An unpopular vote

I write to explain my recent vote today against the proposal for a National
Popular Vote (NPV).

I voted in the losing minority in a lop-sided 116 to 37 vote.  The NPV had
the support of  Common Cause, the Speaker, many other leading Democrats and
all of the constituents who wrote to me about the issue.  In fact, at one
point, it had my support:  I signed on as a co-sponsor to the proposal.

I developed doubts after hearing the first day of debate about it last
week.  During a one week postponement of the vote, I consulted experts in
favor of the proposal.  My consultations clarified the issues, but did not
relieve my doubts.

In the end, I concluded that, whatever its merits in principle, the proposal
creates unacceptable down-side risks for our country, risks that are not
outweighed by its alleged benefits.

What is the NPV proposal? The proposal would change the way we select the
President of the United States — a technical issue, but one of great

In November, we will each cast a vote for president of the United States.
The votes will be counted in each state.  In each state, a winner will be
identified and the state will send electors to the electoral college to vote
for that winner.  In order to be elected president, a candidate must achieve
a majority of votes in the electoral college (270 of 538).

Each state gets a number of electors equal to the number of senators that it
has (two) plus the number of representatives in Congress (proportional to
population).  In effect, this formula places the small-state to large-state
balance of power in choosing a President mathematically mid-way between the
balance in the House and the balance in the Senate.

The requirement of winning with an absolute majority in the electoral
college forces presidential candidates to assure themselves a geographically
broad victory.

If no candidate achieves an absolute majority, the election is decided in
the House of Representatives in a special vote in which each state
delegation gets one vote.  This one-vote rule further advantages smaller
states.   An absolute majority must be achieved.  Only the top three vote
getters in the Electoral College may be considered.

Under the NPV proposal, states would enter an interstate compact to the
effect that they would instruct their electors in the electoral college to
vote for whomever gets the most popular votes nationwide.  This does not
require a constitutional amendment because the constitution does not control
how the states choose electors.

If states controlling a majority of the electors in the electoral college
enter the compact, then the compact would take effect.  The members of the
compact would in the next election instruct their electors to vote for the
candidate who won the national popular vote, regardless of whether that
candidate won a majority among those states.  Since the members are a
majority of the electoral college votes, the national popular vote winner
would receive the constitutionally required majority in the electoral

By its terms, states could withdraw from the compact at any time except
during the six months preceding the inauguration date for the president —
between July 20 and January 20 every four years.

The NPV increases two kinds of serious risk.

I am frightened by the possibility of a deviant regional candidate – a
George Wallace – under the NPV.  The NPV lacks the requirement of geographic
distribution that the electoral college enforces.  In a three way race, a
George Wallace, could, for example, win 35% of the vote by winning huge
majorities in one region.  If the major party candidates split the balance
of the vote in other regions, the George Wallace would win.  Whereas in the
electoral college system, the George Wallace would win only the electoral
votes from that region and lack sufficient votes to become president.

The NPV lacks even a runoff mechanism, a requirement that if no candidate
wins an absolute majority the top two candidates compete in a runoff
election.   Few major democracies that elect a President nationally do so
without a runoff mechanism.  But a runoff is not part of the NPV compact.
The compact is designed to be implemented without the necessity of a
constitutional amendment and a creating runoff election would require a
constitutional amendment.

Which brings me to my second concern:   An interstate compact just is not a
robust way to build a system as important as the presidential election
system.  If a state didn’t like the way the NPV system seemed to be headed,
they could withdraw even in June of the election year.  That’s awfully late
in our long cycle and could throw the whole race into confusion.  Moreover,
if a state decided to withdraw even closer to the election or after the
election, the election would end up in the courts.   It is not a settled
question how and whether the Supreme Court would enforce an interstate
compact of this nature, although interstate compacts have been enforced for
more mundane issues.

Once I understood these risks, I reexamined the arguments in favor of the
NPV.  There are three common arguments made in favor of the National Popular

First, some are offended by the possibility that a candidate who did not
actually obtain a majority of the national popular vote could win the
presidency in the Electoral College.  George Bush did this in 2000 and John
Kerry, if he had won Ohio, would have done it in 2004.

