Governors Hold the Cards in Congressional Redistricting

In “split-control” states, Republicans won 6.2 percent more seats than expected when they held the governorship; when Democrats held that office, Republicans won 6.5 percent fewer seats than expected.

Americans will elect thirty-four governors to four-year terms this fall.  They will still be in office after the 2020 Census and will have a say in how states redraw their Congressional and legislative district plans. All states where legislatures draw district lines except North Carolina grant the governor the power to veto any plan.  In states where control over the branches is split between the parties, this process should lead to compromises acceptable to both parties.  As well see, however, the evidence from the redistricting after the 2010 Census suggests governors hold all the cards.

In nine states, politicians play no direct role as redistricting is left up to nonpartisan commissions.  Courts, too, can override the lines drawn by legislatures.  This year’s dramatic redrawing of the lines in Pennsylvania follows similar judicial interventions in Florida and New York.  The New York decision affects my analysis since it applied to elections beginning with 2012.  (The other decisions have yet to come into force.)  I have added New York to the commission list, but have analyzed it, and California, separately as well.

I have also excluded the seven states which have only one Congressional district like Wyoming and Alaska since gerrymandering is not possible with no lines to draw.  That leaves 43 states which can be categorized as follows:

  • maps drawn by nonpartisan commissions or courts (8 states);
  • maps drawn by Republican legislatures facing Republican governors (16 states)
  • maps drawn by Democratic legislatures facing Democratic governors (6 states); and,
  • maps drawn when either the houses of the legislature were held by opposing parties, or where the legislature had unified control but faced a governor of the opposite party (13 states).

To avoid relying too heavily on a single year, I have added together the votes cast for Republican and Democratic House candidates in each group of states for the 2012-2016 elections.  I have applied the same method to seats won, again summing up the number of Republican and Democratic seats won across all three elections.  That method produces these results:

House Votes and Seats Won 2012-2016 by Redistricting Method

 

 

 

 

 

 

 

 

 

 

The first column reports the Republican percent of the total two-party popular vote summed across the three elections, 2012, 2014, and 2016.  In the sixteen states where Republicans held both houses of the state legislature and the governorship, they won 56.5 percent of the two-party House vote and 71.7 percent of the seats.  In solidly Democratic states the Republicans won both a minority of the popular vote and of the seats awarded.  The results for commissions and courts is complex; I will deal with it in a later article.

In various articles here I have described the natural inflation of the proportion of seats won due to the operation of our first-past-the-post electoral system.  As parties win larger and larger proportions of the vote, they gain an ever-increasing share of seats.  I have estimated this inflation factor using both biennial election results back to 1946, and across states in 2012.  Both methods produce equivalent results, for instance.

To estimate the share of seats awarded you need only square1 the value of the ratio (Republican Votes)/(Democratic Votes) to get the ratio (Predicted Republican Seats)/(Predicted Democratic Seats).  This approach gives rise to the third column in the table, the proportion of seats that are predicted to be won by the Republicans after applying this “square law” rule.2 In the entry for Republican control, that party’s 56 percent share of the House vote should produce a share of about 63 percent of seats.  In practice, the Republicans won nearly 72 percent of the seats.  The final column measures the over- or under-representation of the Republicans in the House as a percentage gain or loss compared to the predicted share.  In this case, the Republican’s 72 percent is about 14 percent higher than the expected 63 percent. This figure provides a criterion for evaluating how over- or under-advantaged a party was compared to expectations.

The normal expectations for states with unified control are confirmed:  Republicans win a disproportionate share of seats in states where they controlled the redistricting process, and won disproportionately fewer seats than expected in states where the Democrats were in control.  Notice that the size of Republican advantage in states that party controlled is larger than the disadvantage the Republicans faced in states controlled by Democrats, +14 percent versus -4 percent.

By this measure states with some form of split party control show hardly any partisan advantage at all.  Republicans won a share of the seats awarded in these states nearly equal to their expected share.  However, it turns out this overall result hides a lot of significant variation.

We can identify two different forms of split control:

  • ones where both chambers of the state legislature are held by one party but the governor is of the opposite party; and,
  • ones where the chambers of the state legislature are held by different parties.

