Since the years of the George W. Bush administration, pundits have been poring over electoral maps, and drawing inferences – alternately illuminating, plausible, and at the very least interesting (though quite often false) – from the information conveyed in those representations. By this point, enough pieces focused on the fundamental cultural and political divisions between rural America and urban America have been published that one could probably assemble an edited volume on the topic.
The latest of these, Red State Blue City, by David Graham, has just come out in the Atlantic. It raises much of the same issues as the previous decade and a half of similar pieces. If instead of focusing on the state-by-state electoral map, and you focus on the counties, it is clear that, with a few exceptions that seem to be based on racial demographics, there’s a significant and well-established split at the county level. Rural counties go hard red, the majority of their voters supporting Republicans. Urban counties show the opposite trend, deep blue, supporting the Democrats. The urban counties, of course, wield a lot more votes since they are densely populated, but there are far more rural counties, at least in most states.
The suburbs, the border zones, or areas not only of sprawl, but transition, often get ignored in these sorts of analyses, but really would bear much more scrutiny (perhaps a good topic for a later post and discussion). The main point that many of these county-wide electoral analyses do come to some consensus upon is that when it comes to the deepening cultural divisions within our country, reflected along political voting lines, one of the most significant differences is the urban-rural geographical divide.
There definitely is something there – all you have to do is pore over the state electoral maps – and this also seems to be a self-perpetuating division as well. If you grow up, live in, or move to an area where you’re surrounded by people who – very broadly speaking – embody and espouse the agenda and mindset of the Reds or the Blues, presumably over time that environment exerts its effects upon one, and after some time, you start thinking and seeing the world in a similar way.
And, when you look across the divide to “those other people” – perhaps making a daytrip to see how the proverbial other half lives, or going by the now infinitely available images on social media or programs – you can observe an entirely different mindset at work, pushing values, making claims, and setting out arguments opposed to your own. This is, granted, a gross oversimplification of the complex factionalization that has been occurring within our country for decades, but you get the point.
What this particular Atlantic article stresses is that we ought now to expect to see a lot of political clashes – outside of election season – between city and countryside (with the suburbs perhaps poised between them). I tend to agree, for example, on the bedeviled issue of sanctuary cities that is already coming to a head in the new administration (there’s a lot more to be said about that issue as well, both in general, and in the times of the new Trump administration particularly) – that is going to involve serious conflict nationwide, and may take years to come to some resolution.
There is one aspect of this overarching narrative of fundamental blue urban vs. red rural cultural and political schism that I find difficult to buy into, and my misgivings about it stem from observing two things here in Wisconsin, one of the key “battleground” states that delivered Trump his electoral victory over Clinton. I live and work in downtown Milwaukee – a very blue area, so blue that for a number of local offices, there was only one candidate on the ballot, a Democrat – and I also know quite a few people who intended to vote, and then did vote for Trump (I wasn’t one of them). What was interesting was that, for the most part, they kept that fact to themselves, both in face-to-face conversations and in social media, mainly to avoid conflicts, arguments, harassment, and the like.
The flip side of this was that, until the actual election results, most of the the Clinton supporters I know were in full triumphal mode here, confident of their impending victory. Even after the election, there were many people who openly supported Clinton who made the rather implausible complaint – “I don’t know anyone who voted for Trump” – when quite likely, they knew quite a few who did.
Now, that is admittedly anecdotal information I’m providing here about these two aspects of blue urban electoral politics – first that many people in a blue area did vote along red lines, and second that they largely kept that voting intention and preference to themselves, revealing it only to fellow Trump supporters, and to those others (like me) who they trusted not to adopt hostile, dismissive, or inquisitorial attitudes towards them. So, let’s get away from personal experiences and stories, and look at sheer numbers. What do they reveal to us?
Clinton solidly won Milwaukee county, as expected. She took 66.4% of the vote, 288,986 votes – going by Politico’s data. That sounds quite good, and if you look at the map, that sets Milwaukee county as a solidly blue island in a sea of surrounding red – the suburb counties of Waukesha, Ozaukee, and Racine. That picture is a bit misleading, though. Trump got 61.6% of the vote in solidly Republican Waukesha county, 57.1% in Ozaukee, but only 49.8 in Racine (where Clinton took 45.4).
