Introduction
Well, so much for going back to writing about old roads. I was hoping to put my election file away and literally move on. The “stunning” upset of Tuesday night has forced me to pull out the ‘Election 2016’ file and look over it again. I also reviewed the three articles I wrote about this election season to see what I said, what I think I got right and what I think I got wrong. I believe that my approach to political analysis is useful, more so now than I did three days ago when I wrote my last piece. I am not sure why I want this to be a public event as I am uninterested in competing with anyone, or taking credit or blame for any ‘prediction’ I have made. I am not paid, my readers are not acolytes who must be soothed or inspired, I would prefer to spend my time in the eighteenth century. In truth, however, I was always going to pull that file out regardless of the outcome, so I might as well post my post-election analysis to document the evolution of this project.
The short answer is that I see that my predictions were colored by optimism. Although I wrote of my doubts as I will show, I tried to push them under the rug, which was a mistake. However, I do not believe that the fundamental arguments I am making are in anyway less true than they were three days ago. If anything, I believe the results actually strengthen the case for my methodology. Therefore I hope to go through my arguments piece by piece and review each argument, use the data from the elections to see where things went well and where they did not conform to my hypothesis, and review the comments and criticisms I made about the approach to election prediction of others.
Opinion
It is not my place, or even my interest, to persuade the reader of the merits of the selection we have made. We are all adults and should be able to assess individually what impact the election might have on our lives. However, I do have my opinions, and I would be lying if I said I was neutral or indifferent to the results, so let’s get that out of the way. Trump is a boorish oaf who has been elected primarily because of his very boorishness. He is completely unfit to serve as President of the United States. However, I thought that about Ronald Reagan and George W. Bush as well so nothing new there. Perhaps he will prove to be worse than those very bad presidents. In that event I call upon people to do the right thing, to stand up to tyranny, to assert the rights we have inherited through the arduous efforts of those who came before us, and to stand up to intolerance, bigotry, and cupidity by being politically engaged members of the loyal opposition.
It is my opinion that all three of these presidents are part of a continuum of the same principles and the same agenda, but now the filter has finally been removed from the rhetoric and the real argument is in plain sight. I am a fan of transparency and we can all plainly see what the issues are thanks, ironically, to the impolitic Donald Trump. Those on the right can no longer hide behind flags, bibles, or books by Milton Friedman, William Buckley or Ayn Rand. They lost that privilege when they elected Donald Trump on Tuesday. Race has long been used as a tool to fire up the masses and it was never more apparent than in this most recent election. The fundamental problem in American politics is that racial hatred and opposition to the ‘other’, immigrants, Muslims or Jews, LGBTQ, but especially black Americans, has always been used as a wedge to divide and conquer those who should be united in their opposition to the seizure of power by corporate interests. Everything else is a smokescreen designed to distract us from the steady insinuation of the tentacles of global corporatism into the halls of democracy. Those promised relief from what they believe has been the loss of the American Dream and the usurpation of their privileges by the urban elite and their multicultural mooching accomplices have been snookered by the same old con and will suffer the consequences of their choice.
I have less than zero sympathy for people, particularly in places where the vote is often tighter than it is here in Massachusetts, who did not vote, who ‘protested’ by not voting, or who voted for a Johnson or a Stein and were unhappy with the result. Other people did vote, and they voted for the guy who won, so it is on you. I am a white, male, heterosexual, over-educated descendant of a Mayflower passenger with a house in a booming city of the future. I am probably the least likely to suffer from the potential effects of Trump (and the worst case scenario, nuclear annihilation, does not discriminate, so we are all screwed there). I voted for Hillary Clinton. Do I like her? No. I think she is a wooden candidate with a grating voice and is an uninspiring and mediocre public speaker, I think her entire platform was “I am not Donald Trump” (which should have been enough but clearly was not), I think she never clearly articulated what she would actually do as President to make our lives better, I think she is a centrist (which in my book means Republican-lite), I think she has a hawkish foreign policy and I know she voted to send troops to Iraq, I think she is overly influenced by corporate lobbyists, I think she represents just about everything wrong with politics in Washington. On the other hand, she is not Donald Trump, who represents everything that is wrong with America, who represents everything that is wrong with the world in truth, which is a lot worse. I will be the first to say I told you so when things fall apart.
I hope my biases do not adulterate the analytical work I am trying to present. However,it is inevitable that I will couch the terms of my argument in a manner that will be seen as trying to influence people’s opinions. Truthfully, I am merely attempting to try to test a hypothesis that data can be employed to analyze elections and that the data reflect something about the dynamic of the populations in different geographical areas, be they neighborhoods, or nations. If you are offended by the occasional left-leaning viewpoints that accompany this analysis then you probably do not want to read these entries. I am not interested in engaging in an online insult fest and so any comments that I find offensive or obnoxious will never see the light of day. I also do not aim to turn this project into a polemical mouthpiece for my opinions. I put them here for the record so any body who thinks I have a secret agenda will be disappointed, but I do not wish to bore people with a daily dose of my thoughts on the happenings in Washington D.C.