As it turns out there are nearly equal number of each type of split control; in seven states unified legislatures faced an opposition governor, while in six states the chambers themselves were split.  Once we break out these various patterns, the power of governors becomes clear.  In both types of split control, the governor’s party is disproportionately advantaged during redistricting.  In fact, if we group these split-control states together simply by the partisanship of the governor, Republicans were over-represented in seats awarded by 6.2 percent where they held the governorship; when Democrats held that office, Republicans were under-represented by 6.5 percent.

These results are rather striking.  They suggest that opposite-party governors can force a redistricting map that is actually more favorable to the governor’s party than to the legislature’s.  Similarly when the two legislative chambers are held by opposite parties, it is again the governor who appears to determine which map wins approval.  It appears the governor’s veto is a more powerful weapon in the fight over Congressional district lines than the legislature’s control over drawing the lines themselves.  This fact could weigh heavily over redistricting fights in states like Colorado, Michigan, Florida and Georgia where Democratic governors may win election and end up facing Republican legislatures.  In Massachusetts and Maryland the reverse will likely hold true.

 


1The longitudinal estimate was 2.04; the cross-sectional estimate was 2.08. For simplicity I have rounded down to two, which is well within the confidence intervals for each estimate of beta.

2The tendency for first-past-the-post systems to disproportionately advantage the winning party was first observed in elections in the United Kingdom. There the coefficient reached three, giving rise to the name “cube law” rule, since cubing the ratio (Labour Votes)/(Conservative Votes) does a good job of predicting the ratio (Labour Seats/Conservative Seats). Following this tradition, I have named the US version of this relationship the “square law” rule.

How Well Do Generic-Ballot Polls Predict House Elections?

I have compiled the results of generic-ballot polls taken near to an election and compared them to the actual division of the Congressional vote.  The table below presents the margin between support for the President’s party and support for the opposition.  For each election I have used about half-a-dozen polls from different agencies taken just before voting day.  Averaging the differences between these two quantities shows that these polls have fared rather well since 2002.  The average deviation in the four midterm elections is 0.6 percent; in Presidential years that falls to 0.1 percent.

Still these averages hide some fairly wide fluctuations.  In four of the eight elections the difference between the polls and the election results exceeds two percent.  The error was especially egregious in 2006 when the polls predicted nearly a fourteen-point Democratic margin compared to 8.6 percent in the election itself.

In the most recent election, 2016, the polls predicted a slight positive swing in favor of the Democrats, but the outcome went slightly in the opposite direction.  All the cases where the polls erred in picking the winner occurred in Presidential years and usually when the polling margin was close.  The four polls taken during midterm years all predicted the correct winner, though the size of the victory was off by more than three points in two of those elections.

 

The Election in Pictures

Some data updated through September 2.

Sunday, July 29th, marked the point when there are just 100 days left until the November midterm.  In this post I will try to pull together my various writings and predictions for both the 2018 House and Senate elections.  I begin with the most important factor that influences both types of races, the President’s job-approval figures.

Presidential Approval

The President’s job-approval rating is an important predictor of midterm election results in both my model for Senate elections and the one for House elections.  The Democratic advantage on the “generic-ballot” question about voting in House elections has waxed and waned as Donald Trump’s approval rating first fell after he was inaugurated then rose again over the past few months.

Donald Trump enters the 2018 election with about the same level of public support Barack Obama had in 2010. Though Obama was much more popular when he was inaugurated, that goodwill faded over the following eighteen months. In the 2010 midterm that followed, the Democrats were “shellacked,” losing sixty-three seats in the House of Representatives. Support for Trump also ebbed away during 2017, but he has rebounded slightly from his nadir last December.


Though Trump’s overall approval rating heading into the first midterm is largely identical to Obama’s, Trump is more intensely disliked.  Pew reported that the proportion of people saying they “strongly” disapproved of Obama’s performance in office grew from 18 percent in April, 2009, to 32 percent by September, 2010.  For Trump, CNN found that he took office with over forty percent of Americans already strongly disapproving, a figure that has remained relatively unchanged.  In its June, 2018, poll CNN reports a “strongly disapprove” figure of 45 percent.

