Are Republicans Blowing It In The Midwest?

I think this question is a lot more complicated than most people think, and there are a lot of ways to attack it. I’m going to look into the urban/rural divide, talk through how good/bad the results were for each party in each state and then think through what it might mean for 2020 and beyond.

Losing a Two-Front War

I sometimes think of elections in these swing states as a two-front war.

The first front is the suburbs. These are areas where Republicans have traditionally performed well, but where Trump lost a lot of Romney voters and where the 2018 Republicans also performed poorly. The second front is the comparatively rural areas and towns (some of which have a lot of white, blue-collar Obama-to-Trump type voters) where Trump made gains (though those areas generally had been moving toward the GOP for some time). There are obviously a lot of voters who don’t fit into those exact camps (e.g. evangelicals who have voted GOP for a while, black Democrats, etc.) but these seem to be some of the swing-ier groups in recent elections.

In 2016, the Republicans made enough gains in one front (the more rural side) to offset losses and win swing states like Pennsylvania, Michigan, Ohio, and Wisconsin. But they took damage on both fronts in 2018.

The map tells a pretty simple story: Republicans suffered basically everywhere (you can spot increasingly blue counties or clusters of counties that represent major cities), but their margins tended to recede more in the more sparsely populated areas. Republicans had already lost a lot of ground in the urban/suburban front of their two-front war in 2016, but it seems as if they lost a bit more in some major Midwestern metros in 2018.

These losses translated into some rough aggregate numbers.

Those topline numbers aren’t great for Republicans. In most of these contests, the Republican margin at the top of the ticket fell by mid-single digits to low-to-mid double digit margins since 2016. They suggest that rural, less evangelical voters who pulled the lever for Trump aren’t guaranteed converts, and that Trump’s actions and style (maybe unsurprisingly) haven’t brought Romney-to-Clinton areas back to the GOP.

Is it All Terrible for the GOP?

That being said, swing-ish mid-Atlantic and Midwestern states didn’t all move in lockstep. We can start to see a lot of variation (and some not-as-bad signs for the GOP) when we compare the 2018 election to a different baseline.

This table compares the 2018 Senate election to the Senate result in the last election when that seat was up. That’s 2012 for every race except for the Minnesota special election (the race that’s mapped in the .gifs), which is 2014 (when now-disgraced Al Franken managed to keep his Senate seat during a good GOP year).

I think the two most interesting states in this table are Ohio and Michigan.

In 2012, Democratic senator Sherrod Brown posted a mid-single digit win in Ohio as Barack Obama won the state by three points. That result didn’t require much explanation. If you add a decent-but-not-crazy Senate incumbency bonus to Obama’s margin, you get something reasonably close to Brown’s margin.

In 2018, the national environment was significantly more Democratic but Brown won by a very similar margin. Some of that movement can be explained by the specifics of candidates (e.g. I’ve heard some argue that Josh Mandel, Brown’s 2012 opponent, was a less than great candidate), but I tend to think that some of it may just be Ohio. Brown’s unchanged margin plus Mike DeWine’s gubernatorial win and Trump’s eight point win there in 2016 all suggest that the state might be generally becoming more Republican. That is, Ohio might be moving out of the swing-state-that-Trump-got-a-wonky-result-in into light-red-but-not-always-safe-for-the-GOP category.

In Michigan, Republican John James lost to Democratic senator Debbie Stabenow by about six points. I don’t want to talk Renacci or James up too much because on some level an L is an L in politics and both lost. And in Michigan, Gretchen Whitmer, the Democratic candidate won by about 10 points. But Senate races tend to be more tied to national politics than governor’s races. And James’s margin suggests that Trump and the GOP have an opportunity to make Michigan genuinely swingy (which is really something considering Obama won the state by almost 10 points in 2012) and keep Trump’s 2016 win from being a weird one-off.

But not every state looks like Ohio and Michigan. In Pennsylvania—arguably the most important state on the electoral map—Democratic senator Bob Casey and Democratic governorTom Wolf won re-election by 13 and 17 points respectively. That represents a huge swing toward Democrats from 2016, and both Wolf and Casey improved on their previous margins. If Republicans are looking for good news from this cycle, they shouldn’t look at Pennsylvania.

Wisconsin and Minnesota are also less interesting to me because they’re behaving roughly as I would have expected them to. Tina Smith’s 11-point win in Minnesota seems consistent to me with how a light blue state should vote in a semi-open (Smith is an appointed incumbent, which is less of an advantage than incumbency) Senate contest in a year like 2018 when Democrats won by a solid margin nationally. Similarly, Tammy Baldwin (an incumbent) won by 11 points in a state that’s only a point or two less blue than Minnesota. Wisconsin has arguably been marginal for a while and Minnesota has been reddening for some time, so the 2018 results didn’t change my perception of those states much.

So what’s it all mean for 2020?

There’s an obvious first takeaway here: Republicans don’t have a lock on the Upper Midwest or Pennsylvania, and some slippage there could lead Trump into a very precarious situation in the 2020 election. But we have no clue what the national environment is going to look like in 2020. If it’s good for the GOP then Trump could win these states and win again, but if it’s rough for Republicans then they could easily lose these states and the White House with them.

I would also argue that the Ohio results and the Michigan results aren’t as bad for the GOP as those in other states. We can’t extrapolate too much off one midterm result, but it’s possible that those states have moved right more generally and that the 2016 results (where Trump made significant gains) aren’t completely tied to Trump and Trump alone.

Finally, it’s worth noting that Republicans could do well on both fronts. It hasn’t happened recently (e.g. Trump arguably won these states by making gains on the rural level and not the suburban level, other recent Republican presidential candidates ended up losing these states), but it’s not impossible to imagine a strategy that fuses some elements of Trump-ism to traditional conservatism and ends up with a bigger winning coalition. I don’t think that this is the only way for the GOP to win a majority, and at this point I’m not confident that Trump is the sort of Republican who could pull that sort of a feat off. But I don’t think that Republicans are stuck at 46 of the popular vote (or less) permanently. Eventually they’ll get more votes, and it’s worth trying to anticipate how or what that might look like ahead of time (more pieces on that theme to come in the next two years).