Posts © 2011-2012 by Gerald G. Day

Saturday, January 7, 2017

January 7, 2017
Despite its flaws and its anti-democratic character, the electoral college has its defenders. Some claim that it hasn’t messed up as often or as badly as its detractors claim, which is a bit short of justifying its continued existence. An extreme form of that argument is the comment by George Will that, although in the five elections of this century, the winner has lost twice, no big deal: "Two is 40 percent of five elections, which scandalizes only those who make a fetish of simpleminded majoritarianism."[1] Count me among the simpleminded; a 40% failure rate sounds like a lot.
Others argue that it would be unfair to allow all those votes in California to decide the election. One could argue in rebuttal that it was unfair to allow Texas’ 38 electoral votes, awarded by a margin of 9% of its popular vote, to decide the election, which it did last year. Then there is the argument that we shouldn’t allow cities to determine the outcome, but should we award votes to acreage?
Reference to the original design is of little aid, as it no longer describes the process. Present-day electors are not the wise men of Hamilton’s argument in Federalist 68, and political parties, which were unknown and unwanted, now control the system. The present-day electoral system resembles the Hamiltonian concept in that it interferes with popular, national election but, as we have seen this year, it too can bring a demagogue to power.
Republicans support the electoral college because it helps Republicans, just as Democrats attack it for the same reason. Leaving partisanship aside, can it be justified? Here’s a reasoned defense of the modern electoral college by a serious observer, [2] with some comments.
"1) Certainty of Outcome
"A dispute over the outcome of an Electoral College vote is possible—it happened in 2000—but it’s less likely than a dispute over the popular vote. The reason is that the winning candidate’s share of the Electoral College invariably exceeds his share of the popular vote." That makes no sense. The second sentence, while true, doesn’t support the first. I don’t know of any instance of the national popular vote’s being challenged but, in addition to 2000, the electoral vote was challenged in 1876 (and in 2016, but this article predated that election).      
" . . . Because almost all states award electoral votes on a winner-take-all basis, even a very slight plurality in a state creates a landslide electoral-vote victory in that state. A tie in the nationwide electoral vote is possible because the total number of votes—538—is an even number, but it is highly unlikely." If the issue is avoiding ties, then a popular vote would be better, as Judge Posner acknowledges: "
Of course a tie in the number of popular votes in a national election in which tens of millions of votes are cast is even more unlikely." So what is the problem?
"[I]f the difference in the popular vote is small, then if the winner of the popular vote were deemed the winner of the presidential election, candidates would have an incentive to seek a recount in any state (plus the District of Columbia) in which they thought the recount would give them more additional votes than their opponent. The . . . result would be debilitating uncertainty, delay, and conflict—look at the turmoil that a dispute limited to one state, Florida, engendered in 2000." Recounts are demanded under the present system. The turmoil in Florida was because of the electoral vote; whatever the final count in Florida, Gore would have won the national popular vote.
"2) Everyone’s President
"The Electoral College requires a presidential candidate to have transregional appeal. No region (South, Northeast, etc.) has enough electoral votes to elect a president. . . . This is a desirable result because a candidate with only regional appeal is unlikely to be a successful president." It doesn’t seem likely that anyone could win the popular vote with only regional support, if that word is used realistically. Advocates of the electoral system point to the sources of Hillary Clinton’s votes as proof that her support was over-concentrated, but she carried states on the west coast, the mountain west, midwest, northeast and upper south.
"The residents of the other regions are likely to feel disfranchised—to feel that their votes do not count, that the new president will have no regard for their interests, that he really isn’t their president." Leaving aside the reference to regions, that describes millions of 2016 voters.
"3) Swing States
