Fixes for US Presidential Elections

Copyright 2000 by Hugo S. Cunningham

first posted 001229

letter mailed on 17 Dec 2000

This letter was in response to a column proposing, without much elaboration, that State electoral slates be split evenly for pluralities of less than 1%.

1. letter

To the Editor of the "Boston Globe":

To reduce acrimony in close Presidential races, columnist David L. Chandler proposes splitting the electoral vote in States carried by the winner with less than 1% ("Resolve the mess simply: if vote is this close, split it," Sunday 17 Dec 2000, "Focus," p. C3). For this year, he suggests that if Florida's vote had been split, Al Gore would have gotten a winning 280 electoral votes without any controversy.

This is a promising idea, but Mr. Chandler hasn't thought it all the way through.

This problem is easily remedied, however, by prescribing a graduated scale of reduction for margins of less than 1%. This would be especially easy to do if a Constitutional Amendment [See Appendix 1, paragraphs (1) and (2)(a)] provided for automatic awarding of electoral votes to winning candidates, and for fractional votes. If we assume 6 million votes for Florida, the 25 electoral votes could be awarded as follows:

Bushconverts to
popular vote margin %-age margin electoral margin electoral vote
electoral vote
60,000 1.0% 25.0 25.0 0.0
55,200 0.92% 23.0 24.0 1.0
6,000 0.1 2.5 13.75 11.25
480 0.008 0.2 12.6 12.4

This would practically eliminate the possibility of deciding the Presidential election by two hanging chads. Nevertheless, court decisions awarding 5000 votes one way or the other can still be decisive in a close race [Appendix 2].

To reduce the stakes and acrimony, the newly elected Congress should be empowered by Congressional Amendment [Appendix 1, paragraphs (2)(d) and (3)] to mandate Presidential power-sharing in close races. Rules for power-sharing could either be prescribed in advance, or negotiated after a close election. Today, the most important power for sharing might be judicial appointments, but 40 years ago judicial appointments were uncontroversial.

And finally, although it has nothing to do with close State votes, it is time to throw out the Twelfth Amendment's bizarre mechanism, in case of electoral-college deadlock, to choose the President with a State-by-State caucus of House members. The last time it was used, in 1824, we had four years of bitterness over a "corrupt bargain." The choice of a joint session of Congress would have far more legitimacy [Appendix 1, paragraph (4)].

--Hugo S. Cunningham


Appendix 1. Proposed Constitutional Amendments

Appendix 2: How this year's election might have looked, if electoral votes were reduced for states with less than 1% victory margins:

State full electoral vote winner margin %age reduced electoral vote
WI11Gore0.253 (-8)
NM5Gore0.11 (-4)
OR7Gore0.11 (-6)

Ignoring Florida, this revised tabulation would yield 246 electoral votes for Gore and 246 for Bush. Depending on whether manual recounts are allowed, Florida might give anywhere from 0 votes for Bush (a draw) to 5 votes for Gore (allowing him to win). For the year 2000 election, therefore, it is no obvious improvement. Nevertheless, in conjunction with a graduated power-sharing requirement for narrow electoral college victories, it would sharply reduce the impact and likelihood of cheating on recounts.

Appendix 3: My earlier Usenet post on reducing electoral votes in close States

Note -- as a tentative addition to a long post, it did not evoke much enthusiasm at the time. Also, the reduction of State electoral votes was less attractive than my (and Mr. Chandler's) current proposal of splitting.

From: (Hugo S. Cunningham)
alt.politics.libertarian,alt.politics.democrats.d,, alt.politics.usa.constitution,alt.politics.elections
Subject: Constitutional fixes for Electoral College
Date: Tue, 14 Nov 2000 01:19:06 GMT
Lines: 96
Message-ID: 8upo5l$

[most of post omitted ...]

I am also unsure whether to press the following:

(5) To reduce the temptation to distort recounts in Presidential races, Congress shall have the authority to specify a graduated reduction of a State's electoral vote, for victory margins of less than 1%.


Note 2: Detailed mathematical specifications don't belong in the Constitution.
One approach might be:
Divide 1% by a State's specified number of electoral votes (Congressional districts + 2). Call the resulting %-age "Q".
A State with an exact tie has zero electoral votes. For each fraction of Q that the winning candidate is ahead, give him one electoral vote, up to the maximum for a 1% margin.
Example: Florida has 25 electoral votes.
For Florida, Q = 1% / 25 = 0.04%
If Bush leads Gore 49.04% to 49.00%, he would get 1 electoral vote.
If he leads Gore 49.50% to 49.00%, he would get 13 electoral votes.
If he leads Gore 50.00% to 49.00%, he would get 25 electoral votes.
Note that if only 1 electoral vote rather than 2 x 25 = 50 electoral votes are at stake, the incentives to engage in recount skullduggery are far less.

--Hugo S. Cunningham

I appended a similar post to someone else's thread in Usenet group ne.politics:

From: (Hugo S. Cunningham)
Newsgroups: ne.politics
Subject: Re: No Electoral College? Be careful what you wish for.
Date: Tue, 14 Nov 2000 01:24:14 GMT
Lines: 99
Message-ID: 8upof9$

Go to Hugo S. Cunningham
home page.