The Electoral College - Part 2

 

The politics of America's other peculiar institution

In part 1 of my look at the Electoral College I basically concluded it’s dumb, but we’re stuck with it. Or perhaps to put it slightly more elegantly, the Electoral College was a necessary compromise in 1787, but the reasons for its inclusion in the Constitution no longer exists. Despite this, it’s been impossible to pass an amendment to the Constitution to abolish it due to alternating periods of complacency and partisanship. Now, with the 2020 election a little over seven months away the two parties must determine a path to 270 electoral votes. To this end I’ve examined the results of the last four Presidential elections and the 2018 mid-term elections to look for trends in determining which states are likely to swing the upcoming election one way or the other.

However, before I delve into the election data, I want to take a quick look at why this is important. After all, the electors are based on Congressional representation, which is based on population, As long as you win the popular vote isn’t it overwhelmingly likely you’ll when the Election, even if there is a possibility that it could go the other way. Not really. If you win the popular vote by a large enough margin then the likelihood of winning the Electoral College also goes up, but several recent articles have shown how Donald Trump could lose the popular vote by even more in 2020 than he did in 2016, perhaps as much as five or six million votes and still win the election. 

There are two factors that make this possible. First, although the House of Representatives is proportional based on population, the Senate is not, and it’s the total of both chambers of Congress that make up a state’s Electoral College representation. This means Wyoming with a population of 567k has three votes in the Electoral College or 1 per 189k residents. California on the other hand with a population of 40 million and 55 votes gets one vote per every 727k residents. Or said another way, the Electoral College is weighted in favor of states with a small population. As a result, it’s better to win several smaller states with a total population of 40 million than to win California with the same population.

The other, probably more significant, factor that can allow someone to win the popular vote, but still lose the election is the winner take all nature of the Electoral College. This means the margin of victory in a state is irrelevant. If a candidate wins a few states by large margins, then loses a greater number of states by a small margin they could easily have a greater number of total votes but have fewer electors. The perfect example of this is looking at the five largest states from the 2016 election. Hillary Clinton won California and New York by huge margins winning 84 electors. Donald Trump won Texas by a medium amount and won Florida and Pennsylvania in squeakers winning 85 electors. In those five states, Clinton got more than 5 million more votes than Trump, but received one less elector.

One other consideration that can affect the result of the election but is harder to define is the role of third parties. When there’s a third-party candidate, who is ideologically similar to one of the major party candidates, they can bleed off some of their support potentially throwing some key states to the opposing candidate. Republicans have long claimed that Ross Perot’s run in 1992 cost George H.W. Bush the election by getting votes that would have otherwise gone to Bush. However, there’s no way to say with any certainty what would have happened without the third-party candidate. Similarly, in 2016 it appears a lot of left leaning voters went for either the Green Party of the Libertarian Party. Over the previous 3 Presidential elections third-party candidates averaged around 1.5% of the vote, in 2016 it was nearly 6%. Again, it’s impossible to say who they would have voted for between Clinton and Trump, but given how close many of the key states were it’s not hard to imagine that minor parties could have swung some of those states to Trump.

When we discuss swing states, there’s a few states that probably come to mind, Ohio has likely held that status for the longest time. However, over time the states that fall into that swing-state category tend to change as some red states become more blue and vise-versa. I’ve created the following categories to describe the likely pattern of voting in this year’s presidential election:

Hopefully the categories in the above chart are self-explanatory. For our purposes the states in the strong and solid categories don’t really matter. That’s not to say the opposing party couldn’t win some of those states, but if they do it’s going to be part of an epic landslide and won’t be important in determining the outcome. It’s really the states that are leaning towards one party, the true swing states and the three traditionally blue states that Trump won in 2016 that are going to be the key to this election. 

I do need to clarify that this chart only applies to the Presidential election, the party dynamics in these states don’t always follow strict partisan lines. In Kansas for example Republicans haven’t lost a Presidential election in decades and Trump currently has a double-digit lead, however they’ve alternated between Democratic and Republican governors for a long time, including the Democrat who was just elected in 2018. Florida conversely has gone blue more often than not in Presidential elections but hasn’t elected a Democratic governor in over two decades.

For President Trump, the formula for winning is pretty straight forward, do what he did in 2016. He won all the normal Republican states, swept the 3 swing states (FL, IA, OH) and turned 3 normally Democratic states red (MI, PA, WI). Although not the historic landslide Trump has claimed it was, it did give him the win with an extra 34 electoral votes to spare. If Trump once again holds all the traditionally red states, he could lose any one of the six swing or blue states he won and still win the election. If he loses more than one then it’s going to be close depending on which states he loses, and again this is assuming he doesn’t lose any of the Republican leaning states. 

It’s too early in the election season to take general election polls too seriously, but they can show where we’re starting from. In these six states, only Iowa is looking fairly strong for Trump, the other five are all leaning Democratic in a Trump/Biden race but mostly within the margin of error. In the three Republican leaning states (AZ, GA, NC) only Georgia is looking strong for Trump. Arizona looks like a toss-up with maybe a slight edge to Trump, but Biden is leading in each of the three most recent polls in North Carolina by 2 to 4 percent. Taken together Trump’s path to re-election looks difficult, but enough of these states are within a couple of points that Trump’s re-election is certainly well within the realm of possibility.

For the Democrats, it’s basically the opposite. I’m basing all this on Joe Biden getting the nomination. As of this writing Bernie Sanders is still in the race, however all the polling shows Sanders trailing by double digits in nearly every state and is unlikely to get the nomination unless there some significant change to the political landscape. That’s good news for Democrats because Biden is polling 2 to 4 percent higher than Sanders in most swing state against Trump. In most states the polling is all over the place making it easy to cherry pick the polls you like based on your desired outcome. However, there are two states that have consistent polling showing Biden ahead of Trump, Pennsylvania and North Carolina. These two states alone would be enough to sway the election for Biden, with everything else being the same as 2016. Also of note for the Democrats, recent polling is showing normally solidly red Texas being within the margin of error. Even if Republicans take Texas, Trump may be forced to spread his campaign thinner to keep states like Texas in the red column.

It’s all these swing state calculations that’s really the problem with the Electoral College. The vast majority of states will only get minimal attention during the election because it’s only about 10 states that are likely to sway the election one way or the other. If we had a national popular election a Republican might choose to campaign in California because even if they still loss the state, picking up an extra couple of percentage points would help their campaign. A Democrat for the same reason might do a swing through Kansas and Oklahoma. A city of 100,000 people in California or Alabama would be just as important as an equal sized city in Ohio. The future of the Electoral College will be a debate that will continue for the foreseeable future. But love it or hated, for 2020 the Biden and Trump campaigns need to have their electoral math down pat to win the White House.