Clinton did even better, understandably enough, in the City of Milwaukee itself, winning 76.5% and 188,657 actual votes. But notice something. Trump got 18.4% of the votes in the city itself, and 29.0% of the votes in Milwaukee county. That’s significant, in a state where the contest is close. He got 45,411 votes in the city of Milwaukee, and 126,091 in Milwaukee county. That is just 19,428 fewer votes in Milwaukee county (where Trump lost) than in Waukesha county (which he won by a significantly smaller percentage margin than Clinton won Milwaukee).
Where else did Clinton win in Wisconsin? She won Dane county – where Madison, the state capital, major university town, and a deep blue liberal enclave, sits. And she got a higher percentage there than in Milwaukee. Trump still got 23.4%, or 71,270 votes. Looking south and west where four other counties went blue, it was a lot closer in all of those. Interestingly, three of those blue wins are clearly rural counties, and Rock county probably could be considered to be so as well. You might argue that they present that anomaly by being dragged into the electoral wake of Dane county, for one reason or another.
Where else? Those thinly populated blue counties at the top of the map, where Clinton got in the low 50s, and Trump in the low-middle 40s. Menomonee county went resoundingly blue, 78.4%, but gave Clinton just over 1000 votes. La Crosse and Eu Claire – both counties with university towns – also went blue, as did Portage, which has UW-Steven’s Point. With each of those, again, low 50s for Clinton (except Portage, where she got 48.6% to Trump’s 45.4%).
There were a few red counties where Trump just barely eked out a victory – Door County is a prime example, 49.3% Trump to 46.1% Clinton – but if you look at many of them, there are double-digit gaps between Trump’s and Clinton’s percentage of the vote. And, when you add up all those small contributions, they are significant. But, consider this – the two most important counties that Trump lost, Milwaukee and Dane, gave him 126,091 and 71,270 votes. Nearly 200,000 votes from those blue, urban, liberal counties.
Clinton lost Wisconsin by a bit over 27,000 votes. She won those two urban counties put together by over 309,000 votes (those two counties alone account for 36.6% of her total votes in the state), but she lost almost 200,000 votes in those counties as well to her main opponent. That is a lot of people in what was supposed to be the strongholds of Democrat territory, in a state where the vote clearly mattered, who did vote for Trump over Clinton.
When you look at other state maps, particularly those where the state counts were close, you see similar patterns. Trump and the Republican party sweep most of the rural counties, with a few exceptions (for example, those in the south with large African-American populations) – and those votes do matter, but aren’t enough to give him a win. He also does well in some of the suburb counties, depending on how they lean. He loses the major urban counties, and even some of the university town counties, but for Clinton’s purposes – at least in quite a few states – he doesn’t lose them by enough. He garners significant support in those urban counties where presumably Trump supporters are few and far between, and in several states, this tips him over the edge, and gives the Republicans an electoral victory in the 2016 Presidential Election.
This is a win for Trump and for the Republicans that came as an upset for anyone who was relying upon the polling prior to the election. Nate Silver’s up-to-then solidly predictive models, for one example, had given Trump a 28.6% chance of winning the electoral college (and specifically a 16.5% chance of winning Wisconsin) on November 8 – and that was bullish compared to many of the his predictions of the last few weeks. What happened?
I suspect – though this would be quite difficult to find data to support – that quite frankly many of the polls were off because people who intended to support Trump told pollsters that they weren’t doing so – that they were voting Clinton, a third party, or were undecided. I don’t think this was the case with the level of support for Clinton in the polls. Why would people do that? This is where we get back into these red vs. blue generalizations, analyses, and inferences about political and cultural divides. It seems quite clear – in our blue urban counties, there is a sizable minority of people who do support Trump, who do identify much more with the Republican agenda and coalition, but who don’t make those allegiances public.
Why is that the case? Does it make sense for them to adopt that stance? And what are the implications for our fractured and factionalized liberal democratic society? Those are issues that still remain to be explored, but clearly addressing them is going to require coming to terms with it actually being the case, enough so to exert significant effects upon presidential elections.