"The winner-take-all method of awarding electoral votes induces the candidates . . . to focus their campaign efforts on the toss-up states . . . . Voters in toss-up states are more likely to pay close attention to the campaign—to really listen to the competing candidates— knowing that they are going to decide the election. They are likely to be the most thoughtful voters, on average (and for the further reason that they will have received the most information and attention from the candidates), and the most thoughtful voters should be the ones to decide the election." His argument is an odd, semi-populist variation on the original principle of the electoral college: the best people should decide. Here it’s the residents of a handful of states; Judge Posner knows better than to praise the electors.
In swing states last year, voters were more divided than most, but evidence of their thoughtfulness is lacking. One could argue as easily that a system which allows a few states to determine the outcome increases the chance that demagoguery will succeed.
"4) Big States
"The Electoral College restores some of the weight in the political balance that large states (by population) lose by virtue of the mal-apportionment of the Senate decreed in the Constitution. This may seem paradoxical, given that electoral votes are weighted in favor of less populous states. . . . But winner-take-all makes a slight increase in the popular vote have a much bigger electoral-vote payoff in a large state than in a small one. . . ." True, but it also results in many votes being meaningless.
What is the benefit? "So, other things being equal, a large state gets more attention from presidential candidates in a campaign than a small states does." That isn’t entirely consistent with the Judge’s theory of attention to swing states. In any case, his model isn’t realistic; safe states are ignored, even if large.
"5) Avoid Run-Off Elections
"The Electoral College avoids the problem of elections in which no candidate receives a majority of the votes cast. . . There is pressure for run-off elections when no candidate wins a majority of the votes cast; that pressure, which would greatly complicate the presidential election process, is reduced by the Electoral College, which invariably produces a clear winner." The electoral vote isn’t always clear, but the tendency of the electoral vote to be decisive and known early offers the best argument for the present system.
There are several solutions to the problem posed. We could declare the winner of a plurality of votes elected, which would produce the same result as the present system in many elections (seven in the Twentieth Century), apart from the illusion of a majority created by electoral votes. It also would avoid electing someone who lost the popular vote. A runoff election would be more decisive but might simply repeat the result by electing the winner of the plurality, and would cause delay. An "instant runoff," in which voters pick first and second choices, would be better.
I’ll contribute another argument for retaining the electoral college: Vote counting is much faster now than in the past, but the exact result would not be known for some time. However, since the beginning of the Twentieth Century, the popular vote margin has fallen below 500,000 only once, so it’s unlikely that the outcome would be in doubt for long. Also, the electoral college doesn’t eliminate delay. Technically, the winner isn’t known until the electors meet, this time on December 19. Hyper technically, the result isn’t known until the electoral vote is approved by Congress, which occurred yesterday.
The 2000 outcome was in doubt until December 12, when the Supreme Court ended the Florida recount. This time, there was some uncertainly due to recounts. Both were the product of the electoral system.
Judge Posner summed up as follows: "Against these reasons to retain the Electoral College the argument that it is undemocratic falls flat." The reasons offered hardly overcome the principle that the person with the most votes should win. That is the principle on which virtually every other election is run.
The present system might be justified by the theory that votes should be by states, that federalism somehow mandates that the leader of a nation be chosen by a coalition of states. However, states, as such, do not vote; we merely allow them, by means of the winner-takes-all rule, to disfranchise a minority of their voters. An intermediate position, taken by Nebraska and Maine, is to award some electoral votes by Congressional District. However, even leaving aside the effect of gerrymandering, any jurisdictional boundary within the country, including those between states, is arbitrary and, in this context, irrelevant. Only people are natural units. When we elect a president, we should give effect to all of their votes.


