, but rather for the candidate who wins the most popular votes nationally. This system, with enough states as signatories, would generally mean that the winner of the Presidential contest would be the person who had won the largest number of votes from individual voters nationwide.  In that way, the plan would ensure that every voter—regardless of the state in which she lives—would have her vote count equally to that of every other voter in the country. Importantly, the agreement, by its own terms, would not go into effect until a sufficient number of states to comprise a majority of the electoral college—that is, states whose electoral college allotments collectively total 270 or more—ratify it.

To date, eight states (Hawaii, Illinois, Maryland, Massachusetts, New Jersey, Washington, Vermont and California) and the District of Columbia—comprising 132 electoral college votes altogether (almost half the needed 270 votes)—have adopted the plan.

Various policy arguments have been raised against the NPV idea, and in favor of the status quo.  More importantly for present purposes, however, in the past few years some analysts have raised constitutional objections to the NPV plan.  In the balance of this column and in a subsequent column, I will address some of these arguments.  Many of the constitutional criticisms that have been advanced relate to the compact aspect of the NPV plan, and focus on whether or not Congress would have to approve the coordinated interstate action before the current NPV plan being adopted were implemented.  I have touched on some of these criticisms before, and will return to others in a later writing.

A Challenge to the Ability of Any State, Whether Acting Alone or in Concert with Others, to Allocate Its Electors With an Eye to the National Electorate

But in the space below, I respond to a different line of constitutional challenge laid out by Willamette Law Professor Norman Williams, who wrote in a 2012 law review article that, putting aside the compact aspect of the NPV plan, it is unconstitutional for a state to pick its electors based on the national (as opposed to in-state) popularity of a candidate because such a decision by a state would violate Article II of the Constitution—the provision that tells each state to appoint its electors “in such manner as the legislature thereof may direct”—and the understanding of Article II embraced by the constitutional “framers.”

Under Professor Williams’ reading of the Constitution, even a state that were not coordinating with other states would be prevented, under Article II, from picking its electors with an eye towards the national vote tally.  (In an earlier law review article, to which I wrote an academic journal response, Professor Williams had suggested, completely unconvincingly to me, that a state that based its electoral college decision on the national electorate would be violating the Equal Protection Clause of the Fourteenth Amendment; this newer article relies instead on Article II and the expectations of the Framers in 1787.)

Professor Williams observes that his reading of Article II “may strike many as counterintuitive.”  That may be true, but that is not necessarily fatal from a constitutional viewpoint.  To my mind, the problem with Professor Williams’s line of argument is not that is it is not intuitive, but rather that it is not strong.

There are many parts and subparts of Professor Williams’s claim, but his thesis can be distilled into two large pieces, each of which I treat in turn.

The Novelty of a State’s Looking to the National Electorate

One thing Professor Williams points out is that, until now, no state has tried to pick its electors based upon national voter popularity.  That is true, but it is also largely beside the point.  Sometimes a device has never been tried because it is unconstitutional, but as even Chief Justice Roberts in the Obamacare ruling made clear, novelty is not itself a constitutional infirmity, because “there is a first time for everything.”  Indeed, most of the aspects of the presidential election process we take for granted today would have been novel for much of America’s history.  The rise of political parties; the near-automatic access that the major parties have to put their candidates on presidential ballots; the development of the “short-form” ballot — by which voters register preferences for actual Presidential candidates, rather than for individual presidential electors (the folks who make up the so-called electoral college); the requirement of many states that electors take pledges to support particular candidates, etc. were all unheard of for generations after the founding.  And yet no one thinks that any of these devices is unconstitutional simply because it was new when introduced.

