Let's Remake the Vice Presidency

 Everyone likes to knock the position of Vice President in the American governmental system. They have two basic powers: they replace the president under certain circumstances, and they can preside over the Senate, but can vote only to break a tie.

In perusing the Federalist papers, it seems that it was considered all-important that the VP be elected in the same manner as the president, in case he would be required to stand in for the president. In fact, that makes sense as a goal (but note that in a worst-case scenario, further presidential succession would violate it).

Furthermore, since the Senate would have so few members in the 1790s, the specter of gridlock due to tie votes was very real. So having a president of the Senate who could vote only to break ties, but not to create a tie, was a way to avoid them altogether (unless the VP declined to vote, of course).

They did not think it at all important that the VP should have a genuine voice in the Senate, or to have any actual job to do on an everyday basis. The experience of the first VP, John Adams, established the futility of a VP attempting to participate fully in the senatorial legislative process. As he put it, the Vice Presidency was “the most insignificant office that ever the invention of man contrived or his imagination conceived”.

Now, with two and a half centuries of hindsight, I think we could do much better (although I am perfectly aware that since so many Americans consider the constitution to be holy writ and replacing it as rank heresy, making major changes such as I will suggest are extremely improbable).

If we were redoing the constitution, there are many even more fundamental changes that have more urgency than fixing the presidency of the Senate, but in this post I consider only that one.

Original version:

The Vice President of the United States shall be President of the Senate, but shall have no Vote, unless they be equally divided.  The Senate shall choose their other Officers, and also a President pro tempore, in the Absence of the Vice President, or when he shall exercise the Office of President of the United States.

New version:

The Vice President of the United States shall be President of the Senate, and shall have one vote on all matters. The Senate shall choose their other Officers, and also a President pro tempore, in the Absence of the Vice President, or when he shall exercise the Office of President of the United States. In the event the Senate be equally divided, the President of the Senate (or the President pro tempore) may cast a second vote.

This is a rather small change to the text, yet it would vastly improve the lot of vice presidents, since they would be active members, rather than largely passive nonmembers of the Senate over which they preside.  In fact, they would be the Senate's most powerful member.  Furthermore, this change would retain the ability to break ties in the Senate, thereby neatly avoiding that particular type of gridlock.

One issue that this change would address is that of determining whether the VP is in the Executive Branch or the Legislative Branch. Under the original constitution, this is vague, since neither the executive or the legislative function is exercised often. Under the revision, the VP would be a full member of the Senate, just as much as any ordinary senator, only rarely standing in for the president. Therefore, the position would be more legislative than executive. Also, presidential elections would surely begin to focus almost as much on the VP as on the president, since the VP would be much more powerful. The election would be about choosing two rather than one president.

Probably the most controversial new vice-presidential power would be that of casting a second, tie-breaking vote. If the ability to avoid gridlock by casting an additional vote is to be maintained, then we have to decide, who will cast that vote? There is no one more obviously suited for this than the VP, since although each of the other senators represent a single state, the VP, like the president, represents every state.

I had originally favored empowering the president to break ties. This would have avoided giving the same individual two votes. However, requiring active presidential participation in the Senate could create time and attention demands that could interfere in a potentially dangerous way with the presidency. The VP would already be a participant, and the power to break ties is a natural extension of the position of President of the Senate.

Automatic Sharing of Seats in Close Elections

I haven't posted here is quite a while, but I had this odd idea and thought I'd reactivate this blog to post it.

The idea was triggered by the observation that in a democracy, elections are often close. This means that almost as many voters favor each option, with only a slight difference among the support. To the extent that our democratically elected representatives are supposed to represent all voters, the frequency of close elections undermines democracy itself.

That is, a voter whose option loses by, say, more than 20% of the votes (60% to 40% in a two-option system) understands that their position is in a true minority, larger than the random variations we often see in elections. However, if the difference is 2% (51%-49%), then the reaction is very different. Except in the sense of the “rules of the game”, there was no practical difference in the degree of representativeness of the two camps.

This principle also applies in three-way or other nonbinary elections: if the winner's vote count is 20% of total vote greater than any of the losers (e.g., 47% to 27% to 27%), once again there is no question about the victory. Furthermore, the generally good solution of doing a run-off between the two highest if there is no majority is also vulnerable to this complaint: imagine 35% to 33% to 32%: is a runoff between 35% and 33% significantly more democratic than between 35% and 32%? I don't think so. The two-person run-off will indeed find a majority winner, but very often up to a third of all voters will abstain in protest, on the grounds that their position is, for all practical purposes, just as representative as the two in the runoff.

Here's another, more democratic but more complicated approach.

First, set a hard criterion for “significant difference” in the electoral outcome. This could anything, from a percentage difference relative to all registered voters, a percentage difference relative to all who voted, even an absolute difference in the number of voters based upon some agreed-upon criteria.

In order for an individual to win an election, they would need to obtain (1) a majority of all votes cast, and (2) a number of votes greater than the criterion different from the next highest number of votes. If this occurred, then the seat would be occupied by that individual. For example, given a criterion of 20%, 60% of the votes would guarantee that candidate a seat. If a candidate received, say, 59% of the vote and the second place candidate received 39% or less, that would also give the seat to the one with 59%. However, if the leading candidate received 59% and the second place candidate received 40% (with the remainder going to lesser candidates), then no individual would occupy the seat.

In the case where no candidate received a significantly greater number of votes than the next highest candidate, the seat would be shared.

A shared seat would operate as follows: the candidate with the greatest number of votes would be the principal occupant of the seat, and would decide who would speak first for the seat during debates or to the public (however, each occupant will also be allowed to speak). However, every action of the seat would require the concurrence of a simple majority of its occupants; otherwise the seat would abstain or take no action.

For presidents, this would mean that before any routine action (e.g., an executive decision, approval of a rule, nomination of a judge, and so on), a majority of the co-occupants would need to agree. And whenever physically possible, each co-occupant is expected to weigh in on every matter—abstentions lower the N of the co-occupants, so would hand over the decison-making power on that issue to the remaining co-occupant(s).

For obvious reasons, the principal occupant of the presidency would have the power to act alone in grave emergencies; however, misusing this power would explicitly be grounds for impeachment and removal; this process could be begun not only by Congress, but also by a voter referendum or by any co-occupant of the presidency. Furthermore, the principal co-president would be required to confer with the other co-president(s), if possible, even in emergent situations.

In Congress, the general principle of majority rule among co-occupants of a seat would apply. For example, in the Senate, if a seat were occupied by two co-occupants who disagreed vehemently with each other, it is likely that that seat would abstain from many votes, perhaps voting only on trivial matters or on the rare occasions of agreement among or between the co-occupants.

That's the basic idea. In effect, when an election is split, the seat is split. We could have two or even three presidents, for example, and the same for every seat in Congress. Obviously, seats occupied by a single individual could be more efficient than split seats. But in the winner-takes-all system, a large number of voters would be represented by someone whose positions they disagree with. The question is whether in a democracy hugely greater representation is better than greater efficiency.

The same principle could be applied to votes on policy, but since there can be no “shared occupancy” in that case, the effect would simply require a threshold difference of a certain size between the winning versus losing position (i.e., a supermajority); a smaller difference would nullify the vote. This would apply to votes in Congress and to voter referenda.