Saturday, June 29, 2019

NC Makes Constitutional Law History Again

In my previous post, I contemplated whether North Carolina's partisan gerrymandering case, known as "Rucho v. Common Cause" and combined with a Maryland case, would make constitutional law history. And indeed, it did.

It made history because the court, until a new majority assumes power, said "we aren't getting involved in these partisan gerrymandering cases because they are too political." Which further made the court a political institution, in the eyes of some, and an odd savior of partisan gerrymandering to others.

In an interview the day of the U.S. Supreme Court's decision not to involve federal courts in partisan gerrymandering because of the "political question" doctrine regarding justiciability, I mentioned the fact that we have a dual judicial system, with both a federal court system and a state court system. And that the state case, challenging partisan gerrymandering based on state constitutional law, was working its way through the state system and could end up at the North Carolina Supreme Court. Immediately after my comments, former state senator Bob Rucho (the "Rucho" of the case name) said the following:
"But the courts in North Carolina, since we all live under the U.S. Constitution, should not be imposing political decisions, whether it be on the federal or state level, and I think that will probably get to a (U.S.) Supreme Court decision at some point." 
In my Introduction to American Politics, as well as my State Politics, Judicial Process, and Constitutional Law: Powers classes, I discuss the vertical division of governing power based on the concept of federalism, and the fact that all Americans live under two judicial systems: federal and their respective state court systems. And as noted by the U.S. Courts website:

State courts are the final arbiters of state laws and constitutions. Their interpretation of federal law or the U.S. Constitution may be appealed to the U.S. Supreme Court. The Supreme Court may choose to hear or not to hear such cases. (emphasis added)
Now, in general, it has been my observation of modern Republican Party principles and politics that they prefer to have power 'reside closest to the people' and that usually means preferring the states over federal government actions and policies. But to hear a former state Republican senator espouse the hope that the federal government, especially the institution of the U.S. Supreme Court, would overrule a state constitutional action, struck me as going against the grain of GOP principles, to say the least.

But, it may be more the principle of "I won here, and now you can't win there" that could be guiding that political philosophy. And I'll discuss the venue of state constitutional law later.

But the ideological "change of heart" aside, the North Carolina partisan gerrymandering case did make constitutional law history, most notably with five justices--the conservative wing of the court--making it clear that, in the words of the majority opinion's author Chief Justice John Roberts:

"The districting plans at issue here are highly partisan, by any measure." (page 2)

"Excessive partisanship in districting leads to results that reasonably seem unjust." (page 30). 
What the court held, however, was that it is not the place of federal courts to insert themselves into an issue based on "political questions," which the courts have discretionary power to say "we will leave it to others (namely, 'political' entities like Congress or the states) to resolve this politically-charged matter."

Some would contend that by not deciding on partisan gerrymandering, the courts have inserted themselves into politics. And while Chief Justice Roberts made some questionable rationale for not doing so, in this era of hyper-partisan and polarizing politics, making a decision either way would have injected the nation's highest court clearly into politics. So it was the proverbial "rock & hard place" that the Court found itself, and with a clear five conservative majority, they chose to block the federal courts as the potential solver of this issue.

I make the statement that the Chief Justice "makes some questionable rationale" for his decision not lightly, but based on personal research and findings in the political science discipline that challenges many of the assumptions made by Roberts in one key paragraph (found on pages 24-25 of the majority opinion):

Voters elect individual candidates in individual districts, and their selections depend on the issues that matter to them, the quality of the candidates, the tone of the candidates’ campaigns, the performance of an incumbent, national events or local issues that drive voter turnout, and other considerations. Many voters split their tickets. Others never register with a political party, and vote for candidates from both major parties at different points during their lifetimes. For all of those reasons, asking judges to predict how a particular districting map will perform in future elections risks basing constitutional holdings on unstable ground outside judicial expertise. 
Let's take each of the first three sentences and explore what the research in political science has to say about voter choice, ticket splitting, and "independents" in our electorates.