1. The Washington Post, 12/16/16

Richard Posner, "In Defense of the Electoral College," the_electoral_college.html

Friday, December 30, 2016

December 29, 2016
This year’s vote already has been analyzed many times in an attempt to learn how it came out as it did. Some of those analyses were based on early, incomplete data, and were correspondingly misleading. Exit polls, even assuming them to be reasonably accurate, have been interpreted as disclosing various reasons for the result. Here’s a look at the election based on final numbers,[86] which make sweeping interpretations more difficult. (Comparisons by year are between this election and that of 2012; all numbers pertain to the presidential election. Until noted otherwise, the electoral vote numbers assume that electors voted as they pledged to do so). 
The total national popular vote was 137,053,916, up 6.05% this year. National voter turnout (percentage of eligible voters) was higher, at 58.9%, compared to 58.0%.
Trump won the electoral vote 306 to 232; 270 are required to win. Clinton’s electoral vote was lower than Obama’s by 100, 232 to 332.
Trump won 56.88% of electoral votes to Clinton’s 43.12%; Clinton won 48.043% of popular votes to Trump’s 45.953%. Clinton won more popular votes than Trump by 2,865,075. Clinton did not win a majority, but 54.047% voted against Trump.
Clinton’s national popular vote was lower than Obama’s, 65,844,954 to 65,918,507. Trump’s national popular vote was higher than Romney’s, 62,979,879 to 60,934,407. Clinton’s national popular vote margin was 2.09% compared to Obama’s 3.86%.
Clinton’s popular vote percentage exceeded Obama’s in 12 states plus D.C., and fell below it in 38. Of the 13 gains, 7 were in jurisdictions that she won, all of them Democratic in 2012 and 2008. None of the other 6 brought her close to winning that state. Of the 38 declines, the rate was more than 10% in 8 states, above 15% in 3.
Total votes increased in every blue state except Hawaii, and in every red state except Iowa, Mississippi, Ohio and Wisconsin. Turnout was higher in some blue states, but lower in others. The same was true of red states.
"Other"(third party and independent) votes amounted to 8,229,083 (6.004% of the national total), much increased from 2,384,728 (1.85%) in 2012.
Many states voted as they had in 2012; 24 states with 206 electoral votes went Republican both times; 19 states and D.C. with 232 Democratic. Maine, which gave its 4 electoral votes to Obama, split its votes this time and awarded 1 to Trump. Clinton lost 6 states which Obama won, and won none that he lost. The six losses were crucial; see next item.
The election turned on 7 "battleground states," Florida, Iowa, Michigan, North Carolina, Ohio, Pennsylvania and Wisconsin. Clinton did not win any of them. Obama won all of those states except North Carolina; Clinton lost it by a wider margin. Here are the results:
Pop. VoteMargin ElectoralVotes
Clinton  Obama    Clinton         Obama
North Carolina-3.66%-2.04%00
Those 6 switches and Maine’s 1 gave the election to Trump with votes to spare: 206 + 99 + 1 = 306.
Clinton’s losing margin, expressed in votes, in battleground states and those which proved critical, was as follows:
Florida -112,911
Iowa -147,314
Michigan -10,704-10,704
North Carolina-173,315
Ohio -446,841
Wisconsin -22,748-22,748
Compared to 2012, total votes and turnout in battleground states, expressed in percentages, were as follows:
Change Change Margin
Florida12.11% 2.71%-1.19%
North Carolina5.24%0-3.66%
Related to turnout is the "undervote," the number of ballots containing a choice for various races but not for President. Those voters who skipped the presidential line were more numerous this year in most states. In battleground states for which the information is available, the numbers are as follows:
Florida 1.10%0.50%
Iowa 0.60%0.30%
Michigan 1.00%0.70%
North Carolina0.40%0.60%
Ohio 1.30%0.60%
Wisconsin **
Average rate0.88%0.54%
National average1.10%0.60%
* unavailable
In 4 battleground states, Florida, Michigan, Pennsylvania and Wisconsin, the total number of votes for Stein (Green) and Johnson (Libertarian) exceeded the margin by which Trump beat Clinton. In three states, Michigan, Pennsylvania and Wisconsin, the Stein vote alone exceeded that margin. In one, Florida, Johnson’s votes alone did so. In North Carolina, the total "Other" vote did.
Unemployment rates in battleground states are as follows; in "Rank," the state with the lowest rate would be no.1 :[87] 
Florida 4.80%258.50%38
Iowa 4.10%145.10%5
North Carolina4.90%279.30%47
Ohio 4.90%277.40%25
Wisconsin 4.10%147.00%17
Average rate4.76%7.74%
Average, all states4.71%7.37%
The actual electoral vote was 304 for Trump, 227 for Clinton. Neither candidate gained any votes from the other. Seven votes were thrown away: 1 each to Bernie Sanders, John Kasich, Ron Paul and Faith Spotted Eagle, and 3 to Colin Powell. Ironically, given the attempt to persuade electors to desert Trump, he lost only 2, Clinton 5. Washington had the dubious honor of producing the most defections, 4, the majority of the Clinton defections, 3, and the oddest vote (Faith Spotted Eagle).
This mini-drama did not change the outcome and did not alter the overall or state-by state pattern described above.
As 44 states and D.C., with 438 electoral votes, were in the same column in both elections, much of the result probably was historic party preference rather than a reaction to these candidates.
Counting spending by the presidential campaigns and "single-candidate super PACs," Clinton outspent Trump about 2 to 1. By most accounts, she had a more substantial and better organized campaign but, at least in retrospect, its strategy appears flawed, focusing attention in the wrong places. Her candidacy did not excite people; Trump’s did, and it doesn’t help to point out that it did so for wrong reasons.
Mrs. Clinton did not do well in battleground states. She lost those seven by close to a million votes. Eliminating Ohio, the margin is a half million; omitting North Carolina, which perhaps never was in play, the deficit still is a third of a million.
If we drop Florida and Iowa, the remaining popular vote margin becomes tiny, but the electoral vote result was fatal. A shift of 77,774 votes in the three remaining battleground states (Michigan, Pennsylvania and Wisconsin, totaling 46 electoral votes) would have given the election to her (232+46=278). To win the popular vote by 2,865,075, but lose the election by 77,744 is as clear an indictment of the electoral college as one could want.
