The July 17 of the Weekly Standard has this article, attacking the National Popular Vote. The author is Tara Ross. Her title is, “How to Win the Presidency With 15% of the Vote.”
She assumes that if the National Popular Vote Plan were in force, there might be many strong general election contenders in November for president, so that the vote would be so split up that the first-place winner might have polled only 15% of the vote.
The states of the United States have, together, held thousands of gubernatorial elections. A check of gubernatorial election returns for all states, for the period 1824 to the present, shows that no one has ever won a gubernatorial general election with less than 30% of the vote. When people understand that there is not going to be a run-off, there is a strong tendency for most voters to deduce which candidates are strong enough to potentially win. Then, most voters choose to vote for one of those potential winners. Thus, even when ballot access is completely unhindered, the normal election produces a pattern in which only two candidates receive the lion’s share.
Three-way contests (for offices for which there is only one winner), in which any one of three candidates is perceived to be a potential winner, are not uncommon. But four-way contests, in which any one of four candidates is perceived to be a potential winner, in partisan general elections with no run-off of any type, are extremely rare, not only in the U.S. but in the world. The scenario of winners who only received 15% of the vote does not happen.
Ross is not the first, and will not be the last, defender of the Electoral College to pretend that a national plurality vote would lead to a multiplicity of viable candidates (some of whom, of course, would be “extremists”). For a long time I wondered where they were getting this idea. Then I realized that they make it up out of whole cloth because they have so few real arguments at their disposal.
Ross’s statement about a regional candidate winning the nationwide vote is especially hilarious, for the reasons Richard outlines above. In fact, the real problem is that a regional candidate can, under the right circumstances, get enough electoral votes to deny a majority to any other candidate, sending the election to the House of Representatives. In the House, each state, regardless of size, has one vote — Wyoming’s representative has the same voting power as California’s entire 53-member delegation.
I believe it’s been the case where district winners in some UK districts have won their seats with about 20-25% of the votes. Not sure about Canada but since the UK has so many parties (2 major ones but we must not forget the regional Scottish, Welsh or Northern Irish parties which have a huge say in their respective regions), it’s very possible for a winner (which is determined by first-past-the-post) to not have a mandate.
Louisiana 1995.
I’ve been through the British election returns for each of the last 3 British House of Commons elections, and I don’t recall seeing anyone elected with such a low percentage. Do you have any particular years or districts in mind?
I didn’t consider Louisiana gubernatorial elections 1975 to the present, because those are elections in which the voters knew if no one got 50%, there would be a runoff.
The Louisiana 1975-present elections are the only ones I excluded.
REAL Democracy — versus the EVIL minority rule gerrymander math in picking the Congress and Prez/VP.
Uniform definition of Elector
P.R. and nonpartisan A.V.
Much too difficult for the EVIL lovers of minority rule and the New Age MORONS who love stunt schemes like NPV (which blatantly violates the Equal Protection Clause in 14th Amdt, Sec. 1 — Sorry — one person one vote INSIDE each State).
Highland, Inverness, Nairn and Lochaber 1992.
Why did you start with 1824?
Why did you not consider Democratic primary elections in the South – after all they had much higher voter participation.
What percentage of the gubernatorial elections have a winning candidate with more than 65% of the vote? Might we expect the same for presidential elections if the NPV scheme were implemented?
What would happen under the NPV scheme if there were another election where the popular vote was as close as that of 1880, which was about as close as the Minnesota senatorial race of 2008-2009?
In Minnesota, the election was putatively held under the same conditions, and included the same candidates on the ballot in every county. Ultimately, the issue was whether those common standards had been uniformly applied across Minnesota.
But what would happen if the standards were not even uniform to begin with. Might we not have the Supreme Court of every member State of the compact examining the election results of every non-member of the compact, and perhaps coming do different conclusions?
One election that comes to mind right away where there were 4 strong candidates is the 1860 U.S. election for President. For although Mr. Lincoln won the election he was not on the ballot in every state. Although it is very unlikely, at least theoretically, in view of the great difficulty for third party candidates to get on the ballot in a few states a strong Third Party or Independent campaign could finish very strong and run up the votes in the areas where the candidate was strongest.