Analysis: The Electoral College
The easiest way for me to do this analysis is to review statements I made in previous entries. This first entry will review one or two of the more general comments I made about elections before I dive into the weeds more deeply in the next entry. The first one is from the entry Boston Rambles 2016 Presidential Election Analysis:
“Now begins the analysis of the result, the need to find the one single factor that propelled the candidate to victory when all was seemingly lost. Demographic subgroups of all types are brought forth as the linchpin of the election, a new group is crowned as the critical group required to achieve victory, and these voters are assiduously cultivated for four years until the cycle begins again.”
It has begun: Clinton was a flawed candidate, she was too busy wooing Latinos when she should have tried to shore up her vote with the white working class voters of the Rust Belt, she did not have the great turnout operation she claimed she did, her campaign was all hat and no cattle and, inevitably, ‘it’s all the fault of the media’, particularly Nate Silver. Let us dispense with all these unhelpful claims with a few facts. The principal fact we should take from the election is that Hillary Clinton actually won the election, she lost the Electoral College. Voter turnout was quite low in California and yet Clinton still won the popular vote in this election. This leads me to another statement I made in a recent entry
“In fact, I would love the discussion to evolve into a conversation about the elimination of the archaic and unequal electoral college entirely. Then we could have an actual fair and free election where each vote counts equally instead of the current system where massive swathes of the country have no role at all in the outcome of the election. We could stop playing the electoral college map game and candidates would have to visit places where actual voters lived in significant numbers, and not just to pick up a check.”
Two of the last five elections have now been decided not by the will of the people but by the whims of the structure of the Electoral College. I could cite any number of reasons why the Electoral College is a failure, but the main argument is straightforward: it is unequal and gives far greater weight to sparsely populated parts of the country. Here is a map of the likely final results:
Click the map to create your own at The primary problem is the ‘sea of red’ and the misleading it impression it gives. The following contiguous red states have a total of 58 Electoral Votes(EVs) on the map: Utah, Idaho, Montana, Wyoming, North Dakota, South Dakota, Nebraska, Iowa, Kansas, Oklahoma, Arkansas, and Mississippi. This is 3 more EVs than California (55). California has a population as of July 1, 2015 of 39,144,818 inhabitants. These 12 states have a population of 25,698,956 residents. Thus an electoral vote from California represents the opinion of 711,717 people, while an electoral vote from the twelve combined states above represents the opinion of 443,085 people. Voters in this collection of states get three votes for the price of two!! If you live in California, or Texas, or Florida, or New York your vote is worth far less than if you live North Dakota or Wyoming or Alaska. How democratic. This is a problem with the legislative branch of government, specifically the Senate, but here it carries over into the executive branch as well. It is unrepresentative, it leads to the candidates ignoring large segments of the population that are not in ‘swing states’, and it gives the false impression that America is primarily a conservative country. Get rid of the Electoral College! In an age when people move all the time from state to state without any difficulty why should we assign to these random geographical entities disparate value in the power of the vote? The days of the thirteen disparate colonies are long gone. It is time to make sure that one person’s vote is worth the same as another person’s vote.
Incidentally I could have produced a different set of states to show that Texas is poorly represented as well. Nevada and New Mexico are blue, as is Vermont, DC, Delaware, and Rhode Island so this is not strictly a partisan issue. It is partisan in that the urban centers of the country, which hold the majority of the population and are responsible for the majority of the GDP in this country, are dramatically underrepresented by the current structure of government. Urban centers overwhelmingly rejected Donald Trump as I will show later, and now cities and their diverse populations are at the mercy of the representatives of a minority of the population that is primarily white, less well-educated, and rural or suburban.
Urban versus Suburban and Rural voters
From ‘Boston Rambles 2016 Presidential Election Analysis’:
‘Charismatic politicians can have an impact but as a rule they succeed mainly in turning out more of their base than less charismatic candidates, not in changing the dynamic of elections by peeling off voters from the other side. Evidence: Western Pennsylvania and Trump support, which will likely be high. This will be reported as an example of the ability of Trump to pull in voters from the other party, but the fact is that Democrats have been losing voters day by day for more than a decade in Western Pennsylvania as demographic shifts result in an older, more conservative electorate. Trump is not responsible for this change, he is the beneficiary. The number of voters in Western Pennsylvania voting for the Republican candidate in the last three general elections has EXCEEDED the number of registered Republicans every time. This is an indication that some voters are local Democrats and National Republicans, much the same way conservative voters in South Boston typically vote for the Republican in general elections and yet all the local elected officials are Democrats.’