It has been argued that voter suppression tipped the balance. There doesn’t seem to be any way to verify that claim, but suppression exists, and provides another reason not to perpetuate the electoral college. Of the critical states, the Wisconsin vote is the most likely to have been affected.
Media coverage and debate questions, emphasizing scandal and ignoring more important issues, tended to help Trump, as did the Comey circus and misinformation roaming around the internet. It was a mistake for Clinton to get into a slanging match with Trump; he’s better at it, and it helped push the media into he-said-she-said mode, but again it’s difficult to know how many voters these factors influenced.
Low turnout has been blamed for Clinton’s loss but, as shown above, national total votes and turnout were up. In battleground states, the only possible correlations between low turnout and the result are in Wisconsin and Michigan, and in the latter, the total vote was up. Undervoting — casting a ballot not including a presidential vote — was up generally and in some battleground states; it could have been significant in Michigan. Lower turnout and increased undervoting could reflect dissatisfaction with the choices offered by the Republicans and Democrats.
"Other" votes were up sharply, clearly reflecting that dissatisfaction. If all Stein voters had voted instead for Clinton in Michigan, Pennsylvania and Wisconsin, where Stein’s votes exceeded the Trump margin, Clinton would have won the national electoral vote. It’s true that anyone not wanting Trump as President should have voted for Clinton, but we can’t change what they did. Any argument that third party votes cost Hillary the election is based on guesses as to what third party voters might have done if they were not third party voters. This isn’t 2000, which turned on the result in one state, in which the margin of victory was 537 votes, and the Green party candidate drew 97,488.
In Florida and North Carolina, the total "Other" vote this year exceeded Trump’s margin — in Florida, Johnson’s alone did — but any theory that Clinton could have won those states depends on moving significant numbers of Johnson’s or other non-Stein votes to her, an even more speculative outcome.
The election was decided in the battleground states, but whether that was due to employment, economics in general, racial resentment, gender or any of the other theories offered is not obvious. Illegal immigration numbers are not relatively high; according to 2014 figures, of those states only Florida exceeds the national average for percentage of the population.[88]   The racial mix varies widely. 
Unemployment is not notably high in those states, except for Pennsylvania, and is significantly lower in each than it was in 2012. However, resentment at the closing of plants, the loss of jobs and stagnant wages may well have been a factor. If so, it is late in appearing, but Mitt Romney would not have been as able as Donald Trump to capitalize on it, and the contrast between Obama’s promotion of TPP and Trump’s supposed opposition to free trade would resonate. It’s possible to argue that a single, specific factor determined the outcome in a given state, for example low turnout in Wisconsin, but none of the measurable factors will reach across enough states to reverse the national result.
A demand for change (not always clear about or to what), a sense of decline, disdain for government and for liberals, even a sense of desertion, seem to have been major themes favoring Trump. He was seen by his fans as an agent for change who had no involvement in government; his party is very much involved, and in a way that has done little good for ordinary people, but very few seem to realize that.
One of Hillary Clinton’s problems was that she represented to some voters the social-issue focus of the left, a policy at times seeming to be a fixation on the minority group of the moment, but to others she represented the financial establishment. Centrism can work, but not if it alienates both flanks. Here, the two reference points coalesced into the "elite," and Trump played to the resentment of that class. He is very much part of the financial elite, and a manipulative player, but his followers didn’t notice that either, or somehow discounted it.
In 1976, Carter won when Democrats still could count on the South, at least with a Southern candidate. That remained true in diminishing degrees for the two Clinton wins in 1992 and 1996. In 2000, Al Gore, from Tennessee, could not carry that state or Clinton’s Arkansas, although he won the national popular vote. The Southern-red pattern applied this year.
Assuming that Republicans will retain the Romney/Trump states and their 206 electoral votes, and that a future Democratic candidate will retain the states won by Clinton this year and their 232, the next Democrat, in order to reach the magic 270, must win 38 votes from the 7 battleground states which were lost. Two of those states, Ohio and Pennsylvania, add up to 38, and several other combinations of two or three produce more than 38. Four of those states were lost by margins of 1.2% or less, and three, totaling 46 votes, by 0.8% or less. The situation Democrats find themselves in is serious, but not hopeless. However, they need to rid themselves of the notion that all they need do is wait for demographic change.
In order to win the next election, Democrats must turn their attention away from identity politics. Concern for civil rights and liberties must remain a major focus, but in a way that cannot be dismissed as "political correctness," as it was by the Trump campaign. It’s tempting to suggest a return to Bill Clinton’s "it’s the economy, stupid," mantra, but his version would be a retreat rightward. Certainly economic issues must be at the forefront, but not in the coziness-with-Wall-Street mode; they must be oriented toward working people, for example by raising the minimum wage, by creating disincentives to moving jobs overseas, and by revised tax policy to lessen inequality. Vote suppression must be ended. On the federal level, those policies must remain aspirations for the time being, but they should be noisy ones. In the states and locally, some progress can be made.
Democrats must find a way to relate positively to small towns and rural areas; it’s difficult to win the presidency based only on the city vote, at least while the electoral college survives. As long as "Democrat" is seen to equate to "urban elite," they will have an uphill battle.