Winning few if any states outright that person could win the Electoral votes of a state where the voters could not vote for that person. In 1992, Mr. Perot won nearly 19% of the National vote without carrying a single state. Had he not sabotaged his own campaign by his actions that summer it is reasonable to assume he was capable of winning well over 30% of the National vote and do the same as Governor Ventura later did in Minnesota in 1998. At least as far as the popular vote went. It’s unlikely even with that high a vote count Mr. Perot would have won many more states than Wallace did in 1968.
The 1880 election was the closest popular vote margin in U.S. presidential election history, with only one-tenth of 1% separating the two major party popular vote totals (specifically, .1026%). If the 2008 election had been that close percentagewise, only 134,670 votes nationwide would have separated Obama and McCain.
But, that’s very different from the Minnesota US Senate race in 2008, where the original count showed Coleman and Franken at 41.9881% versus 41.9807%, a margin of .0074%. The number of votes separating them was 215 votes (before the recount). 1,211,590 versus 1,211,375.
The 1860 presidential election had anti-Lincoln fusion slates of candidates for presidential elector in Connecticut, New Jersey, New York, Pennsylvania, and Rhode Island. Texas had an anti-Breckinridge slate (a combined slate in support of Bell and Douglas). So it’s tough to draw any lessons from 1860, because without the electoral college the voting would have been different.
I started with 1824 because it was just easier, because I had Congressional Quarterly’s Guide to U.S. Elections at hand, and it starts with 1824. But a future paper issue of BAN will include gubernatorial elections from 1789, and will show the lowest winner’s percentage in each state.
Another election that comes to mind as well is the one in 1876 which made James Garfield the incumbent in the 1880 race in the first place. If the allocation of electors state by state is to be given “credit” for “saving” us from a protracted post-election tussle in a close election like 1880, then it should also be given “credit” for the fraud, intimidation and violence which was perpetrated in four states to give an EC win to Garfield…after he had lost the popular vote by 51-48% to Tilden. The winner-take all allocation of EC votes state by state (48 of them, anyway) should also be “credited” for the protracted 2000 election in which Gore carried by over 500,000 popular votes, and it should probably also be given credit for Bush’s reelection in which he scored an EC victory only through widespread, systematic electoral fraud in the state of Ohio. Ironically, had it not been for what happened in Ohio, Bush would have lost to a candidate who scored about 3 million fewer popular votes than he.
And yes, I believe that would have been wrong.
The current application of the winner-take-all EC scheme in 48 states provides no special buffer against post-election problems, litigation, delay, etc. It does, however, continue to effectively and systematically disenfranchise millions of American voters. And I believe it also abets electoral fraud in this day and age, in which key states can be identified well in advance, and with great precision. These two factors are, in my opinion, too large a price to pay to avoid the unpleasantness of protracted post-election scuffles, however often or seldom they may occur.
#11 The margin in 1880 was 0.0206% – Presidential Elections Since 1789, Congressional Quarterly.
I have seen credible sources that show the margin almost as large as your. But I’m not sure that we should be showing precision to 10-thousands of one percent (about 9 votes) when we have values that range up to 8000 votes different. Moreover, the 8000 variation is larger than the 1898 vote margin that Congressional Quarterly shows.
But though the national results were close, few statewide results were close. You had a situation similar to Minnesota where there was a wide variation in the margin of victory. During a recount, what would have stopped states like Kansas, Nebraska, Minnesota, Rhode Island, or Vermont, all of which Garfield carried by 60%+ from being liberal in accepting disputed ballots?
#13 Who shot Rutherford Hayes? James Garfield was a member of the commission in 1877 that determined which of the disputed electoral votes were valid.
Would the “popular vote” margin been 500,000 in 2000 if voters in Georgia and North Carolina had been permitted to vote for Ralph Nader? Turnout in the Central Time Zone in Florida may have been suppressed due to the premature erroneous announcement in Florida. What about other States. And would the same candidates have been on the ballot had the election been contested as a popular vote race from the beginning?