What Drives Voters' Choices?

First, the statement about what drives voter choices. Since the "behavioral revolution" in the discipline of political science in the 1950s, political scientists have been exploring what drives voters to make the choices at the ballot box; in other words, what makes voters "behave" the way that they do. A classic cornerstone analysis was "The American Voter," which put forward one of the fundamental findings about the electoral behavior of American voters: partisan identity drives voters' choices. And while there is an acknowledgement that party identification isn't the be-all-end-all when a voter picks a candidate, recent elections have shown the intensity of party loyalty by voters who self-identify with a political party.

And now, some evidence (because, unlike some, that's what political scientist do in their research and advancing arguments). First, here's the data from the 2008, 2012, and 2016 American National Election Studies, which is considered the premier data collection for presidential year voters in the nation, regarding the party identification and strength of party ID of voters, with their presidential vote choices:

So, let's break these three graphs down by "strength of party identification" categories:

  • Strong Partisans (Strong Democrats and Strong Republicans) have typically voted 94 to 98 percent of the time for their respective party's presidential candidate. In summation: these are the true party loyalsts in the electorate. 
  • Not Strong Partisans (who I describe as not considering themselves "foam at the mouth" partisans): 75 to 88 percent of the time they vote for their respective party's presidential candidate. If I'm going to Vegas with at least a 75 percent chance of winning, I'm booking an airline ticket now. 
  • Independent Partisans (or those who self-identify first as "independent," but when asked "do you lean to one party or the other?" say they lean to the Democratic or Republican Party): 78-89 percent of the time they vote for their respective party's presidential candidate. Now, whenever I show this graph to audiences, I ask a simple follow-up question: are these voters really independent? 
  • Pure Independents (the ones who don't lean to one party or the other): here's where the electorate's "swing voters" are often located, but note their percentages of the electorate: less than 10 percent of the electorate are made up of these folks, per ANES. 

So, as a political scientist, I'd say the track record of partisan voting loyalty is pretty intense. Not convinced with one data set? OK, how about another massive data collection, known as the Cooperative Congressional Election Study for 2008, 2012, and 2016 (same principle as the ANES graphs):

The end result for the American electorate as a whole: among the party identifiers (strong, not very strong, independent-partisans), there's a 79 to 98 percent chance that they will vote for their party's presidential nominee.

"But wait, that's the entire country! What about North Carolina?"

Well, here's the CCES analysis of North Carolina voter behavior by party identification and strength when it comes to president vote choices in 2008, 2012, and 2016:

So, within the Old North State's electorate for the past three elections, partisan identifiers are 70 to 98 percent likely to vote for their party's presidential candidate. Also notice, like at the national level, independents within the North Carolina electorates tend to be less than 13 percent of the overall voters.

And while there are other factors that may drive voter's decisions and behavior when it comes to casting ballots, political scientists agree that party identification is a strong driver of voters' choices.


Next, Chief Justice Robert's claims that there are voters who "split their tickets," meaning that a voter picks a Republican candidate for President, a Democratic candidate for US Senate, perhaps bounces back to Republican for the U.S. House, and just bounces back and forth between the parties down their ballot. And what, to assume the chief's intention, this means is that voters cast ballots for the "candidates," and not necessarily the "parties" when it comes to voter behavior.

As the above graphs point out fairly clearly, the only "ticket splitting" being done by voters are those 'pure independents' in the middle. Partisan "defectors" (those voters who identify with one party but tend to vote for the other party's presidential candidate) are typically 'not strong' partisans, but in the national electorate, only less than 20 percent of the votes from not-strong partisans goes to the opposition. This has also been another well-documented research finding in political science, thanks in large part to voters sorting themselves into ideologically-coherent parties.

North Carolina's not-strong partisans tend to have a higher 'defection' rate: in 2008, 31 percent of not-strong Democrats voted for John McCain. But interestingly, over 2012 and 2016, that 'defection rate' has dropped among not-strong Democrats, from 24 down to 19 percent, moving in line with the national numbers.

