Friday, November 06, 2020

What's Wrong With The Electoral College And What Can Be Done About It?

 by Michael C. Dorf

No, this is not simply a blog post arguing that the U.S. would be a better, more democratic polity if we elected our President via a national popular vote. I believe that. I support a constitutional amendment to make that change. Failing that, I support the National Popular Vote interstate compact as a means of circumventing the Electoral College (EC) as a second-best option for getting there. But I doubt that either change -- constitutional amendment or interstate compact -- will occur unless and until a Republican Presidential candidate clearly wins the popular vote but either loses or nearly loses the EC vote, because only then could there be enough of the state legislative support necessary for the change by either method.

However, that's not what I want to say today. Nor do I simply want to complain that in 2000 and 2016 the clear winner of the popular vote lost the EC and that, pending the outcome of recounts, litigation, and shenanigans, there's still a chance that could happen again this year. The consequences have already been catastrophic because of the substantive results of the 2000 and 2016 Presidential elections. The catastrophe that is Trump needs no elaboration right now. And while George W. Bush was a normal President when measured against the yardstick of Trump, let's not forget that W was a bad President measured against non-psychopaths: he ignored intelligence that might have prevented the 9/11 attacks, launched an unnecessary war in Iraq that caused enormous suffering and spawned what became ISIS, and abandoned Americans to their fate during Hurricane Katrina.

However, I'm not going to lay the blame for disastrous policy and, in Trump's case, the broad sabotaging of our democracy as well, on the EC. Responsibility for the last four years rests with the nearly half of American voters who either supported Trump for the reasons that decent people find him intolerable or were so in favor of low taxes for the wealthy, deregulation, guns, and/or socially conservative policies on civil rights and abortion that they were willing to overlook Trump's manifest flaws. (I blame my fellow Americans much less for choosing Bush in 2000, when he ran as a "compassionate conservative.") Were Trumpism confined to the fringes, then the fairly small boost the EC gives to it would not make much of a difference.

So what's actually wrong with the EC?

That's a stupid question, you might think. What's wrong with the EC is that it sometimes delivers the Presidency to the loser of the election. Yet that framing assumes that the winner of the popular vote deserves to be President. Why? For the sake of argument, let's question the assumption that EC winners/national-popular-vote winners like Bush 2000 and Trump 2016 necessarily would have lost a national popular vote.

Consider an analogy. Suppose that the loser of the World Series scored more total runs than the winner. Would we say that the winning team lost? No, of course not--and not just because the counting mechanism counts a 2-1 win as equal to a 10-0 win. We also can't say with any confidence that the team that scored more total runs would have scored more total runs if the Series were played under different rules. Maybe the ultimately Series-losing team ran up the score in winning Game 6 because the manager of the ultimately Series-winning team rested his best relief pitchers for the decisive Game 7 when he concluded Game 6 was likely out of reach. Under a total-runs rule, however, the manager would have proceeded very differently.

Politics is much like sports. Modern Presidential campaigns aim to capture the EC. Elections produce a popular vote winner, true, but only as a side-effect of campaigns that try to win the EC. If there are competitive down-ballot races in California, that will drive up turnout and thus Democratic votes in California, including for President, but there's no reason for a Republican Presidential candidate to spend any time or money futilely trying to win extra votes in California. If there were, Republican Presidential candidates might do better in the national popular vote. I believe we would probably still see a modest benefit to the GOP given current alignments, because there are parallel countervailing effects for Democrats. Still, given that popular vote totals are currently an artifact of the EC, we can't say with great confidence how similar or different they would be in a system with a national popular vote.

Thus, it's at least possible that the EC is harmless in the sense that it nearly always produces the same  Presidents with which we would end up in a system in which candidates campaigned for and were chosen based on the national popular vote. But harmless doesn't mean helpful. If likelihood of producing the president who would have won under a national popular vote is the measure of the EC, then of course we'd be better off with a national popular vote.

Should that be the measure? Many people who hold power in a broadly democratic government are chosen through means other than direct election by vote of the People on an equal-population basis. So why not choose the President of a federal rather than strictly unitary system of government through a mechanism like the EC, that gives some weight to the states?

The line of reasoning that underwrites that rhetorical question could justify some system other than direct popular vote for a ceremonial head of state. However, the Presidency of the U.S. today is something else entirely. Several factors have turned the U.S. Presidency into an extremely powerful institution: the growth in federal power over the last century; broad Congressional delegation to the executive branch; and especially recently, aggressive assertions of executive power that go unchecked by a gridlocked Congress. The fact that a modern President exercises such enormous power at home and abroad on behalf of the People cries out for plebiscitary legitimation. Even if the EC almost never produced a President who lost the national popular vote, the failure to have a national popular election mismatches the modern institution of the Presidency to the selection process.

Meanwhile, the EC distorts campaigns. As the NPV website notes, battleground states receive a very disproportionate share of attention from the candidates, which must necessarily distort not just the campaigns but policy. Just as the first-in-the-nation status of the Iowa Caucuses leads to ethanol subsidies, so we expect Presidents and those with Presidential aspirations to pay particular attention to the wish lists of the folks lucky to inhabit swing states.

