Ghosts of Hanging Chads, and the Problem with Bellwethers

Samuel Rubinstein 19 September 2020
Image Credits: NDLA

Florida is the bellwether state: whoever wins Florida wins the presidency. That has been the case in every election since 1992, and the image of Florida as the pivotal swing-state was imprinted firmly on our minds in 2000, when Al Gore, despite winning the popular vote, lost to George W. Bush precisely because he lost Florida. Twenty years on and the Democrats are still haunted by that election: by the confusion, the hanging chads, and the fear that Florida could steal the election from Joe Biden as it did from Gore. California and New York are reliably blue, and Texas is reliably red. But Florida is purple. Two recent polls have shown an exact 50-50 tie[1]: its bundle of 29 electoral votes really could go either way. Therein lies the problem.

Both parties are desperate to capture the state. Mike Bloomberg has invested $100m in Biden’s campaign in Florida[2], and Donald Trump’s relocation to Palm Beach – he will personally be voting there, rather than in New York, this November – has been interpreted as part of a broader strategy to coddle the local electorate and keep the state red[3].

Image Credits: Wikimedia Commons

Figure 1: Presidential Election Results in Florida, 2016 (Image Credits: Wikimedia Commons)

Even analysts who are optimistic about Biden’s chances adhere to the conventional wisdom of Florida’s supreme importance, suggesting that Biden, however encouraging the polls may seem, has a Florida-shaped Achilles heel which, if struck, erases his lead. The Economist’s model is bullish on Biden, consistently giving him a six in seven chance of winning[4], but, if Trump wins Florida, Biden’s odds plummet to one in two[5]. If Biden wins Florida, Trump is in serious trouble. If Biden loses Florida, his advantage, according to this model, will evaporate, and you might as well toss a coin.

Florida does not feature in Biden’s most obvious path to victory. It makes far more sense for him to focus his efforts on the comparatively straightforward task of flipping those states in the Rust Belt which Clinton was expected to win in 2016 but didn’t: Pennsylvania, Michigan, and Wisconsin. Biden’s personal connection to Pennsylvania, his birth-state, could be of some benefit. Michigan likewise leans towards Biden: it was here that Trump’s 2016 margin of victory was narrowest, and Biden should be buoyed by the fact that the state’s black voters, many of whom stayed at home in 2016, prefer him to Clinton. Wisconsin, the whitest of the three, could prove the most difficult, especially given the recent unrest in Kenosha following the shooting of Jacob Blake, but even there Biden enjoys a commanding lead in the polls[6]. Provided that he wins those three states, which the polls suggest he will, and that he does not lose any of the states that Clinton won in 2016, he does not need to win Florida to obtain the presidency.

Even if Florida is not essential to a Biden victory, a loss in Florida could indicate a national trend in Trump’s favour. Nevertheless, Florida has certain quirks which render it quite uncharacteristic of America as a whole: it has the highest proportion of people aged 65 and over of any state, and its Hispanic population is uncharacteristically right-leaning, predominantly due to the high number of Cuban exiles. Both of these features should make Florida fertile ground for Trump. It is worth noting also that if, on a national scale, only one in every two hundred 2016 Trump supporters decided to vote for Biden this time[7], Biden would win Michigan, Wisconsin, and Pennsylvania (and therefore the election) while allowing Trump to retain Florida. It is entirely conceivable that Trump will win Florida owing to these peculiarities, and, in so doing, that he will lower Biden’s total electoral vote-count by 29 without making much of a dent to Biden’s odds of winning a majority in the electoral college.

That Florida has reflected the national outcome in the last six presidential elections is not something to be sniffed at. It is legitimate for analysts to leap excitedly at patterns like these. But precedents are hardly prescriptive, as the Democrats learned in 2016 when the thirty-year tradition of Wisconsin, Michigan, and Pennsylvania voting for the Democratic candidate cracked. It may well have seemed in 1992, before Floridians chose George H.W. Bush over Bill Clinton, that the precedent of Florida determining and reflecting the election was an axiom of American politics; after all, Florida’s electoral votes had not accrued to the losing candidate since Nixon in 1960. Likewise in 1960, the law of the Floridian bellwether might have appeared watertight and inviolable, since Florida had supported the winner in every election since 1924. In the strange and complex mires of US elections, it turns out that black swans aren’t too hard to come by.

My suspicion that the Florida bellwether will not hold true in November comes in part from the fact that it was not predictive in the last federal election. In 2018, amidst a national ‘blue tsunami’[8] that saw Democrats recapture the House of Representatives, Florida elected a Republican senator in Rick Scott and a Republican governor in Ron DeSantis. Both of these GOP victories had only wafer-thin margins. Still, the fact remains that if the swing in Florida had resembled the national political landscape, Andrew Gillum would have made history as Florida’s first black governor, and Bill Nelson would have kept his place in the Senate Democratic Caucus. Instead, the state appointed its fourth Republican governor in a row, and, for the first time since 1875, both of Florida’s Senate seats are held by Republicans, in another instructive example of historical patterns being broken. Scott, and especially DeSantis, are staunch Trump loyalists. Trumpism in Florida is alive and well.

Image Credits: Wikimedia Commons

Figure 2: Gov. Ron DeSantis Speaks at a Turning Point USA event in 2018 – George Skidmore

Whenever Trump appears to be gaining ground in Florida, and whenever the Florida polls shift at all, a media frenzy ensues. The state has adopted a meaning of its own as the ultimate key to the White House. Other swing states have somehow managed to escape such infamy. Ohio, which took on something of Florida’s 2000 role in 2004, is the object of far less fascination, in part because Trump is roundly favoured to win it. Likewise, states with nail-bitingly tight polls, like North Carolina, receive much less attention than they deserve.

Insofar as Florida is important in this election, its importance is asymmetrical. If Trump loses Florida he will find it all but impossible to win the election. But if Biden loses Florida, he will still enjoy a considerable advantage over Trump: one far greater, I think, than a simple coin-toss. Florida is neither essential to Biden’s path to victory, nor a microcosm of the American political environment. That 2000 hangover, the ghosts of hanging chads, ought to be shaken off. If any state is to assume Florida’s symbolic meaning, it should be one of the three battleground states in the Rust Belt. It is on their terrain that the election will be won and lost.

 

[1]http://maristpoll.marist.edu/wp-content/uploads/2020/09/NBC-News_Marist-Poll_FL-Likely-Voters_NOS-and-Tables_2020090712221.pdf; https://business.fau.edu/departments/economics/business-economics-polling/bepi-polls/bepi-polls-2020/florida-2020-its-the-economy.php
[2]https://apnews.com/ab11b7e46a39d25538deb174de9977b6
[3]https://qz.com/1740938/trumps-official-move-to-florida-is-conveniently-timed/
[4]https://www.economist.com/leaders/2020/09/03/americas-ugly-election
[5]https://twitter.com/gelliottmorris/status/1304467336980815877
[6]https://morningconsult.com/form/wisconsin-presidential-election-tracking/
[7]https://fivethirtyeight.com/features/why-florida-could-go-blue-in-2020/
[8]https://edition.cnn.com/2018/12/06/politics/latest-house-vote-blue-wave/index.html