In my research on North Carolina voter behavior, another technique that I use is to compare the performance within various districts (congressional, state house, and state senate) between the Republican presidential candidate and the GOP candidates at different district levels.

So, in taking the 13 congressional districts in 2016, I lined up Trump's performance in the congressional district against the Republican congressional candidates performance, using a "simple scatter plot with a line of fit (meaning, what's the trend when you compare all 13 districts together):

There's almost a "perfect" fit between how Trump performed and how the Republican congressional candidate performed in the district, which may indicate that voters cast ballots for Trump and then for the Republican congressional candidate at a pretty high level.

Next, I did the same for the 2016 Trump and 2016 State Senate Republican candidates:

Again, with some minor exceptions, most of the state senate GOP candidates had similar performances to Trump in 2016's election (meaning, 95 percent of the 2016's GOP State Senate candidate's vote is explained by Trump's district vote).

Finally, take the 2016 State House vote and comparison to the Trump vote within each house district:

In looking at the 'line of fit,' you'll see more districts are further away from that line, which explains that 82 percent of a state GOP house candidate's vote was explained by Trump's district vote. Still, 80 of a Republican state house candidates' vote being explained by the voter picking the Republican presidential candidate seems a strong relationship.

Two years later, with the 2018 mid-term election, I did the same analysis, this time taking the 2016 Trump district vote and looking at the 2018 GOP candidates' votes in the three districts.

First, congressional districts:

Next, state senate:

Finally, state house:

So, what these three scatterplots tell me, as a political scientist, is that the 2018 vote for Republican district candidates--be they congressional, state senate, or state house--are driven largely (97 percent or greater) by the vote for Trump two years ago.

Meaning: split ticket voters in North Carolina, at least, are few and far between.

Independents in the Registered Voter Pool and the Electorate

Finally, Chief Justice Roberts contends that many voters "never register with a political party," and that is certainly true in North Carolina.

Registered unaffiliated voters are now the second largest group among the 6.6 million registered voters in the state, surpassing registered Republicans. Partisan registration has decreased significantly since 2000. Most of this change is being driven by younger voters:

Millennials and now Generation Z voters are much more likely to registered as unaffiliated (voter registration status by generational cohorts as of May, 2019):

But even with this rise of the 'unaffiliated' among younger voters, that tendency to be 'independent' when it comes to vote choices does not pan out, as per ANES data on generational cohorts and presidential vote choices demonstrates:

The key difference between Millennials and other generation cohorts is that their Democratic identification (strong to lean to) is 55 percent, the highest of any generational cohort, and among Democratic identifiers, 86 percent of them voted for Hillary Clinton in 2016.

So, to sum up: the research conducted in political science shows an electorate of much more intense partisans, voting their party loyalty and not defecting, and while more voters (those under the age of 40 so far) are choosing to register independently of partisans, their vote choices demonstrate their party loyalty.

But this shouldn't be surprising that the Chief dismisses social science research, thanks to his portrayal of such areas as "sociological gobbledygook" in an oral argument (page 40) over a previous partisan gerrymandering redistricting case.

I want to end with a final thought regarding the state constitutional law aspect to all of this. Some commentators, and even the U.S. Supreme Court's four liberal justices, lead by Justice Kagan in her dissent, expressed "doom and gloom" for the nation's polarized environment. Our polarized environment will likely continue into the near future, even with a generational shift occurring. But within the states, as the Chief Justice noted, there are attempts to resolve partisan gerrymandering. One prime example of this was in the Commonwealth of Pennsylvania, whose supreme court struck down partisan gerrymandered congressional districts based on the state constitution's "free and equal elections" clause.

And what does North Carolina's constitution say? Article I, Section 5: "All elections shall be free."

In other words: when it comes to North Carolina's partisan gerrymandering, stay tuned. We're not done yet with partisan gerrymandering challenges in the Old North State.