Ricochet is the best place on the internet to discuss the issues of the day, either through commenting on posts or writing your own for our active and dynamic community in a fully moderated environment. In addition, the Ricochet Audio Network offers over 50 original podcasts with new episodes released every day.
Trump Victory in Michigan: Rural Revolt or Urban Apathy?
By Ali Zifan – Own work, CC BY-SA 4.0, Link
Discussing @tkc1101’s prodigious prognosticating abilities with regard to this election on a separate post I asked him if he thought that Trump’s victory was all do to a working class rural revolt or if perhaps he also benefited from urban apathy which saw a slackening of Obama’s Millennial/Minority vote? He could not say, and said he would wait for some further analysis. Curious about my own question I decided to quickly browse through the county results in Pennsylvania from 2012 and 2016. Based on a cursory analysis, my conclusions were that Hillary Clinton in PA did not actually under perform in urban areas as compared to Obama. Rather, her loss came from Trump over performing Mitt Romney in the more rural areas of PA and her under performing Obama in these same area. This indicated to me that for Pennsylvania at least based on a quick skim of the data Trump won because of a Rural Revolt.
With that bit of tantalizing information, I wondered what the results from Wisconsin and Michigan would look like on a by county basis. So, earlier this evening I sat down and decided to have a look. I have only had time to look at the data from Michigan in any detail so I will only be discussing those results. The question I asked was the same one posed for Pennsylvania: Rural Revolt or Urban Apathy?
To help answer this question I looked at the vote total from each county in Michigan from both 2012 and 2016, comparing Trump’s and Hillary’s performance to those of Romney and Obama (respectively). I also tabulated the the results for all third parties effectively treating them as one candidate and comparing their totals from 2012 to those in 2016.
Results:
Donald Trump is currently winning the State of Michigan by 11,837 votes (I believe votes are still being tabulated and have not yet been certified). Total voter turnout in 2016 was only about 60,000 to 70,000 more votes than in 2012.
Donald Trump over-performed Mitt Romney in all but seven of Michigan’s 83 counties, and those seven all tended to be suburban/urban counties based on their larger vote totals. It should be noted though that in Wayne County, which is the county for the City of Detroit, Trump actually over-performed Romney. Comparing 2012 to 2016, the general pattern for Michigan’s rural counties was that Trump’s gains over Romney were slightly less than Clinton’s losses compared to Obama, with third parties making up the difference, usually indicating a net gain of voters for the county. The obvious interpretation of this is that Hillary’s loss was more or less Trump’s gain, with also substantial gains for the Third Party tickets as a whole. In fact most counties gained in total votes cast compared to 2012, which is of course what you would expect, given that the vote total was higher overall.
Hillary Clinton under-performed Obama in all but 6 of 83 counties. Of these counties, four were also ones where Trump under-performed Romney. Trump’s worst county was Kent County which he still managed to win if more narrowly than Mitt Romney had done. Possibly indicating the effects of defecting Republicans. Another of these four of interest is Washtenaw, home of the University of Michigan. Here again, Clinton got a boost possibly indicating defections among Republicans to Hillary. It should be noted that the third parties did well in both of these counties compared to 2012.
Clinton’s problem in Michigan can be summed up in three counties: Wayne, Saginaw, and Genesee, all three of which Obama carried in 2012. She under-performed in all three while Trump over-performed Romney in each; the combination was enough to flip Saginaw from blue to red. Turn out was down in all these counties: In Wayne County alone, by about 40,000 votes compared to 2012. These are all urban/suburban counties, if I am not mistaken. And between them, you have a loss in total votes nearly five times greater than the margin of victory for Trump in the whole state. The lost votes appear to have all been Obama’s.
Third parties over-performed their 2012 totals spectacularly, leaping from 51,000 total votes in 2012 to around 240,000 in 2016. Jill Stein alone went from around 22,000 votes in 2012 to more than 50,000 in 2016. The difference in her performance alone was nearly three times Trump’s margin of victory. Gary Johnson was not on the ballot in 2012, but was a write-in candidate and then only garnered less than 7,774 votes; while in 2016 he got more than 20 times that vote total. Consistently, in every county, Stein and Johnson over-performed their 2012 results. Most notably, they did very well in counties where both Trump and Hillary under performed like Oakland County. They also showed strong gains in the three counties that saw an overall vote decrease compared to 2012 mentioned above.