The House of Representatives

As the first chart shows, the president’s job-approval rating bears directly on the “generic-ballot” question.  Based on polls through September 2nd, the Democrats’ lead on the generic ballot has grown slowly since Inauguration Day and inversely with Presidential job approval.  The grey area in the chart below represents the likely range of outcomes.  Since April, Donald Trump’s net approval rating has ranged from about -8 to -14.  Those values define the left and right sides of the shaded region.  When I include estimates of any methodological or unique “house effects,” I find that polls conducted over the Internet show a pro-Republican tilt of about 2.6 percent.  Gallup’s polling shows an ever greater Republican edge of nearly five percentage points.  I use the values for live polling, which are most favorable to Democrats, and those from the much-less favorable Gallup, to define the height of the grey area.

Notice that, according to these results, even if Trump were to achieve a net approval rating of zero, Democrats are still predicted in live polling to lead on the generic ballot by about six points on Election Day. That reflects the slow growth in support for Democratic House candidates over Trump’s presidency from about four points on Inauguration Day to a predicted six points this November. In past elections the margin of victory in generic-ballot polls has proven to be a pretty accurate predictor of the actual division of the vote.

An earlier version of this model showed a small, marginally significant positive boost for Democrats in polls of likely voters.  That difference has disappeared as the number of polls has increased.  Since Democrats are generally disfavored in likely-voter polls, especially ones conducted in midterm years, a finding of no difference between registered and likely voters is actually positive news for Democrats.

One other major problem for House Republicans has been the historically large number of their Members who are leaving, or have left, the House.  Forty Republicans will not be returning to the House next January creating an excess number of more vulnerable open seats on that side of the aisle.  Only 18 Democrats are leaving the chamber.  With 22 more retirements than the opposition, the largest midterm gap since the New Deal, the Republicans face a loss of 39 seats based on the historical relationship between the two measures.

The Senate

There is no national generic-ballot question for Senate elections because only two-thirds of the states have a Senate race in any given year.  Looking back historically over Senate elections, the fate of the President’s party depends directly on his job-approval rating and, unlike for House elections, the state of the economy as measured by the growth in real disposable personal income per capita.  Any plausible combination of approval for Donald Trump and income growth predicts that the Republicans will fail to win a majority of the popular vote for Senate in November.  One reason is that the popular vote for Senate candidates of the President’s party runs four points lower when the President is not on the ballot.


Much has been made of the 4.1 percent increase in Gross Domestic Product reported for the second quarter of the year.  That figure represents the growth in nominal GDP; after adjusting for inflation the figure is 2.8 percent. Unfortunately for the Republicans little of that growth appears to be “trickling down” to ordinary Americans.  Here are the recent trajectories for both real GDP and real per-capita disposable personal income.

Personal income has hovered around a two-percent growth rate for the last three quarters, while real GDP grew more quickly.  If voters respond to changes in the amount of money in their pockets, then the economy will not be sufficient to power the Republicans to victory in the fall.  From the chart above, a two-percent growth rate in per-capita income and even a 45 percent approval rating for Donald Trump still leaves Republican candidates short of a majority in the national popular vote for Senate.

From Votes to Seats: The House

Americans became much more cognizant of the word “gerrymander” after the redistricting that followed the publication of the 2010 Census.  Drawing lines for partisan advantage has become easier as voters have segregated themselves geographically by party.  Together the two forces have combined to create a Republican “bulwark” in the House. Democrats need to win the popular vote by more than fifty percent to take half the seats in the body.  How much more is subject to debate, but most estimates put the needed margin of victory in the range of six-to-eight percentage points, or an election where the Democrats win somewhere between 53 and 54 percent of the popular vote.

For instance, The Economist currently projects the Democrats to win 54.3 percent of the popular vote, or a margin of 8.6 percent, but take just 51.3 percent of the seats in the House.  Both Dave Wasserman at the Cook Political Report and Nate Cohn at the New York Times cite a seven-point margin as the minimum required for a slim Democratic win. My model agrees.  It uses historical voting data back to 1940 to estimate the relationship between seats and votes. I include adjustments for redistricting after each Census and for the decline in political competition since the 1994 “Contract with America” midterm. Using those data I estimate the Democrats need at least 53 percent of the national popular two-party vote to win a majority in the House of Representatives.