Voters in 2000 in Georgia and North Carolina were permitted to vote for Nader. North Carolina wouldn’t count the Nader write-ins because he didn’t file his declaration of write-in candidacy in time, but there was no physical barrier to anyone in North Carolina writing in Nader. And Georgia counted 13,273 write-ins for Nader in 2000.
The 1880 national vote totals were Garfield 4,454,433, Hancock 4,444,976, according to Svend Petersen’s A Statistical History of the American Presidential Elections, the most accurate source for presidential election returns 1824-1960. Petersen collected the vote for each candidate for presidential elector. I don’t believe anyone else ever did that. He was a master.
The Electoral College actually magnifies the effect of fraud. If a party manufactures 60,000 votes for its candidate or disenfranchises 60,000 of another party’s supporters in a close election in Florida, then it can shift 5% of the Electors into its column. If it did the same under a national popular vote system, then it shifts the popular vote in its direction by about 0.05%.
15 –
Ooops! Bad on me. Never type on just one cup of coffee, I guess. (But I’ll bet Garfield would have gladly changed places, considering what happened later).
Jim, I’ve always been amused by the western Florida fantasy. I try to imagine the sound of screeching tires of thousands of pickup trucks as Republican voters in western Florida (most of whom evidently wait until the last hour of polling to vote) tuned in their radios, heard that Gore had “won,” and simultaneously returned home without voting. In the compressed time span of a just few minutes, it must have made an ungodly racket, and one doozey of a traffic jam. And it would have been even worse since Gore voters would have gone home too and…oh wait! Of course all the Gore voters would have continued on to the polls, wouldn’t they, even though they “knew†their candidate had already “won?â€
It’s a fanciful but dubious story.
As for Nader – hasn’t he always told us over and over again that he took as many votes away from Bush as he did Gore? For my own part I wonder how many Nader voters would have voted for either major party candidate then, or in future elections, but Jim – if you think there were potentially 500,000 Nader voters in Georgia and North Carolina, well then maybe I might even convince you after all that Hayes was the incumbent in 1880!
Now….enough talk of popular votes! I’m surprised that a stout defender of the winner-take-all application of the Electoral College system (“WTAECâ€) would even stoop to such talk. Let’s get back to the topic YOU introduced…post-election headaches in close elections.
How did the marvelous, ingenious WTAEC fail to prevent what happened in 1876, in 2000, and the other electoral abuses which we’ve experienced in presidential elections in our history? Since you oppose the NPV scheme because of what “might†happen in close elections it’s a fair question to ask you. How did we end up with these problems under the WTAEC?
Why have not recounts, protracted litigation and fraud been prevented by the (“WTAECâ€)?
Why do you countenance a system which actively encourages fraud (Ohio, 2004, for example)?
And what about the little detail of the millions of “minority†voters in solid blue and solid red states? What about the Republican in Massachusetts who has, effectively, never been able to vote for his or her presidential candidate? Or the Democrat in Idaho? Why don’t those votes matter to you?
Another question to consider…If you truly believe that voters don’t bother to vote at all because they know their candidate “can’t win,†then it logically follows that the WTAEC systematically suppresses the vote in dozens of solid blue and solid red states. Furthermore, if it is true, as you believe, that the proportion of voters dissuaded from voting is greater among the party that is likely to lose, then the WTAEC has the effect of disproportionately suppressing votes (and probably voter registration) among members of, or sympathizers with, minority parties in those states where the statewide presidential winner can be predicted with certainty months in advance. This would consequently put that minority party’s candidates for state and local office at a perpetual disadvantage, making it more likely that the party of majority in the presidential race will also prevail in lower offices, and that the state itself will remain red or blue for the succeeding presidential cycle.
All because of the WTAEC.
And that’s a system you like.
Incredible.
17 –
Precisely. What happened in Ohio in 2004 prior to and on Election Day would not have been worth the effort had the NPV been in effect. With the entire election likely turning on the outcome in that state, the White House was the prize, and every conceivable fraud was indeed perpetrated by a Republican Governor and a Republican Sec’y of State to put that state into the “R” column.
