The Strange Case of 1976

1976 was a horrible year for Senate Republicans; adjusting for that fact makes a slight difference to my 2018 predictions.

Re-examining the results for my original model of Senate elections, it was hard to ignore how poorly the model fit the data for 1976.  Here is a graph of the model’s predicted vote for Senate and the actual vote that shows what an “outlier” 1976 is.  While Truman rallied Senate Democrats in 1948, even that event just hovers on the edge of the statistical “margin of error.”  The Republicans’ failure in the 1976 election after Richard Nixon was forced from office stands truly alone compared to the rest of the postwar elections in my dataset.

If 1976 had been a normal presidential election year, the Republicans’ Senatorial prospects would have looked fairly rosy.  Gerald Ford was running for re-election, real personal income was growing at two percent, and the Democrats were defending seats won in 1970 by a (two-party) margin of 56-44 at the height of the anti-war and anti-Nixon fervor.  That generally pro-Republican climate predicts the GOP should have won nearly 52 percent of the popular vote for Senate.

But, of course, the 1976 election was anything but normal.  It was the first presidential election after the Watergate scandals had forced Richard Nixon from office in disgrace.  Rather than winning the popular vote by the predicted four-point margin, the Republicans could muster only the same share of the vote they won back in 1970, 44 percent.  Though a number of seats changed hands, at the end of the day the Democrats held the same 61-seat Senate majority they did before the 1976 election.

I can adjust statistically for the anomalous 1976 election by adding a “dummy” variable to my model that is one in 1976 and zero otherwise.  Adjusting for 1976 radically improves all aspects of my model.  Its predictive power as measured by adjusted R-squared rises from 0.43 to 0.56, and all the coefficients are more precisely estimated.

Adding this dummy variable implicitly treats Gerald Ford as different from other Presidents running for re-election.  Ford was apparently so compromised by Watergate that his presence at the top of the ticket did not generate the kind of support his fellow Republican candidates for Senate might have expected.  With the 1976 adjustment, the overall effect of Presidential candidacies rises from 2.5 to 3.1 percent, suggesting Ford’s performance was suppressing the estimate for other Presidencies.

Adjusting for 1976 also increases the compensating effect (“regression-toward-the-mean”) of the prior vote for a Senatorial class.  With the suppressing effect of 1976 removed, I now estimate that the Democrats’ lopsided Senate victory in 2012 should be worth about 2.3 percent to the Republicans this November, compared to the 1.9 percent figure I presented earlier with no adjustment for 1976.

For comparison to the chart above, here are the predicted and actual values for the model adjusted for 1976:

Including a dummy variable for 1976 sets its residual to zero and places its predicted value precisely on the line.  The largest positive outliers are now 1978 and two Presidential years, Truman in 1948 and Barack Obama in 2016.

The effects of this modification on the predictions in my earlier article are quite modest.  Without adjusting for 1976, I predicted the Republicans will win 48.1 percent of the popular vote if Trump’s approval rating is at forty and real disposable personal income rises two percent.  With the adjustment that figure rises to 48.4 percent.

The overall conclusion remains that no likely combination of factors predicts that the Republicans will win a majority of the popular vote for Senate this fall.

 

 

This Dog Won’t Wag

Donald Trump’s polling figures showed little change after last year’s attack on Syria.

The 1997 movie Wag the Dog introduced the public to a notion common among pundits and historians, that Presidents like other political leaders might engage in military actions overseas to distract from problems at home.  With a President beset from many sides, pundits have opined whether Friday’s attack on Syrian chemical weapons facilities might have been motivated, at least in part, by the same diversionary motive.

Friday’s action reprises the Tomahawk missile attack conducted by American forces against a Syrian airfield a year ago.  That equally brief sortie had little short-term effect on President Trump’s job-approval ratings and no long-term effects at all.  Here is the trajectory for “net approval,” the difference between the percent of American adults who say they approve of how the President is “handling his job” minus the percent who disapprove.1  President Trump has been “underwater” since his Administration began with more Americans disapproving of his performance than approving.

The most optimistic “wag-the-dog” interpretation of this chart might credit a two or three point positive bump in President Trump’s net approval rating after the attack on the Syrian airfield on April 7th of last year.  Yet a couple of other prominent events might give us pause.  Trump’s approval rating actually improved after Michael Flynn had to resign as National Security Advisor, but it soon fell back down. Last year’s attack on Syria itself took place at a time when Trump’s rating was already recovering.  Perhaps the apparent gain after the attack was just the continuation of that trend.

When partisans are routinely described as “tribal,” we can hardly expect many of them to change their minds about President Trump just because of a single military attack. Republicans endorse his performance in office by margins of 85-10, while Democrats disapprove by an even greater margin of 9-90 unfavorable. For people so solidly entrenched in their partisanship even a missile attack against a sovereign nation can have little sway.

Perhaps, then, we should look at the opinions of self-described “independents.” Maybe their opinions will be more sensitive to current events like a strike on Syria. Sadly those seeking a wag-the-dog effect will once again be disappointed.


Independents’ appraisals of President Trump tracked rather closely to the Gallup figures above (with the exception of the weird spike at the turn of the year).2 Again, whatever small gain the 2017 Syrian attack may have imparted to Trump’s approval ratings among independents quickly dissipated a few weeks later.

 


1Since the beginning of this year, Gallup has reported only weekly job-approval ratings. I appended the 2018 data to the 2017 weekly averages.