From Vote to Seats: The Senate

Unlike House districts, Senate seats cannot be gerrymandered because they constitute entire states.  That makes the Senate more competitive than the House.  There is no equivalent “bulwark” in the Senate; winning half the popular vote generally translates into about half the seats.  Since the model predicts that the Democrats will win a majority of the 2018 popular vote for Senate, we should expect the Democrats to win a majority of the 35 Senate seats at risk.  Winning 18 of those seats would not be enough to flip control of the Senate because of the Vice President’s tie-breaking vote.  Nineteen seats would give the Democrats control.


From Here to November

Can the Republicans rebound between now and election day?  Unfortunately, recent history suggests they will face even larger obstacles in November than they do todayPresidential approval generally declines as the election nears, and the opposition party’s advantage on the generic ballot grows.

Job approval for first-term presidents fell on average about five points between May/June and October of midterm election years.  Trump might see a smaller decay because his current popularity is historically low, around 42 percent.  Presidents whose approval was above fifty percent in May/June saw a 3.3 percent drop in approval; those who started below fifty percent in May/June saw their ratings fall an average of just 1.9 percent.


In off-year elections, generic ballot polls for both of Obama’s midterms fell as the election neared.  For George W. Bush in 2006, Republicans recovered slightly from their early summer deficit, then watched support for their candidates crater in October.  Most observers credit that sharp decline to the Mark Foley scandal that fall.

 

From Job Approval to the Ballot Box

“Generic-ballot” polls predict a nine or ten point Democratic victory in House elections this fall, enough to flip the chamber.

As most observers know, support for the Democrats on the so-called “generic-ballot” question has moved inversely to the public’s opinion about Donald Trump’s performance in office.


Can we use this link between support for the parties on the generic-ballot question and Trump’s job approval figures to forecast possible election results in November?

There are 228 generic-ballot polls in my current dataset covering the period from Inauguration Day, January 20, 2017, through June 9, 2018.  Of those 228 polls, 209 also collected data on presidential approval.  Those 209 polls constitute the sample for this analysis.

A weighted-least-squares model using my standard predictors — days in office, polling method, and polling sample — plus the net Trump job approval score “explains” about half the variance in the size of the Democrats’ lead on the generic-ballot question.  Overall, the Democrats’ lead has increased at a slow, but discernible pace since the Inauguration.  Even after taking Trump’s job approval into account, I find support for the Democrats grows about 0.3 percent every hundred days.  At that pace, the Democrats will have picked up about 2.2 percentage points by the time the election takes place on November 6th, some 655 days after Trump took office.

Two other factors influence the size of a poll’s Democratic lead.  The Democrats do about 2.2 percent worse in online polls than in ones conducted by telephone. On the other hand, polls restricted to “likely” voters show a marginally significant (p < 0.11) boost for the Democrats of 1.3 percent.  That might indicate a slight Democratic advantage in voter mobilization going into the fall, but the effect is still too small, and too variable, to be taken seriously at the moment.

These factors remain constant in the face of changes in Trump’s job approval and the passage of time.  We can thus set them at their values on Election Day and see how the margin in the generic House ballot varies with changes in Trump’s job-approval rating.  This chart presents the likely Democratic lead on the generic-ballot question in polls conducted with live interviewers on Election Day, November 6th. The dotted line represents the effect of adding the small, marginally-significant boost for polls of likely voters:


As of July 20, Trump’s net job approval (approve-disapprove) is averaging about -11 (42-53). My estimates indicate that difference translates into a Democratic margin of 8.8 percent in generic-ballot polls.  Notice that the remaining factors in the model predict a Democratic margin of about six to seven percent even if Trump’s net approval score is zero. That might be a decent estimate of the so-called “blue-wave.”

Models of the relationship between House votes won and House seats won find that, to take back the House, the Democrats will likely need to win by a margin of +7 or better to overcome gerrymandering and partisan geographic self-selection.  For instance, The Economist’s simulation model for the House midterms predicts that the Democrats will win 54.3 percent in November, or a margin of 8.6 percent.  In their model that gap translates into a Democratic seat margin of 224-211, enough to retake control of the House.

History suggests that both the President’s approval rating and support for the President’s party in the House are lower on election day than they are in the spring.

 

Are Some Republicans Leaving Donald Trump?

Recent polling suggests Trump has been losing support among Republican voters since the spring.  Likely Republican voters show less support than other Republicans.