Tables turned, Dems would probably have done the same thing.
A CD (Congressional District) basis of allocating electors would be even worse. Fraud would be even easier and cheaper to perpetrate with even more accurate effect.
The WTAEC system is a quaint but musty, rotted piece of antiquity. It stinks. The one defense you’ll never hear made for it is the only one which makes any sense. It is this:
“Hey…my party has this system racked up right now, so I don’t want to mess with it, OK?”
#16 Nader had 3 times as large a share of the vote in South Carolina as he did in Georgia. You’d think that Atlanta would be more fertile territory for Greens than Spartanburg.
So presumably, Peterson gives individual electoral totals for California (where one Garfield elector and 4 Hancock electors were appointed). Which did he (or you) use in producing your totals of 4,454,433 and 4,444,976? In California, the top Hancock elector receive 0.6985% more popular votes than the elector who finished 6th.
Even if Peterson was careful, he could not account for situations like California in 1912 where it was claimed during the recount that some election judges when counting votes had simply made a tick mark for the head of the elector slate under a presumption that most voters had voted for all Republican or all Democratic electors.
#17 it is much easier to conjure up 1000s of extra votes in areas where you are dominant, rather than try to scrape a few votes up in each precinct.
This may be why the NPV schemers don’t promote 1960 as an instance where the popular vote and electoral vote results were contrary to each other. If the Kennedy-Johnson presidential slate lose “their” 82,000 vote plurality in Alabama, then it calls in question some of the popular vote that the Democratic machines ginned up in Illinois and Texas.
Look at Florida in 2000, where the Gore people concentrated their efforts on counties he had won. If 60% of the vote has clearly punched-out chads for candidate A, then probably 60% of the hanging chads will go the same way.
The lawyers who originally promoted the NPV scheme claimed that the electoral college was sexist, because had the popular vote been used, then the States would have rapidly expanded/debauched the franchise to include non-property owners, women, illiterates, imbeciles, felons, minors, illegal aliens, the deceased, and others to increase the influence of their State.
The popular vote in Colorado doubled between 1892 and 1896 (Colorado extended suffrage to included women in 1893). If we assume that half the votes were cast by women, and that they were in similar proportions to men, then they alone reduced the national plurality by over 10%, in a not very close election, and a not too populous state.
#13 Did Georgia in 2000 report the 13,273 Nader votes on their certificate of ascertainment? They aren’t included in the results maintained by the the Clerk of the House of Representatives.
#15 It is reasonable to assume that Ralph Nader would have received at least as high a percentage of the vote in North Carolina and Georgia as he did in South Carolina. Let’s say 2% (similar to Virginia). That is 100,000+ votes.
19, 20, 21…
Say Jim…since you’re running up hash totals and searching for “reasonableness,” let me ask you this:
Is it “reasonable” to say that the 2,576,630 voters (36.7%) in NY who voted for McCain would have used their time more productively by cleaning their bathrooms than going to vote for a candidate whom everyone on the planet knew had no chance of scoring NY’s electoral votes in the precious EC system?
Or how about the 3,521,164 (43.8%) who voted in Texas for Obama? Wouldn’t they have been better off washing their car that Tuesday?
Or, turning to the small states which are supposedly “protected” by the EC, how about the 157,317 (35.3%) in RI who voted for McCain? Or the 122,485 (38.0%) who voted in Alaska for Obama? What were THEY thinking?
NPV may resolve some problems, but may also cause more. Who knows if they will be worse or better.
24 –
I would agree with the first part of your statement if just slightly differently worded:
“NPV may resolve some problems, but may also cause others.”
I don’t concede that more problems will result. I do agree, however that it might (but not necessarily) usher in problems more difficult to solve than the ones we’ve already dealt with under the current EC system. That said, I believe that any other problems which might result from implementing NPV are worth the price of coming closer to the goal of ensuring that every person’s vote counts as much as every other person’s vote for president. I’d also posit that it will make localized electoral fraud, which has been rife in many of our elections, more difficult to achieve. That is also a worthy goal worth the costs of implementing the NPV, and one which will NEVER be achieved under the current EC system.