Why Trump’s performance matters in Mass.
Clinton expected to win, but her victory margin could influence legislative races
REPUBLICAN KATE CAMPANALE stunned the state’s political establishment in 2014 by winning the 17th Worcester district House seat, edging out Democrat Douglas Belanger by just 43 votes out of 9,750 cast. Campanale will face Democratic challenger Moses Dixon on November 8, after Dixon defeated Belanger in the September 8 Democratic primary. Campanale is in a stronger position than 2014 as a one-term incumbent, but the fate of the House seat and Campanale may rest on the popularity of another Republican, Donald Trump.
Popular poll-based presidential voting models give Hillary Clinton a 99 percent probability of winning Massachusetts and its 11 electoral votes. Yet even though Clinton is the clear favorite, her margin of victory in the Bay State may influence the outcome of down-ballot races for the Legislature.
A statistical analysis of the 211 contested Massachusetts state legislative races in the presidential years of 2004, 2008, and 2012 shows that a 1 percentage point change up or down in the presidential margin usually translates into a 0.7-point change for the candidate of the corresponding party in a state legislative race. As a result, a 37-point loss for Trump—on the order of Bill Clinton’s Massachusetts margin over Bob Dole in 1996—would likely make for a terrible night for Republican state legislative candidates. A 24-point loss, on the other hand—the margin of President Obama’s Massachusetts victory over Mitt Romney in 2012—would lead to a more typical outcome, as judged by elections since 2004.
The statistical model used to analyze the 211 contested state legislative races from 2004, 2008, and 2012 uses a simple linear regression to predict the Democratic/Republican margin for each state legislative race based on the margin of the presidential race in the district, and an indicator of whether the state legislative seat has a Democratic incumbent, is open, or has a Republican incumbent. While there are, of course, many other factors that help determine the results in each race (candidate quality, voter outreach, etc.), the regression model explains 80 percent of the variation in the margin, making it a statistically useful tool for analyzing the relationship between state legislative margins and presidential race margins.
Moderate Bay State Republicans such as Gov. Charlie Baker, hoping to keep the socially moderate and fiscally conservative Northeast Republican brand in play, have made a principled stand against Trump’s divisive and racist ideology. Tea Party supporters, such as Rep. Geoff Diehl of Whitman, publicly endorse Trump and are working for his victory. Much of the Massachusetts GOP establishment supported more mainstream candidates in the presidential primary, while almost 50 percent of actual rank-and-file Massachusetts Republican primary voters backed Trump.
The Massachusetts Legislature has 200 members—160 state representatives and 40 state senators. There are currently 125 Democrats and 34 Republicans in the House (Democrat Garrett Bradley stepped down, leaving a total of 159) and 34 Democrats and 6 Republicans in the Senate. All 200 seats are up for election every two years.
While there is an election for every district, many seats go uncontested. Running a quality campaign for a state legislative seat takes organization, time, and money, and unseating a well-known incumbent can be difficult.
For the 2016 general election, there are 18 contested races in the Senate and 54 contested races in the House, with 128 uncontested seats. There are a total of 193 candidates for the 72 contested races. Eleven of the contested seats are open, without an incumbent running.
Rep. Campanale’s district is the race with the highest degree of uncertainty as to outcome—essentially a 50/50 coin flip—based on the model. The model takes into account Campanale’s incumbency and the fact that it is a presidential election year, bringing a more Democratic-leaning set of voters to the polls.
The baseline prediction is computed using my regression model, which uses the district’s presidential margin and an indicator of whether the seat has a Democrat incumbent, is open, or has a Republican opponent as variables. I used the district’s average of the presidential margin in the last two elections as input for the baseline calculation. For example, the average 8-point Democratic margin that Obama won the Senate 2nd Hampden and Hampshire District by in 2008/2012 maps to a 3-point Republican advantage for Donald Humason, when adjusted for the Republican incumbency advantage. An increased margin of 10 points in the presidential race at the top of the ticket would move all of the Republican-leaning races into possible Democratic territory, whereas a 13-point weaker showing by Clinton would put the Democratic-leaning races into striking distance for the GOP
Reps. Campanale and Kelcourse are facing their first challenge as incumbents, while the rest of the incumbents have shown the ability to win multiple challenges in their swing districts. Open seats have historically shown more volatility. There are three open House races and one open Senate race, the Cape and Islands seat currently held by Senator Daniel Wolf. All four of these open seats are currently represented by Democrats and would be highly coveted pickups for the GOP.
The statistical model raises a troubling dilemma for Republican politicians in Massachusetts: support the national party’s divisive nominee and risk damage to the GOP brand, or reject the nominee and put more down-ballot state legislative races in jeopardy.
An average of the Massachusetts presidential election polls as of September 1 puts Clinton’s margin over Trump at 22 points, similar to President Obama’s 24-point win over Romney in 2012. If Clinton’s Massachusetts margin in November ends up in the low twenties, we are likely to end up with a split between Democratic and Republican winners and losers for the races in the table. However, if Trump implodes at the debates and Clinton ends up with a 30-point margin or greater, we could see many of the GOP-leaning state legislative districts moving in the Democratic direction. Alternatively, further narrowing of the Clinton/Trump margin would likely lead to more Republican state legislative winners.
But what if the measurable relationship between the presidential race and state legislative races that has held in the last three presidential years does not hold in 2016? This election cycle has already defied most predictions, resulting in a Republican nominee who has said and done things that would have completely disqualified more typical candidates. Trump is historically unpopular and Clinton has her own approval problems after suffering relentless attacks on her integrity for years.
There are surveys of elections in other states that show Republican candidates for the US House and Senate polling at significantly higher levels than Donald Trump. This might indicate a realization among likely voters that Trump is not a mainstream candidate. Alternatively, voters might be acknowledging the likelihood of a Clinton win and be trying to provide a legislative check to a Clinton presidency by electing Republican members of Congress.
Sens. Elizabeth Warren and Ed Markey are not up for re-election this year, Massachusetts’s representatives in Congress do not face serious competition this year, and Massachusetts voters already have a Republican chief executive in Charlie Baker, who may owe some of his victory to voters trying to balance a Democratically-controlled Legislature.
But history points to a sticky relationship between the Democratic/Republican margin at the top of the ballot, and those in down-ballot state legislative races. An unexpectedly big or small margin between Clinton and Trump will likely show itself in the tightly contested state representative and state Senate battles on November 8.
Brent Benson analyzes politics and public policy in Massachusetts using a quantitative approach on the Mass. Numbers blog (massnumbers.blogspot.com). You can follow him on Twitter @bwbensonjr.A correction was added to this story to reflect two changes to the chart of legislative races.