In closely divided elections this is a possibility in our system for two
reasons:  It could sometimes happen that by winning smaller states who carry
a slightly disproportionate number of delegates a candidate could gain an
advantage in the electoral college.  More importantly, lop-side majorities
in some states swell the national popular vote of a candidate, but a razor
thin margin is enough to win all of a state’s delegates.  So, a candidate
who won many states by razor thin margins but lost a few states very big
could win a majority in the electoral college while losing the popular

These possibilities follow from our identity as a union of states.   It is
the states that elect the president.  Generally, this basic idea checks the
power of the federal government and helps to protect local liberty and
freedom to experiment.  A president who claimed to be elected by the people,
as opposed to the states, might be more likely (even than our current
overreaching President) to expand presidential power.

Second, because many states are predictable in how they vote, candidates pay
much of their attention to “swing states.”  Some have gone so far as to
suggest that this is the cause for apathy and low voter turnout across the
country.  However, if low turnout were caused by the electoral college
system, then in true blue Massachusetts, the presidential election would be
a relatively low turnout election.  But the opposite is true, people turnout
in droves to cast their vote for the presidency even if the outcome in
Massachusetts is a foregone conclusion – it is the local elections that have
the lowest turnout.

Third, some suggest that the electoral college system creates incentives for
fraud and abuse in the swing states – alleged cases in point, Florida in
2000, Ohio in 2004.  This is true, but arguably, in a national popular vote,
we could see fraud (or allegations of fraud) anywhere.  Political scientists
feel that fraud is less likely in larger jurisdictions – the sheer size of
the margins makes it harder to successfully commit fraud.  However, that
still leaves the possibility in an NPV system of greater confusion about a
close election’s outcome as allegations and real incidents occur in multiple

I gave careful consideration to voting for the NPV just to keep the
conversation alive about reform of the electoral college, but all-in-all I
really feel that — lacking a runoff mechanism — the NPV proposal is not
sound enough to make a positive statement for change.

Nor, honestly, am I clear in my mind that the electoral college – which
embodies our national history as a union of states – is the root of
political evil.  Certainly the concept of sending electors to a college is
anachronistic, but the formula for the college embodies a balance of power
between large and small states.

I am less interested in changing the mechanics and more interested in
developing and supporting national candidates who combine truthful vision
with the skills to engage to a broad range of people across the nation.

Published by Will Brownsberger

Will Brownsberger is State Senator from the Second Suffolk and Middlesex District.

9 replies on “An unpopular vote”

  1. Hi Will,

    Common Cause tells me the NPV vote is up again- has your thinking changed?


  2. “However, if low turnout were caused by the electoral college
    system, then in true blue Massachusetts, the presidential election would be
    a relatively low turnout election. But the opposite is true, people turnout
    in droves to cast their vote for the presidency even if the outcome in
    Massachusetts is a foregone conclusion…”

    Ah, but the data contradicts your argument, Will. There are, in fact, lower turnouts in non-competitive “spectator” states. In the 2000 and 2004 elections, while the ten most competitive battleground states saw turnout increase by 9% (to 63%), the ten “safest” states like Massachusetts saw voter turnout increase a mere 1.7% (to 53%).

    For the youth vote (age 18-24), the picture is even more bleak. From 1972 to 2004, youth voter turnout across the nation dropped by 5%. But in “spectator” states, youth voter turnout plummeted by a range of 12 to 19%. In “true blue” Massachusetts, turnout shrunk by 14%, nearly triple the national fall-off. Overall, there is a turnout gap of 17% between the 10 most competitive states versus the 10 least competitive states for this age group.

    This data can be found in FairVote’s report “Presidential Election Inequality” (

    FairVote had also conducted a study in 2004 that found that over half of all Presidential campaign resources were spent on THREE states — Florida, Ohio, and Pennsylvania. ( Also, a full 18 states received zero candidate visits and no TV advertising.

    Above, you state that “the NPV lacks the requirement of geographic
    distribution that the electoral college enforces.” How can the outright exclusion of 18 states possibly support the notion that our current system enforces geographic distribution?

    And one last question — how many of your constituents called in supporting National Popular Vote, versus those opposing?

    Thank you, Will, for your continued interest in this important proposal.