And that's not even the worst effect of the EC. The worst effect, playing out now as it did in 2000, is that it turns relatively close but clear-cut elections into photo finishes that in the current climate of intense polarization threaten to lead to very serious harm. Because he is a narcissistic liar, Donald Trump of course claimed that millions of votes for Hillary Clinton were fraudulent so that he won even the popular vote in 2016, but no serious person paid attention because it was a lunatic conspiracy theory that assumed that Democrats were clever enough to orchestrate a gigantic but undetected scam but too stupid to target it to the states where it would make a difference. I haven't been following the latest rantings of Trump, so I don't know whether he's also now claiming that he won the real 2020 popular vote, but my point is that a photo finish in the national popular vote in a country as large as the U.S. is much less likely than a photo finish in Georgia or Florida. Trump will implausibly claim victory under all circumstances. However, even a normal human candidate can in good faith throw the system into chaos in a closely contested state that could make the difference between an EC victory and an EC defeat.

The EC is thus terrible in principle and practice. Yet for the moment we're stuck with it. Can anything be done about that?

Maybe. Congress has the power to mandate timing rules governing the selection of Presidential electors. It has done so in the Electoral Count Act. However, timing rules can only do so much. If Congress could mandate some other uniform rules governing the manner of holding Presidential elections and counting votes, that would be extremely helpful. But Article II says Congress can set only timing rules, so we're out of luck, right?

Maybe not. Article I, Sec. 4 gives Congress the power to override state rules governing the timing and manner of holding Congressional elections. And because Congress specifies the same date for Congressional and Presidential elections, many rules that Congress sets for the manner of holding Congressional elections would necessarily also end up governing Presidential elections. Requirements to use auditable paper ballots, rules about who does and does not get to observe ballot counting, and many other standardized practices for Congressional races would become de facto requirements for Presidential elections and thus greatly boost confidence in results.

But (if Democrats don't win both Georgia Senate runoffs in January) could a polarized Congress agree on such rules? Perhaps. Ex ante, it is unclear which side gains partisan advantage from the sorts of good-government reforms I'm proposing. As a dramatic illustration, note that yesterday Trump's supporters were demanding that ballots be counted in Arizona in the hope that he would make up his deficit there while simultaneously demanding that ballots not be counted in Pennsylvania and Georgia, where they did not want to see absentee votes for Biden erase Trump's temporary lead due to the prior counting of election-day ballots. That's amusing hypocrisy, no doubt, but it also illustrates the larger point that there could be consensus for some reforms when enacted behind a veil of ignorance.

That's a modest proposal, I realize. Elimination of the EC would be much better, but absent a first or second-best solution, one does what one can. And if we manage to make it to January 20 without armed conflict, I'll count it an enormous success that Congress is debating how to ameliorate one of the negative impacts of the EC.

12 comments:

Larry Lennhoff said...

What do you think of the idea of splitting the electoral vote of each state in proportion to the popular vote statewide? Not by congressional districts, which would give further advantage to gerrymandering, but if you won 60% of the vote statewide in a state with 10 electors, then you would get 6 electoral votes and your opponent(s) would get the other 4.

This could be done without a constitutional amendment, on a state level. It would make campaigning in purplish red and purplish blue states more attractive to the minority party. And it would diminish the importance of the very largest states such as California and Texas.

Joe said...

A national voting rights act, however unfortunately watered down to get a few Republican votes in the Senate, is appropriate. The suggestions here, along with updating the preclearance rule and other things should be a bipartisan concern. As Rick Hasen argues, it might be appropriate to push for a Voting Rights Amendment as a long time project.

HuskerDru said...

I've long pondered Larry's proposal, and am inclined to favor it. I believe additional harm caused by the EC is that it tends to suppress voter turnout in such states as mine (Louisiana), where any progressive knows her presidential vote will amount to nothing, and conversely, as California, where the same resignation reigns for conservatives. I personally felt more consequential as a voter during the years I lived in Michigan. So, I'm proposing that eliminating the EC, or allocating electoral votes proportionally, would, likely, tend to increase overall voter engagement and participation.

JS said...

Could Congress eliminate the current winner take all system for awarding Electoral Votes? Perhaps mandate that all 50 states use some form of a district system similar to what Maine & Nebraska currently use?

Joe said...

I guess proportional representation could be more equitable, but since individual states per constitutional rule have to change, there will be a concern about doing it on one's own. And, if it requires a national effort, let's just have a national popular vote.

I question the level of support of a a half-way measure, which in practice will be somewhat complicated with fractions tending to be not as simple as clear tenths.

And, if a state has something like eight electoral votes, I'm not sure how much it would matter. The raw votes might matter to a candidate (let's say a few hundred thousand) than two or three electoral votes. To the degree the small state gets attention.

So, all things considered, maybe it's better, but not by too much, especially to push much the level of effort necessary for a national rule. And, there is a risk of a few states getting it, which would benefit one side unevenly.

egarber said...