Moreover, the idea that states would want to look to the national electorate at this point in history in particular is not hard to explain.  The 2000 election was the first modern presidential contest (after the country has internalized the one-person, one-vote ideal of the 1960s redistricting cases) in which the successful president candidate obtained fewer votes nationwide than his opponent.  Moreover, the country is far less divided today geographically (even as it remains sharply divided by party) than it was for most of the nation’s history.  Businesses, families, and culture are much more dispersed throughout the United States.  People move around within the United States (and thus may want their vote to count equally wherever they move) much more today than in the Eighteenth, Nineteenth and early Twentieth centuries.  Political scientists have documented (and, indeed, quantified) much more carefully today than ever before how current presidential elections tend to favor a few “swing” states, such that non-swing states that are largely ignored in the presidential election have selfish reasons today for moving, either individually or by coordinated action, to a national popular vote scheme.  And so forth.  For these reasons, there is nothing constitutionally suspicious about new ideas and innovations in the voting realm generally, or about the timing or origins of this one idea in particular.

The Rejection of a Direct National Election in 1787, the Fear of Majoritarianism Embraced by the Framers, and the Problem of Who the Relevant Framers Are

The other big component of Professor Williams’s argument revolves around the decision in the Philadelphia Convention, at which the Constitution (including Article II) was drafted, to reject a proposal by Connecticut delegate Gouverneur Morris that the President be elected directly by the citizens of the United States.  Because this proposal was voted down, and based on a criticism of the proposal by delegate Charles Pinckney on the ground that it would unduly favor the populous states, Professor Williams argues that the framers conceived of Article II as prohibiting any system that overly focuses on or empowers a national majority.

There are numerous problems with this argument, none of which Professor Williams adequately anticipates or addresses.  First, the evidence he advances for this “expectation” on the part of the “framers” is very thin.  He does not cite any specific discussion in the Philadelphia Convention of the limits on state legislatures to pick electors on whatever basis they choose, and analyzes no evidence from the ratification debates in the states that bears in a specific or even general way on the question.  The “framers” were not simply the men who sat in Philadelphia who had conversations to which no one else was privy; the framers who count more in constitutional interpretation are the folks who read and understood the words that emerged from Philadelphia and decided to make them the “law of the land.”

And the words of Article II on their face give states broad latitude to pick any kind of electors they want.  Surely that would have been the ordinary meaning attached to the phrase “in such manner as the legislature thereof may direct.”  Under any variant of Scalian textualism/originalism, these words—absent some evidence that persons outside Philadelphia would have interpreted them differently—foreclose Professor Williams’s argument.  (To be sure, subsequent amendments to the Constitution, like the First and Fourteenth Amendments, might limit the kinds of criteria that states may take into account in picking electors—prohibiting, e.g., racial and religious discrimination in elector selection—but these limits do not come from Article II, the provision on which Professor Williams relies.)

The Multiple Explanations for, and Consequences of, the Rejection of a Direct National Election at the Founding

Moreover, even focusing on the Philadelphia Convention alone, Professor Williams himself cites to other reasons—besides a deep-seated rejection of national majoritarianism—that explain some delegates’ aversions to Morris’s plan.  Some of those in Philadelphia rejected direct election because the voters (absent modern communications, transportation, or the advent of national parties) would lack adequate information on which to base their votes.  Other delegates expressed the related fear that voters would, lacking adequate information, simply vote for local favorite-son candidates.  Yet others voiced their concern not in terms of populous versus small states, but rather in terms of Northern versus Southern states; direct election would have enabled the North, with more voters, to abolish slavery, which would have been unacceptable to Southern states and have lead them to reject constitutional ratification.

None of these three objections (each of which or all of which might account for the rejection of Morris’s proposal) is relevant today in the context of the NPV plan:  Voters today all across the country have adequate information to pick a President; a state basing its Electoral College selection decisions on the nationwide electorate is not favoring local, native-son, candidates but instead is casting its gaze at a non-parochial level; and (thankfully) slavery is no longer a salient legal or political feature of America.

