View profile

Franklin & Marshall College Poll: Feeding the Goat

Issue #16 • View online
Franklin & Marshall College Poll
Dear Readers,
This month’s newsletter discusses what we can learn about the upcoming mid-term elections from the state’s recent municipal election. Many perspectives have been offered about what 2021 means for 2022 and, unsurprisingly, some of these perspectives are contradictory.  I discuss how well some of these perspectives hold up when scrutinized by the actual votes cast in this and the state’s 2017 municipal election.
Thank you for reading,
 Berwood Yost

The political media is saturated with reporters and analysts who are impelled to create as much content as possible to feed their platforms’ ever-growing needs, which means every election outcome is now input for predictions about the next. The 2021 municipal election was just another example of this all-too-common ritual, mostly intended to draw eyes, generate social media buzz, and raise interest in the next election.
If you read about the results of the 2021 election in Pennsylvania you may have heard that it signaled a GOP resurgence fueled by the suburb’s turn to Republican candidates.[i] Or you may have heard that it was actually a better year for Democrats than they had in the prior municipal election in 2017.[ii]
Most likely, though, you heard this election was an undoubted preview of the mid-terms. As the New York Times On Politics newsletter reported on November 18,
In Pennsylvania, a Supreme Court vacancy was contested with millions of dollars in spending. In some ways, the contest functionally pitted a generic Democrat against a generic Republican, because even the most engaged voters know little to nothing about candidates for the judiciary…The Republican candidate won by 2.6 [sic] percentage points - in a state Biden carried by 1.2 points in 2020. That represented a nearly four-point improvement for Republicans.
So, which of these particular perspectives is most strongly supported by the data as we now understand it?
What do the Votes Say?
Plotting county vote shares for this election and the most recent similar election, which took place in 2017, is one way to interpret what happened this fall. The scatter plot below shows the Democratic share of the Supreme Court vote won in each county in 2021 and 2017. Dots above the diagonal line on the plot mean that there was a greater share of the vote for the Democratic Supreme Court candidate in 2021 than in 2017.[iii]
The county-level vote shares in 2017 and 2021 are strongly correlated, but the plot makes it immediately clear from all the points that are above the diagonal line that the Democratic vote share was larger in most counties compared to 2017. Statewide, the share of the vote won by the Democratic candidate increased from 47.7 percent to 49.5 percent, which is nearly a two point swing from 2017. This seems to affirm the notion that Democrats’ may have overachieved, not under-performed, in the state in 2021.
Despite the New York Time’s pronouncement, the Republican Supreme Court candidate won by one percent, not nearly three. Further, while Biden won the state by 1.2 percent, the Republican congressional candidates actually carried the total Congressional vote in the state by 1.2 percent (50.6 to 48.4) in 2020. I’m not suggesting which of these is the better comparison, but referencing the generic ballot is normally about expected Congressional outcomes. If that’s the case, the Republican performance was about the same as it was in 2020.
If we are thinking about the implications of 2021 for the next election, seeing what happened in the three counties in the state that voted twice for President Obama and who then voted for President Trump in 2016 (Erie, Luzerne, and Northampton, which are labelled on the plot) could offer some clues. Two of the state’s three Obama-Obama-Trump counties returned to Biden in 2020 and stayed with the Democratic Supreme Court candidate in 2021. The Democratic share of the vote actually increased in each of these three counties compared to 2017.
Bucks County, singled out as an augur of the suburban voter for 2022 and beyond by Politico’s haruspex, yielded almost exactly the same share of its vote to the Democrat in 2021 as it did in 2017. The high voter turnout in Bucks County could be interpreted as being driven by national issues that arose in the races for local school boards that animated Republican voters. But the counter-argument to the nationalization narrative in this election is the fact that Luzerne County, the quintessential Trump County, had the third lowest turnout in the state at 25 percent of registered voters. Philadelphia had the lowest turnout among registered voters, as is often the case, with only about 20 percent of voters turning out compared to the statewide turnout of about 32 percent.
The Predictive Value of the 2021 Municipal Election
Every election cycle is functionally the same, but every election is also unique in the issues that matter, whether voters want change or stability, and how the candidates’ qualities, capabilities, and messages capture voters’ interests. Attempts to look ahead are entertaining, perhaps, but there are too many changes from election to election and race to race to expect our predictions to be of any real value.
Let’s just think about the 2017 election for a moment. Republicans won a Supreme Court race and one of two Commonwealth Court races, exactly as happened this year. There were two special elections for the State House in 2017, as there were this year, and the races in 2017 were easily retained by the Democrats. The two special House elections held this year were also easily retained by the Democrats. Those 2017 races missed the electoral headwinds that Republicans faced in the 2018 mid-terms. The same will probably be true about interpretations of this year’s election.
It is safe to say that the state’s 2021 municipal election reaffirms what we know about the state–it is a closely divided, competitive state whose outcomes will be decided by which of the state’s voting blocks and geographic regions are most engaged and attentive. It is also probable that Democratic candidates are going to get beaten handily in 2022 because of the price that the President’s party nearly always pays in the mid-terms.
It is also safe to say that the stories after the 2022 mid-terms will be all about what those mid-terms mean for the upcoming presidential election. There probably won’t be much value for readers in most of these stories, but every media outlet is trying to generate content around the clock. As a long-time journalist told me when I asked why we get so many of these stories, “The goat needs to be fed – even if it’s only garbage and tin cans.” We should all remember that.
References & Resources
[ii] It’s not as simple as a ‘red wave.’ Here’s what really happened in the Pa. and N.J. elections.
Did you enjoy this issue?