2Because Gallup does not report partisan breakdowns, I have averaged together the soundings archived at Pollster that do. This sample includes 313 polls that reported approval ratings for independents. They were conducted by thirteen different organizations, with four each constituting about a sixth of the observations (Politico/Morning Consult, YouGov/Economist, IpsosReuters and SurveyMonkey).

Can the Republicans Hold the Senate in 2018?

Historical voting patterns and current economic and political conditions predict the Democrats can win 18 of the 33 Senate seats in contention this fall and perhaps take back control of that body.

The departure of Paul Ryan as Speaker provides yet another piece of evidence in favor of the Democrats taking back the House of Representatives this fall.  Faced with this prospect Senate Majority Leader Mitch McConnell and other Republican legislators and strategists argue that the party must now focus its efforts on maintaining its one-vote lead in the Senate.  In this essay I examine the prospects for a Democratic Senate victory as well.

At first glance, the Democrats’ prospects seem quite poor. The current “class” of Senators running for re-election last faced their voters in 2012 when Barack Obama was re-elected President. That helped the Democrats win or retain a number of seats in unlikely places like North Dakota, Missouri, and Montana. Twenty-three Democrats were elected to the Senate that year, compared to eight Republicans and two independents.  With so many more Democratic seats up in 2018, we might expect the GOP to maintain or even expand its slim Senate majority this fall.

However the Democrats themselves held a similar advantage in the 2016 election but failed to win back the Senate. The Republicans needed to defend the 24 seats they won during their “shellacking” of the Democrats in 2010, while just ten Democratic seats were at risk. Nevertheless the Democrats managed to swing only two seats into their column in 2016, suggesting that the relative number of seats at risk may not be a very powerful predictor of the eventual outcome.

One factor in the Democrats’ favor is the absence of Donald Trump at the top of the ballot come November.  Updating the chart from my previous article to include the 2016 election does not change this basic fact:

The Democrats did better than expected in the 2016 Senate elections, winning 53.7% of the popular vote.  That raised the average for the six “open seat” elections, when the President was not standing for re-election, to 48.8% from the figure of 47.9% I reported in 2016.  In general, though, having the President at the top of the ticket adds, on average, four percentage points to the vote for his co-partisans in the Senate. The Republicans are thus starting from behind despite the disparity in seats at risk.

Along with whether the President is standing for re-election, I identified three other factors that have systematically influenced the vote for Senate candidates since World War II:

  • the size of the previous popular vote for the current “class” of Senators facing re-election;
  • the President’s job approval ratings in polls near mid-term elections; and,
  • the year-on-year change in real disposable personal income per-capita.

That framework enables us to examine some possible scenarios for this fall.

There is some evidence that the size of the vote for a Senatorial class does influence how well that class fares six years later. The party advantaged in one election sees a drop in support when facing re-election.  In 2018 that factor helps the Republicans by about 1.8 percent, still not enough to overcome the deficit from not having the President at the top of the ticket.

As I find for House elections, a President’s “job-approval” rating influences the outcomes of Senatorial races.  A ten-point increase in the percent of Americans approving of the President’s performance results in a 1.2 percent increase in support for Senatorial candidates of the President’s party.

On the economic front, the Republicans have repeatedly touted their recent tax cut as providing an impetus to support for their party. As Bloomberg reported after a GOP retreat in February, “With President Donald Trump’s stubbornly low approval ratings and historical trends suggesting they’ll lose seats in the November mid-term elections, party leaders told lawmakers their salvation lies in hammering on the message that the tax cuts passed at the end of last year are putting more money in voters’ pockets.”

My earlier findings confirm that growth in real disposable personal income does have an effect on vote for the President’s co-partisans in the Senate.  In the simulations below, raising personal income growth by one percentage point yields an increase in the predicted vote for Republican candidates this fall by about 1.4 percent.

We can put these findings together and examine the model’s predictions for the Senatorial vote in 2018 given different combinations of presidential job-approval and growth in personal income.


or, visually,

In no likely scenario does my model predict a popular-vote majority for Republican Senate candidates this fall. Which of these scenarios might we see play out in November?

The job-approval figures I use in this analysis come from Gallup, since only that organization has published these ratings as far back as the 1940s.  Like most other pollsters Gallup has reported a slight rebound in Trump’s approval figures over the past couple of months, but they are still running around forty percent.

Gallup has begun reporting weekly ratings this year which explains the much lower variability of its 2018 results.  Also Gallup’s polling shows no systematic partisan bias in either direction but hews closely to the polling “consensus.”  However, Gallup may underestimate approval for President Trump by about 1.6 percentage points because it surveys all adults rather than just registered voters.

Personal income growth, despite the tax cuts, does not provide much reassurance for the Republicans either. Real disposable personal income per capita has shown only modest growth over the past few months, running at about 1.5 percent on an annualized basis.


While there has been a slight downtick since the January peak of 1.6 percent, it seems plausible that real personal income per-capita will show a gain in the neighborhood of 1.5 percent by November.

From the table of predicted vote outcomes, a job-approval rating of 40 and even two percent personal income growth corresponds to a 52-48 split in the popular vote favoring Senate Democrats this fall.  Because there is little “bias” in the translation from votes to seats in the Senate, winning 52 percent of the popular vote translates into a two or three seat margin this fall.


Using the equation shown in that chart, if the Republicans win 48 percent of the popular vote for Senate this fall, they should come away with only fifteen seats (45% of the 33 seats at risk).  A three-seat victory for the Democrats would flip control of the Senate even given the Vice President’s deciding vote.