It has become a commonplace among journalists and pundits to observe that Republican voters have remained largely behind President Trump.  Recent polling still shows job-approval ratings for the President among Republicans remaining in the 85 percent range.  But that focus on individual polls obscures a more complex trend, one that does not bode well for President Trump and his Republican Party.

This graph presents the “net approval” score (percent approving minus percent disapproving) for Republican voters in polls that disaggregate their results by partisanship.  Like in the country at large, support for Trump declined during 2017 but has rebounded this year.  (These data end before the decision to separate children and parents at the southern border became a national news event.)  I estimated the trajectory of support using a fourth-order polynomial based on time in office, with the usual array of dummy variables to adjust for polling methods and “house effects.”

The bold line representing the President’s approval rating among likely Republican voters should be especially troubling.  The Republicans most likely to turn out in November average about five points lower on net job approval than do other Republicans.

Even casual examination of the data points displayed here show that Republicans’ opinions about Trump’s performance in office have displayed wide variability since he entered the Oval Office.  However the most recent polling shows a decline in approval since the spring of 2018.

How Popular will the President be in November?

Most presidents are less popular when mid-terms are held than they are in late spring.

In an earlier piece about the “generic” Congressional ballot question, I presented the trends in support for the parties over the course of the summer for a few mid-term elections.  However that question has been asked on a more limited basis than the standard Presidential job-approval question, which Gallup began asking back when FDR served.  As I showed earlier, trends in the job-approval question track inversely with changes in the generic-ballot item. Given the tie between presidential popularity and mid-term outcomes, we might ask how the job-approval question has tracked before mid-term elections.  Given that President Trump’s approval rating stands at a bit under 42 percent today, where might it be come November?

I have compiled Gallup’s job-approval scores for all mid-term elections back to Harry Truman’s second term in 1950.  I have calculated separate averages for first-term and second-term presidents since the additional experience voters have with second-term presidents should limit the effects of most events before the election.

As expected, first-term presidents show a larger change in their job-approval scores than second-termers.  Indeed, excluding the anomalous 1974 election after Nixon resigned, second-term presidents saw their popularity grow on average by about a percentage point.  Nearly all first-term presidents saw their popularity decline as the election grew near, falling an average of 3.8 percent between May and October.

Among first-term presidents, only Jimmy Carter had an approval rating as low as Donald Trump’s at the end of the spring.  Carter saw his rating improve over the course of the summer, but much of that rise came after the Camp David summit with the then-presidents of Egypt and Israel in September, 1978.  Otherwise first-term presidents generally see a decline of about four points on average in their job-approval scores as the summer wears on.  (Leaving out Carter in 1978 raises that figure to five.) On that basis we might expect to see Trump’s job-approval score in the high thirties come Election Day.

One mitigating factor might be that Presidents who start off with ratings under fifty percent show smaller declines than those with a majority of citizens approving of their performance in the spring.  Presidents whose approval score was above fifty percent in May/June watched their ratings fall an average of 4.5 percent by October.  For those Presidents, like Trump, starting with a rating below fifty the average decline was just 1.6 percent.

Don’t Count Those Chickens Just Yet

Support for the President’s party on the generic House ballot is historically lower on election day than in the spring.

Most “generic ballot” polls attempting to forecast the 2018 general election have shown a narrowing of the gap between Democrats and Republicans in the past month or two.  The gap was widest at the end of 2017 when Trump’s job-approval numbers reached their lowest point.  While Republicans have regained a bit of ground since then, Democrats still have about a 5-7 point lead in recent polls.

Before Republicans get too excited by this rebound, we should examine the trends for the Trump Administration in the context of previous elections.  Here are two charts that depict the trends in the difference between support for the President’s party and support for the opposition in generic ballot polls.  The data for 2006 and 2008 comes from RealClearPolitics; the remainder comes from the archives at HuffPost Pollster that I have used in earlier postings.  In each case I have averaged polls by month, combining together May and June, and July and August, when polling is less frequent than in the fall.

For both Obama years, 2010 and 2014, we see a fairly linear decline in the Democrats’ margin over the Republicans on the generic ballot.  Unsurprisingly the fall in Democrats’ fortunes was much more substantial in 2010, when Obama observed that his party had been “shellacked” in the midterm.

In 2006 the Republicans faced a double-digit deficit in the spring.  Though they shaved a couple of points off the Democrats’ lead by September, the Mark Foley scandal ended any hopes of a Republican come-back.

