A Referendum on the President

A Referendum on the President

By Sean Trende - November 5, 2014

There are two basic approaches to evaluating elections.  The first view holds that elections are choices: Voters evaluate the proposals put forward by the candidates, and carefully select the candidate that best lines up with their own views.

The second view, of which I am a longstanding proponent, is that elections are referenda on the party in power.  That is to say, people might give some thought to the other party’s proposals, but their ultimate choice focuses almost entirely on their opinion of the performance of the president’s party. As the great political scientist E.E. Schattschneider put it, “The people are a sovereign whose vocabulary is limited to ‘yes’ or ‘no,’” and it can only speak when spoken to.

Last night’s elections gave a pretty good test of these hypotheses, because we had some well-developed theories on either side that predicted different outcomes.  A variety of “fundamentals”-based models suggested that Republicans should have picked up a large number of Senate seats.  They varied in particulars -- Alan Abramowitz’s model suggested six seats, an early version of the Washington Post model said eight.

Proponents of the choice model suggested that the unpopularity of the Republican brand and/or policy positions would cause the GOP to underperform what the fundamentals suggested. To them, “war on women” messaging could trump fundamentals in states like Colorado (and so forth).  And, for a large portion of the election, it looked as if they might be correct.

But, in the end, the fundamentals won out.  Back in February of this year, I put together a simple, fundamentals-based analysis of the elections, based off of nothing more than presidential job approval and incumbency.  That was it.  It suggested that if Barack Obama’s job approval was 44 percent, Republicans should pick up nine Senate seats.  Obama’s job approval was 44 percent in exit polls of the electorate, and it appears that Republicans are on pace to pick up nine Senate seats. Moreover, only one Democrat -- Natalie Tennant in West Virginia -- ran more than 10 points ahead of the president’s job approval.

Democrats ran about even with the model in the races that were ignored: states like South Carolina, Texas and Mississippi.  Democrats ran well ahead of the model in states that were thought to be competitive early on but where Republican campaigns fizzled over the summer: Michigan, Oregon, Minnesota. They also ran ahead in Kentucky and West Virginia, states where a local Democrat’s base is probably well above the president’s job approval. In the rest of the states they ran 2-3 points ahead of the model, which probably either reflects fundamentals that my rudimentary model didn’t include (the Democrats’ ground game and monetary advantage), or the fact that these things are non-linear at the extremes: A quality Democrat in Iowa simply isn’t going to fall down to the 41 percent showing that the model predicted.

Of course, this doesn’t prove anything, but I do think it is safe to say that the fundamentals-based approaches performed better than the choice approaches. While the Republicans weren’t as unpopular with the electorate as some polls suggested the might be, they were still unpopular; a “choice electorate” probably would have rendered a split decision.

With this, I think we have the answer to the question of “What did the election mean?” The answer: The president received a vote of no confidence.  It was amplified in Senate races because of the playing field, but the overall House vote and the performance of Democratic senators in purple states is consistent with this approach.

But, saying “no” to the party in power is not the same as saying “yes” to the party out of power. Parties have plenty of evidence of this, yet they always seem to convince themselves that “this time is different,” and that the American people have validated their approach.  The success or failure of the new Republican Congress will probably be determined by how well they have learned this lesson.

Sean Trende is senior elections analyst for RealClearPolitics. He is a co-author of the 2014 Almanac of American Politics and author of The Lost Majority. He can be reached at Follow him on Twitter @SeanTrende.

Why Obama Wanted to Appear on "Colbert"
Alexis Simendinger · December 9, 2014
All Is Not Calm
David Shribman · December 14, 2014
Obama, Hillary "Catch Up" at the White House
Alexis Simendinger · December 4, 2014
White House Goes on Offense After Immigration Vote
Alexis Simendinger · December 4, 2014

Sean Trende

Author Archive

Follow Real Clear Politics

Latest On Twitter