And even if Professor Williams could show (and he doesn’t) that the biggest force shaping Article II was distrust of majoritarianism, that would still not provide much support his ultimate conclusion.  Article II’s drafters did reject imposing a nationally majoritarian regime on the states, but that is a far cry from saying they foreclosed  such a regime if that is what states (acting individually or collectively) want to do.  There is a parallel here to Article III of the Constitution, which reflects a fear that lower federal courts could end up displacing state courts.  This fear didn’t leave the framers to either require or foreclose the creation of lower federal courts, but instead to leave the matter up to Congress.  At most, Article II does not require or foreclose states from looking to the national electorate, but rather leaves the matter up to the states themselves.

What About Other Moves in the Past Centuries Towards Majoritarianism?

Yet another damning counterargument to Professor Williams’s thesis is that if NPV is too majoritarian to survive under Article II, why aren’t all the other developments in the evolution of modern presidential election mechanics that have occurred over the past two centuries also unconstitutional?  The rise of political parties, the placement of major party nominees on ballots bearing party designations, and the advent of the short-form ballot all moved the country towards a system that facilitates a national popular vote-winner becoming President.  Indeed, by making it less likely that no candidate will lack a majority of electoral college votes on the first ballot, these devices virtually eliminate the prospect of an election being thrown into the House of Representatives, which (because under Article II each state would have one vote in the House) was the most important nonmajoritarian electoral device the framers expected would come into play often.  As a result, all these innovations moved the country away from nonmajoritarianism and in the direction of majoritarianism much more than the NPV plan does, which could be seen as trying merely to narrow an already small window of nonmajoritarianism that remains.

And if Professor Williams’s rejoinder to this is that closing the last open slit in a window is qualitatively different (and worse) than moving it from wide open to almost closed, I would note that the NPV idea doesn’t close the window entirely, for three reasons:  First, absent coordination, a state’s decision to base its electors on the nationwide vote doesn’t come close to guaranteeing that the President will be the national vote winner, so an individual state’s decision to use the national tally as a basis for picking its electors cannot be said to guarantee a nationally majoritarian outcome.  Second, a state that adopts an NPV stance can always change its mind in subsequent elections, so the system isn’t inalterably majoritarian.  And third, since electors in the Electoral College are, strictly speaking, independent and can’t be coerced into voting in any particular way—even the way they themselves have pledged—there is no plan, short of constitutional amendment, that would entirely close the window on the possibility of nonmajoritarianism.

Indeed, the essential implausibility of Professor Williams’s argument can be seen simply by focusing on the independence of the electors.  Imagine that an elector today from some very closely divided state, under the way things currently work, thought that the nation would erupt into civil war if the person who won the nationwide vote tally were to be denied the presidency.  Couldn’t that free-agent elector take into account, under Article II, that specter of war in deciding how to cast his electoral college vote?  And if so, then why can’t a state (and its legislature and people) choose to select electors who are inclined to have that mindset?

All of this brings me to a passage in Professor Williams’s article that may reflect part of the problem with his approach.  He says that each state’s electors must be “accountable to the people of that state” and that the electors must “reflect directly or indirectly the choice of each state’s own electorate.”  The reality, however, is that NPV complies with this requirement.  The people of those states who have joined NPV have (through their legislature, which is accountable to them) made their choice to focus on national voter popularity when they select  their electors.  No one is forcing them to do so; to the contrary, we are respecting their (and no one else’s) desires by respecting their commitment to NPV.  (And the problem, in Professor Williams’s eyes, cannot be that the legislatures are committing to NPV without the ratification of the voters of NPV states, because Professor Williams would not accept NPV any more readily if states joined it by initiative, rather than through their legislatures. And indeed, Article II’s reference to “legislature” may foreclose the initiative device here.)

The bottom line is that the electors from those states who cast their ballot for the nationwide vote winner are completely accountable (to the extent that independent agents are ever accountable to anyone) to the people of those states.  The NPV states aren’t delegating their Electoral College votes to voters outside the state; they have made a policy choice about the substantive intelligible criteria (i.e., national popularity) that they want to use to make their selection of electors. There is nothing in Article II (or elsewhere in the Constitution) that prevents them from making the decision that, in the Twenty-First Century, national voter popularity is a (or perhaps the) crucial factor in worthiness for the office of the President.