The Republicans begin the summer of 2018 facing a five-point deficit which might not be enough to swing the House of Representatives to the Democrats.  If the history of past presidents is to be believed, though, the Republicans’ prospects may worsen as we head into November.

 

 

The “Generic” Congressional Ballot Question

Democrats would lead by nearly fourteen points in “generic” Congressional ballot polls next November if the trends seen since Trump took office continue.

I have written earlier about how methodological differences among pollsters can lead to significantly different results.  In my analyses of Presidential approval I showed how Donald Trump’s approval ratings varied depending on the choice of sample to interview and the interviewing method chosen.  In this piece I apply the same approach to the so-called “generic” ballot question, typically “If the elections for Congress were being held today, which party’s candidate would you vote for in your Congressional district?”  Some pollsters mention the Democrats and Republicans in this question, others leave it more open-ended like the example I just gave.

I have focused on the net difference in support for generic Democratic and Republican candidates.  This ranges from a value of -4 (Republican support being four points greater than Democratic support) to a high of +18 in the Democrats’ direction.  Here is a simple time plot showing how support for the Democrats on this question has grown while Trump has held office.

The Democrats held a small lead of just under four points on the day Trump took office.  Since then the Democrats’ lead has slowly increased to an average of eight points.

What’s surprising about these data is that they do not show the usual methodological differences we see in the presidential series.  Here are a few regression experiments using my standard array of predictors.

Choice of polling method has no systematic relationship with Democratic support on the generic ballot question. In contrast, Trump’s job-approval ratings run one to two points higher in polls taken over the Internet.  Another striking difference is the greater level of support found for Democrats in polls of registered or “likely” voters.  Again, the job-approval polls show an opposite effect, with polls of voters displaying greater levels of support for Trump than polls that include all adults.  I have also included separate effect measures for the two most-common pollsters in this sample, Politico/Morning Consult and YouGov/Economist.  Job-approval polls taken by the former organization show a pro-Trump “bias” of about three percent; on the generic ballot their polls place Republican support about five points higher than other polls.  YouGov/Economist polls also have Republican tilt on this question, though they show a slight anti-Trump bias in job-approval polls.

If we extrapolate these results to the fall election on November 6th (655 days after the Inauguration), and include the effect for registered voters, the model predicts the Democrats’ lead in generic ballot polls would reach nearly fourteen percent (=4.07+2.62+0.011*655).  A margin that large would easily overwhelm the built-in advantage Republicans hold based on partisan self-selection and gerrymandering.  Even if the Politico figure is correct, adding in that pro-Republican factor brings Democratic support down to nine points on election day.  That result would still reach nine percent, or a Democratic/Republican split of about 54-45.  That 54 percent figure still exceeds the 53 percent minimum I estimated earlier would result in Democratic control of the House of Representatives.

Using the model for the relationship between seat and vote divisions presented earlier, a 57 percent margin in the national Congressional vote would translate into the Democrats’ winning 55 percent of the House seats for a margin of 239-196.

Republicans Continue to Leave the House

Three more Republican House Members announced they would not seek re-election this week, bringing the total number of retiring Republican Members to 34 according to the New York Times. That figure compares to 16 Democrats, for a net Republican difference of +18.  We have to look back to the Democratic landslide in 1958 to see a mid-term with double-digit net Republican retirements.  For Democrats, only in 1938 and 1978 did the number of their retirements exceed Republican retirements by ten or more.

This increase of three net Republican retirements raises the predicted Democratic seat swing to 41 using the relationship depicted in the previous article.

I have shown in earlier postings that the relationship between seats and votes that advantaged Democrats in the years after World War II moved steadily in the Republicans’ direction beginning in 1980 and, with the help of gerrymandering, became even more favorable for the Republicans after 2010.  That might temper our belief in a prediction for an election being held in 2018.  First, the 2018 retirement margin of -18 is close to the observed maximum of -21 in 1958.  Perhaps the 1958 election is an “outlier” and without it the relationship is less steep than we observe.  However slope and intercept coefficients estimated with 1958 excluded are numerically nearly identical to those estimated with that year included.  So it’s unlikely that the historical model is radically overestimating the likely result next fall.