The fact is, that if more voters from cities like Philadelphia and Detroit had shown up on election day, it is far less likely that Michigan or Pennsylvania would have fallen into the Republican column and we might be discussing the poor quality of Trump’s GOTV or the historic nature of the first female President of The United States. Here is some data to support my claim:
Philadelphia population 2015: 1,567,442, representing 12.24% of the population of Pennsylvania (12,802,503). Registered voters as of Monday, November 7, 2016 according to the Pennsylvania Department of State (DOS): 1,102,569, representing 12.64% of the total registered voter pool of Pennsylvania (8,722,938), so slightly higher than the population share-great job registration workers! Number of votes cast by Philadelphia in election of November 8, 2016 according to the unofficial numbers of the Pennsylvania DOS: 680,227, representing 11.49% of the total vote cast in Pennsylvania (5,918,847) (Note the write-ins are not listed yet, a couple of Philadelphia districts have not been reported yet (21 of 1686) and the number has been changing a bit as time goes by, but as of 11:30 am on Thursday November 10, 2016, these are the figures I have). If the voting rate were to equal the share of the population, Philadelphians would have cast at least 725,000 ballots, 45,000 more than they actually did on election day. If 82% of these votes were for Clinton and 15% were for Trump, a reflection of the rest of the vote in Philadelphia, Clinton would add 37,000 votes to her total while Trump would add 7,000. The final unofficial statewide results have Clinton at 2,817,409 and Trump at 2,890,633, a difference of 73,224 votes. The total difference would have been reduced from 73,000 to 43,000 solely by increasing the ratio of voter turnout of the city of Philadelphia to reflect the actual population. Throw in Pittsburgh, and one or two other small cities like Reading, Allentown, Scranton and suddenly the race is an extremely tight contest.
If you think I am overly optimistic about the propensity for people in Philadelphia to vote I submit the following data. Philadelphia vote in 2008 election: 717,329, 11.93% of total vote (6,010,519), eight years ago when Philadelphia had at least 45,000 fewer residents. They still did not reach their share of the electorate in 2008, yet Barack Obama received 595,980 votes in Philadelphia compared to 560,542 votes for Clinton, a difference of more than 35,000 votes. Obama’s margin coming out of Philadelphia was 478,000 compared to Clinton’s margin of 455,124, a difference of 23,000 votes.
Compare that to Westmoreland County, in western Pennsylvania, a county that gave Trump a 56,921 vote differential on Tuesday (116,427 for Trump to 59,506 votes for Clinton). Westmoreland County voters made up 3.07% of the votes cast in Pennsylvania while the population of the county is only 2.80% of the state population. If the county voted at its share of the population they would have cast about 166,000 votes instead of 182,000 and again, the margin of Trump’s victory would have been much reduced. Interestingly, the population of Westmoreland County has dropped by almost 8,000 inhabitants between 2010 and 2015 (from 365,169 to 357,956 inhabitants) as has the number of registered voters (249,140 in 2008 to 245,991 in the latest figures) and yet in 2008 176,873 voters cast ballots, 5,000 fewer than on Tuesday! Where are the voter fraud investigation teams!!! Voter turnout was almost 74% in Westmoreland County against about 62% for Philadelphia. A higher share of total voters in Westmoreland County voted, their preferred candidate won, while a lower share of voters cast ballots in Philadelphia and their leading candidate lost: voting matters. In the election of 2008 turnout in Philadelphia was about 64% and Obama ran up the score by 478,000 votes against McCain in the city, while turnout in Westmoreland County, where Obama lost by 30,000 votes, was 71%.
The charismatic candidate got more of his voters to turn out in each of the above cases. I believe it is a red herring to argue that Clinton should have spent more time wooing white working class voters. They have been abandoning the Democratic party like rats from a sinking ship for decades. Also, they make up a smaller share of the electorate at each election as demographic patterns change. There is a ceiling to that effort even if it were to prove successful in the short term. Better to spend the energy on trying to get voters that are increasingly likely to support the Democratic candidate to turn out, such as the fast-growing Hispanic community, or immigrants who are here to make a better life through hard work and are put off by racist and bigoted rhetoric, or educated young urban voters who are disgruntled because they feel ignored.
These two counties alone provide a remarkable record of the power of turnout. In 2008 the net gap in votes between the Democratic candidate and the Republican candidate was 449,186. in 2016 the gap was 394,203 votes, a difference of 54,983 in favor of the Republican candidate. Trump won the state of Pennsylvania by 73,224 votes, so it is easy to see that turnout differences in only two or three more counties would have been sufficient to cause the pendulum to swing completely from the Republican to the Democratic candidate. This only compounds the problems of the already flawed electoral college by handing ALL the EVs in Pennsylvania to one candidate based on the small differentials in turnout of a few voting blocs in a few counties in one state.