      1. Hi Will,

        If you have the time, I’d appreciate you responding to my counterarguments, or at least acknowledging them on their merits. In particular, my retrieving data that contradict your “droves” comment as well as my pointing out that the current electoral college-based system does not promote geographical breadth, given that 18 states are excluded from any and all official presidential campaign activity.

        I want to dispatch with two of your feared risks, and perhaps you will once again see that the benefits of NPV outweigh these perceived risks:

        1) NPV could help a deviant regional “George Wallace” candidate

        Will — what would be the difference whether a candidate wins 35% of the popular vote from, say, the American South, or if that candidate wins a plurality of electoral votes from the American South (with the other candidates splitting the remaining states’ votes)? How would having a popular vote versus an electoral vote make any difference if a candidate can win a bloc of states because he/she is regionally popular? I don’t believe there is a difference. The problem of the 35% is only a problem because there is no runoff mechanism. Which is your associated argument:

        2) NPV does not include a runoff component

        Will, the current system does not have a runoff component either. This fatal flaw of the “spoiler effect” applies to either system. Therefore, why do you single out NPV as having the flaw? It is neither an improvement or a degradation of our current system, because the runoff issue is another issue entirely, albeit an important one. We already suffer under the spoiler effect, even with our winner-take-all system, because a candidate can benefit from a third candidate siphoning off votes from an opponent on a state-by-state popular vote basis, and therefore win these states. You may have heard the argument that Perot’s presence tipped Bill Clinton victory in 1992:
        Bill Clinton/Al Gore (D) – 43.0%
        George H. W. Bush/Dan Quayle (R) (Inc.) – 37.4%
        Ross Perot/James Stockdale (I) – 18.9%

        Finally, I want to address your worry that NPV is not a robust-enough mechanism for making this change to the Presidential election system. I want you to look at NPV as a strategy as well as a policy. Obviously, we proponents would like to see NPV passed as a constitutional amendment, but as Dick had posted earlier, it is next to impossible to make this a reality. Therefore, the interstate compact approach is a surefire way to put this issue on the map, in the political sense. When we get enough states passing NPV bills, we hope that the buzz will “trickle up” to Congress, which would then “complete the circuit” by bypassing the antiquated electoral college and bringing us to a vote for president fitting for a democracy.

        I hope my points in this post and my previous post factor into your thinking on NPV, and that you are open to reconsidering your position.

        Either way, it is a pleasure to debate you.

        As an aside, I recently published an Op-Ed on the spoiler effect which you might be interested in reading:

        1. Thanks, Adam, for these thoughts.

          I think the concerns about participation levels in non-battleground states have some validity, but
          I think we have to focus most closely on the things that could go badly wrong in a system. That was what the founders were most concerned about in creating checks and balances and trying to protect against the emergence of unbalanced factions. (See generally the federalist papers.)

          Our system is designed to produce fairly bland, broadly appealing candidates. That has a downside, but it is safer than a system that might result in the election of a dangerous marginal candidate.

          You are right that it is possible to have a spoiler driven outcome in the electoral college, but the likelihood of that is lower because of the absolute majority requirement which the NPV is missing. Additionally, it is hard to simultaneously have both a non-majority candidate and regional candidate in the electoral college. When one does have a regional candidate in the electoral college is a sign of something deeply wrong in marriage of the country (as in Lincoln’s regional election immediately before the Civil War).

          Focus on the downside and the possibility of tragic historical outcomes and avoid those. Don’t worry about relatively ephemeral things like the level of participation in certain states.


  3. This state is sooo corrupt! they change the rules to meet their agenda
    I can see it now, if it looks like the candidate they Don’t want is likely to win the popular vote they will conveiniently drop from compact, once again the corrupt on Beacon Hill are ignoring the voters and making our votes WORTHLESS!!!!

    1. Thanks, Jim.

      I agree that there will be political pressure here and elsewhere to vary the deal if it works out badly. That’s one more reason that I don’t like the NPV. It’s messy and not robust.

      Will B.

  4. After reading your very salient comments here, you have given me reason to rethink my support for the NPV initiative. I will share them with my family and friends which I am sure will promote energetic debate. Thanks for all you do on our behalf Will. Regards,
    Paula Archer

Comments are closed.