At the very least, I say get rid of the 100 EVs granted via Senate representation. That’s what results in votes being treated unequally. Why should a vote in Wyoming carry more weight than one in say, Texas?

Unknown said...

@egarber: One major benefit of the EC, which a straight popular vote cannot provide by definition, is the increased emphasis on issues parochial to individual states. With a straight popular vote, only the most popularly discussed issues nationwide get debated. With the EC, in states which are swing states or states which are tipping point states, and especially in states which are both, the issues which are extremely important to the people in those states which might not be of much (if any) consequence to the rest of the country actually get discussed, unlike the popular vote where these issues never make it to the metaphorical debate table. Since swing states and tipping point states change from election to election, One never knows which states are going to fit into those categories, giving a major self-interest to each state to keep the EC.

Additionally, the "+2" from Senate representation comes from how the EC was originally designed. The original plan was for the Congress to pick the President in a joint session and few of us would object to the +2 if we used that system today. (The Framers decided to create a temporary "shadow Congress" after a two-week break because they were afraid the President would be too beholden to the legislature. However, two centuries of parliamentary theory later has shown that "beholdenness" is not a bug but a feature. But that's a different story.) To object to the "+2" because of the weights of votes in Wyoming versus the weight of votes in Texas, in order to be intellectually consistent, One would have to object to the existence of the uniform representation in Senate altogether.

Plus, since we are talking about a federal system, it is helpful to remember we do not have a single election for President any more than we have a single election for the House, the Speaker, the Senate, and the President Pro Temp. We have 51 separate and simultaneous elections for those who will elect the President, just as we have 435 separate and simultaneous elections for the House (and those who will elect the Speaker) and 33-35 separate and simultaneous elections for the Senate (and those who will elect the President Pro Temp).

egarber said...

<<To object to the "+2" because of the weights of votes in Wyoming versus the weight of votes in Texas, in order to be intellectually consistent, One would have to object to the existence of the uniform representation in Senate altogether.

I don't agree. It's one thing for the great compromise to give sovereign states equal representation in Congress as a structural construct. It is quite another to literally treat human votes transactionally as unequal. The latter directly violates a basic right, whereas the former simply ensures small sovereign entities have an equal say within American federalism.

Therefore, to me an election by the people based on population is perfectly compatible with a legislative structure based on whatever makes sense to ensure robust federalism. The question therefore to me: is there some kind of compelling need to treat VOTERS differently amid all this? IMO, the answer is easily "no". Voting is THE basic right. So I say we dump the 100 to protect individuals, while retaining state by state electoral counts with 438*.

*Sure, that messes up Nate Silver's branding, but progress doesn't come for free. :)

egarber said...

<<To object to the "+2" because of the weights of votes in Wyoming versus the weight of votes in Texas, in order to be intellectually consistent, One would have to object to the existence of the uniform representation in Senate altogether.


To go one step further, the corollary to what you seem to be saying is that we really couldn't have a national popular vote without also eliminating equal representation in the Senate. That's like saying a doctor can't operate on somebody's foot without also say, changing something about a patient's left arm.

Joe said...

"One major benefit of the EC, which a straight popular vote cannot provide by definition, is the increased emphasis on issues parochial to individual states. With a straight popular vote, only the most popularly discussed issues nationwide get debated."

It is clarified that what is debated will be the issues that concern the "swing" states. What special issues are involved there that is not also an issue in a range of other states? How much does that even arise? The candidates will still generally come in "red" and "blue" blocs.

Plus, this is skewered by winner take all. People and regions of CA, TX and NY (probably Florida -- it is at best a tease as a swing) etc. overlap with the interests of swing states. Plus, world-wide, somehow, issue are debated w/o it. So, what is the net value here where we are still waiting for an official announcement even though Biden has four million plus more votes?

"Additionally, the "+2" from Senate representation comes from how the EC was originally designed."

So? A range of things -- slavery, lack of modern communication to unite a country, a much stronger general fear of democracy, some belief that electors will be independent etc. -- was factored in there. We changed a lot since then, even amending the Constitution to change how we pick senators.

And, no, one need not be in for an ounce, in for a pound here. One might. James Madison, e.g., was strongly opposed to not apportioning the Senate by population. That is the rule for state legislatures by "one person, one vote" rules. There is a lot to that. But, the Constitution as a whole is a compromise. So, a middle path could be taken there. Federalism would still be upheld either way. etc.

Joe said...
This comment has been removed by the author.
Joe said...

There is a line, perhaps a bit exaggerated, that after the Civil War, we moved from the United States "are" to the United States "is" & that is something too.

The original Constitution worked with a reality of a more isolated and divided nation. The move to 13 delegations with one vote (Articles of Confederation) to the House and Senate (even senators vote separately, not in a bloc) was a big one there alone. Plus, the population differences were much smaller. Even rights were largely seen as tied to the states. Cf. The Fourteenth Amendment with nationally enforced rights writ large.

Anyway, to get back to the original post, the move proposed to tweak updates things to some degree w/o totally changing the system. Any solution there is likely not to be pure. Even Ivory Soap has a bit of impurity.