Conclusions
The evidence of the Rural Revolt is there through all of the low population counties of Michigan, with Trump seemingly turning Democrats into Republicans. But based on all the data presented above, I think it was also necessary for Clinton to suffer from Urban Apathy, as she clearly did in Wayne and Genesee counties. Additionally, third parties’ strong showings probably sapped from Clinton’s votes; in the rural counties, the third party vote total often helped to make up the difference between Clinton’s loss and Trump’s gain, indicating that in these places she bled votes not only to Trump, but also to these other candidates. So, in Michigan the Rural Revolt primed Trump for victory: without it, he would not have narrowly overcome Hillary, but third parties and a significant under performance in urban areas delivered the win. So I would conclude that both Rural Revolt and Urban Apathy are responsible for Trump’s narrow win in Michigan.
Considering 2020, it is clear that Trump will need to keep his rural base, but even a modest improvement in urban turnout for Democrats can hand them a victory, unless Trump finds votes elsewhere. The catch will be what those 240,000 Third Party voters do. Of them we might guess that 50,000 are probably committed to their third party, leaving 90,000 potential voters for both Trump and his Democratic challenger. The question is whether Trump get enough of them to make up for what will surely be an increased urban voter turnout in Michigan in 2020. Also, is there potentially a new hidden vote of suburban Republicans who left their ballot blank for president in 2016, but who will come and vote for Trump in 2020 if he proves successful? Can they make up for the urban increase that should come?
I hope you all found this interesting. I certainly did. If you have questions about specific counties, I can share my results with you about them in the comments. I am also interested to hear from Michiganders about what they saw and think about this analysis. If I can I will do the same analysis with with Wisconsin too, and go and take a look at PA again more thoroughly. But writing this thing took quite a bit of time so it might be slow in coming.
Published in General
She was awful, and it is my understanding that she did not do much campaigning in MI and WI, so who knows how things would have turned out had she tried a bit harder. In FL which got a lot of attention from both candidates you had improvements on both sides. It isn’t like she lacked the resources to run a campaign in all the states. One wonders what is the point of raising a billion dollars (which when I type it I imagine Dr. Evil saying it), and not go for a more general approach to the campaign.
Yeah. I need to spend some more time looking at data, but it’s hard to escape the conclusion that a lot of people who voted for Obama were just plumb turned-off by her.
(<sarc>I can’t imagine why.</sarc>)
Especially when you consider how badly the Democrats have been doing of late, it really seems that Obama was uniquely appealing to voters. As he’s gone, that works in our favor.
Val,
Agreed on all of your points. I would like to see all of the data too. One thing, we all still don’t quite accept that polling will not tell us everything. Voters are individuals with volition and that means they can do what they want and social science can’t know everything before or after.
Regards,
Jim
We can’t completely dismiss policy but I think that the personality/appeal aspect is more of a factor than we care to admit. For instance, you might be right that not just any policy would fit with a personality like Trumps. But I think that people buy into candidates at a gut level and then you can’t really dissuade them based on policy. Trump waffled on even his signature policy issues at times but that didn’t seem to matter. Trump was a superb entrepreneur who spied an untapped market and knew the proper appeal. We can learn from that.
She did not visit Wisconsin a single time during the general campaign. Tim Kaine visited a couple times. Crowd estimates numbered in the dozens.
I have post on the member feed about Wisconsin results. Between Madison and Milwaukee, Hillary actually ran up larger margins than Obama.
Is the data for non-presidential votes even available? Do they keep track of it? I guess you can see what the difference between other statewide elections and the presidential election is. But that still isn’t the actual number. Do they keep the ballots? Would anyone bother to sort through them all to get this answer? Maybe the former but certainly not the latter.
Polls themselves are useful, but they have their limitations. Generally I think the problems has not been with polls but with their interpretation. I had a professor a while back who made a point that stuck with me. Nine times out of ten if you are wrong it isn’t because the data is wrong it is because you made a wrong assumption about the data. The polls are just data, and as long as people are conducting them honestly (ie. not just making numbers up) they are what they are even when they seem wildly off the mark.
This fits pretty well with what I saw in Michigan. In the rural areas through which I drove in south central, central, and mildly north central Michigan, I saw a fair number of Trump yard signs and one or two Hillary signs. In Ann Arbor, the Hillary signs were still up this past Sunday.
It is worth remembering that in the primaries the geographical areas going for the winners — Trump and Bernie — were nearly all the same.
Val,
Your and your professor’s point is well taken. I guess I am driving at the imperfection of knowledge generally and especially when it involves human behaviors. Even in Physics, our ability to know isn’t always satisfying. Einstein was disgusted by Quantum Physics. He didn’t like the idea of having nothing left but the probability of the position & momentum of a particle. He wrestled the Quantum advocates for a good twenty years over this. The famous exchange goes:
Einstein: “Gd does not throw dice.”