Another test is to let the relationship differ before and after 1992 to see whether the structural changes that we observe in the seats/votes relationship in the current era appear in the relationship for retirements and seat swings.  Once again, allowing the coefficients to differ before and after 1992 showed no measurable statistical difference. While the effects of gerrymandering and partisan self-segregation may make the House less vulnerable to “waves” of Democratic support, there is no evidence for that thesis looking at retirements as a predictor of seat outcomes.

These estimates have a lot of uncertainty attached.  The standard error of estimate is about 28 seats.  That means there is about a two-thirds chance that the actual swing will be somewhere between 13 and 67 seats.  Since the Democrats need a swing of at least 24 seats to win control of the House, even a retirement margin of 18 is not enough to ensure a change in party control.

The regression model taking President’s partisanship into account is a bit more conservative; it predicts a swing of 37 seats.

Update (2/26/18) – One more Republican has announced he is leaving the House, along with one more Democrat.  The net difference remains at +18.

What Do House Retirements Tell Us About the Future?

The pace of Republican retirements predicts that Democrats should take back control of the House of Representatives this fall by a margin of eight or nine seats.

This week Edward Royce (R-CA), Chair of the House Foreign Affairs Committee, and Darrell Issa (R-CA), former Chair of House Oversight and Reform, joined 27 of their fellow House Republicans by announcing that they are retiring from the chamber.  About half are leaving public office entirely, while the remainder are seeking another office like governor or senator.  (Issa has threatened to run for the House again in an adjacent district.)

On the other side of the aisle, fourteen Democrats have announced that they will be leaving the House of Representatives.  Only five of them are leaving public office, though Ruben Kihuen (D-NV) may be joining them subject to an investigation into allegations of sexual harassment.

Many pundits have interpreted the much higher retirement rate for Republicans to be a bellwether for this fall’s Congressional election.  If no one else announces a retirement, the Republicans will face a net loss of fifteen seats going into the midterms.  Just how large a threat do such retirements pose to Republican control of the House?

These two charts show the relationship between the net Republican margin of victory in terms of seats in the House and the net partisan difference in retirements.  It turns out that retirements tell us essentially nothing about Congressional outcomes in Presidential years, but they are quite informative in mid-term elections like 2018.  Here is the chart for Presidential years:

In presidential years we see little relationship between net Republican retirements and how well the party fares in the upcoming general election. What matters more are Presidential “coattails” with Republican swings in years when Eisenhower (1952), Nixon (1960, 1972), and Reagan (1980) ran.  Democrats were favored when they ran along side Johnson in 1964, Obama in 2008, and Franklin Roosevelt in 1944.

A much different picture appears if we look at the same relationship for mid-term years.

Now the number of Republican seats won or lost depends much more directly on the number of retirements.  The line is anchored by the Republicans’ success in the 1938 midterm and their dramatic losses in the 1958 election. If the figure for net retirements remains at fifteen, the Republicans are predicted to lose about thirty-two seats next November.  That would give the Democrats control of the House with a margin of eight seats.

However, because the President’s party historically loses seats in midterms, we should expect to see more retirements from the President’s party in midterm years.  When Democrats presided over a midterm, an average of four more Democrats retired from office than did Republicans.  When the Republicans held the White House in a midterm year, retirements from their ranks outnumbered Democratic retirements by an average of six.

So some of the strong relationship we see between retirements and midterm losses arises simply because the President’s co-partisans are jumping ship knowing that their party will do more poorly in the upcoming midterm.  That leaves us with the question of whether retirements have any additional predictive power once we take the President’s partisanship into account.  Retirements still matter in this better specification, though that effect just achieves statistical significance.


Based on these estimates, in a year where the Democrats hold the White House, and the number of retirements on both sides of the aisle is equal, the model predicts that the Republicans should gain about thirty seats.  When the Republicans hold the White House, and retirements are equal, the Democrats should gain about twelve seats (= 30.2 – 42.1 = -11.9).  Regardless of which party controls the Presidency,* Republicans are predicted to lose 1.4 seats for every retirement.  Applied to the current circumstances, this formula predicts a Democratic victory by thirty-three seats, or one more than predicted by the simpler model.

 

 

*Allowing the relationship between retirements and seat outcomes to vary separately depending on which party controlled the White House added no explanatory power.