86. Election numbers are taken from;;

87. Numbers for 2016 are for October

2012 numbers are the annual average



Saturday, December 3, 2016

December 3, 2016
Jill Stein, the Green Party candidate, wants a recount in Michigan, Pennsylvania and Wisconsin. It’s difficult to see why. Certainly she will not win the election nationally or in those states, having won small fractions of the vote there, as elsewhere. If a recount moved all three states into the Clinton column, she would win. The odds of that are small, so it’s difficult to imagine that Stein expects that to happen. Even if it could, why would Stein want Clinton to win, after running against her, and making very negative statements about her?[77]  Is it a desperate save-us-from-Trump reaction similar to the plea that electors from Trump states violate their oaths? Is Stein feeling guilty of costing Hillary the election? It’s true that Stein’s votes were more than the Trump margin in Michigan, and Wisconsin and, as of yesterday, in Pennsylvania. Whatever the motivation, the three states Clinton would need are the only states targeted by Stein.
None of that matters, according to Dr. Stein, who claims that this is merely an exercise in good government. "Our effort to recount votes in those states is not intended to help Hillary Clinton. These recounts are part of an election integrity movement to attempt to shine a light on just how untrustworthy the U.S. election system is."[78] Also, her application is disinterested: "The Stein/Baraka campaign is well positioned to lead the effort as election integrity advocates, without a personal conflict of interest in the outcome."[79]  The identification of the campaign with the recount is odd considering this disclosure: "Though Jill Stein was a Green Party presidential candidate, the party did not endorse the recount initiative. Her 2016 vice presidential running mate, Ajamu Baraka, is not a plaintiff."[80]
As to evidence supporting recount, Stein offers this: "We are conducting these recounts because independent election experts have pointed to 'statistical anomalies' in the presidential election results in Michigan, Pennsylvania and Wisconsin. Experts who have studied our voting system for years have concluded many of our voting machines are hackable."[81]  Is that true in only those states? The coincidence is a little hard to accept.
"Whether these machines were hacked by foreign or domestic agents will be determined by using the mechanisms available to us in each state we conduct a recount. Statistical anomalies could arise through other means, as well."[82]  In other words, all they have to go on is speculation or, to put more kindly, suspicion. This is confirmed by an affidavit by Stein’s expert: "One explanation for the results of the 2016 presidential election is that cyberattacks influenced the result. This explanation is plausible, in light of other known cyberattacks intended to affect the outcome of the election; the profound vulnerability of American voting machines to cyberattack; and the fact that a skilled attacker would leave no outwardly visible evidence of an attack other than an unexpected result."
However, in addition to the absence of "outwardly visible evidence," this all seems somewhat aimless; he adds: "Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong, rather than that the election was hacked. But I don’t believe that either one of these seemingly unlikely explanations is overwhelmingly more likely than the other."[83]
There seems to be a long-term political motivation as well; one of Stein’s recount web pages tells us: "Independently funded candidates like Jill Stein cannot stand a chance if our electoral system is rigged in favor of establishment, corporate-funded candidates."[84]  How does "rigged in favor of the establishment" square with hacking, possibly by "foreign agents"?
Another explanation was offered by Greg Palast, based on comments by a lawyer involved in the recount: it isn’t about hacking, but about possible errors in software used to tabulate votes, and about ballots discarded due to alleged irregularities.[85]  However, the latter is contrary to the description of the problem and the evidence on the Stein web sites, which make only a passing reference to discarded ballots, and the recount procedure doesn’t seem oriented toward them. As to the former, Palast conceded that the Stein forces do not have access to the computer codes.
It’s impossible to make sense of all of this. The decision of the Clinton campaign to join in the recounts is less mysterious, but unfortunate, as is not only looks like sore losing, but also allows Trump fans to say, "See, Clinton thinks it’s rigged too. Didn’t her people tell us to accept the results?" However, the response by pro-Trump organizations to the recounts has been a mixture of outrage and panic.
This continues to be a strange election year.

82. Ibid.
83. pennsylvania_but_has_no_evidence_of_fraud.html  
85. recount