The case of Michigan is even more stark. As of today the unofficial results of the presidential race in Michigan give Trump a lead of 11,837 votes out of almost 4.8 million cast, a 0.25% margin of victory and possession of 18 EVs, coupled with those of Pennsylvania nearly sufficient to win the presidency. In 2016 voters in Wayne County, Michigan, where Detroit is located, cast about 780,000 ballots, of which 517,447 (66.34%) went to Clinton. In 2012, voters in Wayne County cast 40,000 more ballots, about 820,000, and Obama received 595,846 votes. Thus Clinton received 78,000 fewer votes in Wayne County. Had only half of the 40,000 voters who did not show in 2016 voted for the Democratic candidate, Michigan would have swung completely around. Thus, about 20,000 no shows in Detroit cost Hillary Clinton the election in Michigan, regardless of the efforts of the Trump voters in Wayne County, where his vote total was slightly higher than that of Mitt Romney, or in the state as a whole, where his total vote count exceeded Romney’s by 164,000 votes.
The Election is Rigged
I want to make one final point in this article before I end it and resume the next article with more analysis of my own predictions and how they failed to meet expectations. We argue about voter fraud and intimidation, about the overarching themes and events that propelled the election along, about the influence of day to day events, the personality of the candidates, the stakes involved and the massive lurch the country takes every time there is a change of government. The truth is, most of the problems in a typical election are about the lack of interest not too much interest. Small differences in voter turnout alone can have a big impact on the results of a given state. All the EVs in a state are delivered to the candidate that gets one more vote than the other. The entire election can be summarized by saying that in large cities voters do not make up their share of the voting population while in more rural and suburban areas, voters have a higher propensity to turn out.
As we have already seen, the electoral college favors states that have fewer major population centers, disproportionately awarding EVs to these states. Thus, the turnout problem in places like Detroit and Philadelphia area amplifies the already problematic distribution of EVs. America is not represented in the United States government in a ratio that reflects the makeup of the nation. It needs to be addressed or the inequality that exists will get worse over time. Imagine when Los Angeles has double its current population while the population of North Dakota remains static or even shrinks as it has done for most of the last century.
Boston voted for Clinton over Trump by 81% to 14%. Philadelphia voted for Clinton over Trump by 82% to 15%. I could go on, but the point is that New York, Los Angeles, San Francisco, Chicago and EVERY major city in the country voted overwhelmingly in favor of Hillary Clinton. The voters of Wyoming also had their say; in fact they got to say things three times (3.16 times to be precise) for every one time anyone in Massachusetts gave their opinion. Just because you live on a ranch outside Cody or on a farm in Kansas or a small town in Vermont for that matter, does not mean you deserve more of a vote or that you are more American than I am. I live in Boston, Massachusetts, the birthplace of America, and I should have as much representation as any other citizen. If my voice and the voice of my neighbors had been heard equally if my address was here or in Boise, Idaho, it is my opinion that we would not be talking about President Trump but rather President Clinton. That is a fact, not an opinion.
In a final ironic twist, those of you who have done the math will notice that even if Clinton had pulled Michigan and Pennsylvania into the blue column, she would still have lost 270-268. Why ironic? Because Donald Trump took one of Maine’s four EVs, a quirk of the system in Maine, which allocates electoral votes on the basis of congressional district results, which are divided by population. Two EVs are given to the winner of the state, which was Clinton, but Trump took the majority of the vote in the second congressional district of Maine and thus earned a single electoral vote which, had the election gone slightly differently, would have propelled him to the White House. Had the system not been in place in Maine, it would have been a 269-269 tie. In theory one would think that in the case of a tie that perhaps winner of the popular vote would be declared the victor. However, it does not work that way. Instead the election is thrown into the House of Representatives, which is controlled by the Republican Party and would very likely have handed the election to Donald Trump. The system awards one vote to each state and the controlling party of each state’s congressional delegation would decide the vote for each state. As it happens and as is obvious from the map, the smaller states have more Republican delegations while Democratic delegations tend to be found in the larger states. However, had the House delegations been controlled by the Democratic party, they might have voted the opposite way. Either way, the massive failure of the system that would have been obvious to everybody as a result of that eventuality surely would have revived calls for the elimination of the Electoral College. The system used in Maine and one other state, Nebraska, is not perfect, but it is certainly more representative than handing all the electoral votes to one candidate who gets the most votes in the state. Thus, a final indignity of the failure of the Electoral College could have been to have had the candidate who received the lion’s share of the benefit of the problems of the Electoral College benefit from the efforts of one little state to try to address the problem, by avoiding that problem and the likely ensuing debate. Get rid of the Electoral College!
I still have no words.
Permalink