Niels Bohr: “Stop telling Gd what to do.”
Would it be so bad to admit that some phenomena are not subject to measurement period?
Regards,
Jim
Good point. We should also keep in mind that even the people who respond to polls do not always know why they are choosing as they do. I’m thinking in part of the people on Ricochet who said they were neverTrump until they got into the voting booth and were confronted with a ballot and a pencil.
I guess when you practice Science approaching the underlying conceit is that the universe is not mysterious and unknowable. But, this discussion would require a whole different post. I do though agree that with respect to human behavior our knowledge is even more imperfect than what we have in other fields of study. The thing is with respect to the election people will have to come up with some sort of explanation for how and why things turned out the way they did. For Democrats it is crucial in order to regain the presidency and other lost positions, for Republicans in order to defend and expand their gains.
Looking at political science from a scientific perspective its greatest weakness is not even its irreproduciblity, but rather its infrequent rate of generating results. We only hold election every 2 years presidential ones only every 4. You can use scientific processes and statistical analysis on lots of irreproducible phenomenon so long as you can have a large data set to work with. There is no experimentation in astronomy in the classical sense, it is all observational, but you can ask questions and look for data to support it. How many election do we have to look at that we can consider comparable? In a generation (20 years) you only have 5 presidential elections. A data set of 5 is not very good. This is why people are so into polling. It offers a way to essentially generate more data about the mood of the country (which in essence is what our elections do).
I have some anecdotal evidence along that line. My parent’s generation in my family are mostly retired auto workers in Wayne county (Detroit suburbs). They have cast quite a few 3rd party votes in recent presidential elections. Trump won their votes and enthusiastic support.
It is not uncommon that after eight years of one party, that party underperforms – through apathy as you call it. It is also true (I believe, but I’m not an expert) that the party out of power has a surge of energy. As I’ve said elsewhere, this was the Republican’s election to lose.
Johnson is the wrong kind of libertarian, IMO. I wrote in the Egg McMuffin.
I noticed the Constitution Party was on the ballot, and had I researched them beforehand, I might have voted for them. They got about 17,000 votes. But there’s no sense voting against Trump on temperament grounds if your protest vote is for a raving lunatic, and many of these 3rd party people are just that.
One other point heard reported (it could have been on a Ricochet podcast, but news is flooding in from everywhere) was that while Trump’s margin of victory in MI was 12,000 votes, there were some 90,000 “straight Dem” ballots that did not vote for Her.
Sorry, that’s not embracing him – it is a flat out rejection of her.
Which makes me wonder if another GOP candidate would have beaten her. The fact that she was running against Trump may have made her (and the media and polls) overconfident. Everyone, supposedly even Trump’s campaign, believed she would win. If she was winning anyway, why bother voting for her? It lulled Her into a false sense of security, and gave “supporters” a greenlight to vote 3rd party (if at all)
I’ve thought about that the past week. I have only vague recollections of ’76, but every single presidential election from 1980 on, the more charismatic candidate won. Not usually. Every time. Even HW was blessed enough to run against Dukakis.
Yes, and in Michigan, the Greens got three times the margin between Trump and Clinton. Of course, the Libertarians received more than usual, too, because Trump.
1976 Carter was different and a change, so that is sort of Charisma.
1968 was the truly Charisma vacant election.
1964 the Charisma guy lost because he was painted as a deadly nuclear crazy and Johnson was just a conventional war crazy. Charisma won in 1964 from the aura of JFK.
1960- First real television election- charisma won
This is precisely why this type of post-election naval-gazing is nonsense. We almost never learn anything about “the eclectorate“ and its motivations by putting election returns through the meat grinder. Do it if you enjoy it, but if you think you learned something you’re probably mistaken.
Good point, and one that suggests Rubio’s charm would have crushed Hillary. My question becomes, what of Cruz v. Clinton? Both are off-putting in their own special way. Could Cruz have used his youth, intelligence and relative freshness on the national stage to squeak past Hillary?
I like what Cruz off-puts. I would have liked to see Cruz vs Hillary. His main problem would have been squeaking past the GOP Establishment, though.
Cruz would have been better off doing what Nixon did between 1962 and 1968. Eating endless dinners of rubber chicken and raising money for state parties and candidates. Still, I’ve found him a bit off putting (and he’s my senator) and slick. Reagan was a happy warrior and he should be the standard. Granted, not everyone can pull it off. Cruz I think would have a chance if he spent 4 or 8 years as a Vice President.