rate is generally not a good predictor of how presidential elections will turn out, but the more recent changes in the unemployment rate are more influential. Figure 1 shows the relationship between changes in the unemployment rate during the year before each election since 1952 and the incumbent party’s share of the two-party vote in those elections. The line in the plot shows a strong trend that when the unemployment rate is increasing in the twelve months leading up to the election, the incumbent party tends to receive a lower percentage of the vote; but when the unemployment rate is dropping, the incumbent party performs better. In 2016, the unemployment rate had dropped by 0.3 percentage points from where it had been at the same time in 2015. Based on previous patterns, this meant that Clinton should win about 51 percent of the two-party vote. As of our writing of this supplement, she had actually received 50.9 percent of the two-party vote. Thus, the election results were very much in line with what we might have expected from standard economic indicators.
Even as economic conditions improved markedly during Obama’s second term, Obama did not enjoy a significant boost in his approval rating as a result. A president’s approval rating is the percentage of Americans who say that they approve of the job he is doing as president. When presidents enjoy higher approval ratings, their parties tend to perform better in elections. For example, in 1988, George H. W. Bush won the presidency partly because approval of Ronald Reagan was above 50 percent. In 2000, Al Gore won the national popular vote (though not the presidency) thanks in part to outgoing President Bill Clinton’s nearly 60 percent approval rating. In 2008, John McCain struggled against Barack Obama because of President George W. Bush’s very low 25% approval rating. As 2016 began, Obama’s approval rating steadily hovered just below 50 percent. He was less popular than Clinton in 2000, but much more popular than Bush had been in 2008.
Figure 1 The Relationship between Change in Unemployment and the Incumbent Party’s Share of the Two-Party Vote for President
Note: Figure created by authors using data from the Federal Reserve. X-axis shows change in unemployment rate from 3rd quarter of year preceding election to 3rd quarter of the election year. Y-axis is proportion of the vote for the major party candidates received by the candidate from the party that held the White House.
In short, Obama was popular enough and the economy was strong enough to give a Democratic candidate for president a reasonable shot at winning in 2016, but Obama was not so popular that this would be an easy victory for the Democrats. Adding to the challenge for Democrats was the fact that a third factor also tends to play a role in presidential elections—voters tend to not return the same party to the White House for three consecutive terms. Indeed, since Harry Truman won re-election in 1948 there has only been one instance when presidents from the same party won the presidency in three consecutive elections (Republicans did so in 1980, 1984, and 1988). This pattern likely arises because voters develop a desire for change over time, and also because a party’s supporters may become complacent when their party wins several elections in a row. Thus, Democrats were also facing the challenge of trying to retain control of the White House for a rare third consecutive term.
Choosing the Presidential Nominees
The balance of the fundamentals heading into the 2016 presidential campaign suggested a contest that could go either way. Some political science models indicated that Democrats would win a narrow victory in the national popular vote, while other models suggested a narrow advantage for Republicans. Nearly every model indicated that it would be a very close contest.4 Thus, political scientists expected that 2016 would produce a tight contest for the White House; and given that fact, one thing that was likely to matter was who the parties would nominate to run for the presidency.
The Invisible Primary
Political parties select their presidential nominees through a lengthy and arduous process that begins with the so-called “invisible primary.” During the invisible primary, candidates maneuver to set themselves up for success in the actual primaries and caucuses held in the states during the first several months of an election year. Those primaries and caucuses are where candidates attempt to accumulate a majority of convention delegates in order to secure their party’s nomination. But success in the invisible primary generally comes by raising large sums of campaign funds from donors, performing well in the preelection polls, and lining up endorsements from party officials. A candidate who is able to accomplish these things should have sufficient name recognition and public support to perform well in early primaries and caucuses and the financial support necessary to sustain that early success over a long nomination campaign on his or her way to becoming the party’s nominee.
While candidates jockey for position during the invisible primary, the leaders of each party generally attempt to influence the nomination process to ensure that the strongest candidate will win the nomination and represent the party in the general election campaign. This theory about the role that party elites play in the nomination processes is generally called “The Party Decides,” after the name of the influential book that describes the theory.5 Essentially, the argument is that activists, politicians, and other central players in each party tend to line up their support behind the candidate who they think would run the strongest general election campaign. In past elections, when these party leaders rallied behind a favored candidate by endorsing that candidate during the invisible primary, the favored candidate has usually prevailed in winning the nomination. For Republicans, this pattern has held in every election at least as far back as 1980. A notable exception for Democrats was 2008, when Hillary Clinton held a wide lead in endorsements over Barack Obama before the contests began but ultimately lost the nomination to him. In 2016, it was on the Republican side where the theory would once again fall short.
Hillary Clinton’s success in the Democratic primaries was in many ways a textbook example of “The Party Decides” theory, but it was not without underlying tensions. Here, for example, delegates from the party’s left wing hold up signs critical of Clinton at the Democratic National Convention in July 2016: one of several protests against the frontrunner, who was widely seen as representative of the party’s “establishment” during a heavily antiestablishment election.
ROBYN BECK/AFP/Getty Images
The Democratic Party Decides on Clinton
The 2016 contest for the Democratic Party’s nomination was, in many ways, a textbook example of the importance of the invisible primary and the role of party elites in helping to steer the nomination process. Hillary Clinton entered the race as the early front-runner, just as she had in 2008 before she lost to Barack Obama’s insurgent candidacy. Almost as soon as she formally announced her candidacy in April 2015, Clinton had a mountain of endorsements from Democratic Party activists and elected officials. She also raised tremendous sums of money; before 2016 even began, Clinton had already raised over $110 million for her campaign.
Clinton’s overwhelming support among party leaders did not mean that she would easily claim the party’s nomination, however. As she had in 2008, she faced another difficult challenge from an insurgent candidate—this time Vermont Senator Bernie Sanders. Sanders slowly gained ground on Clinton in the national polls during 2015, based largely on his appeal as the antiestablishment candidate. Indeed, Sanders was not actually a member of the Democratic Party at all. In Vermont, he ran for Senate as an independent; although once in the Senate, he generally voted along with the Democrats in Congress. Sanders’ liberal populist platform helped him to attract support outside the mainstream of the party and especially among younger Democrats. His viability as a challenger to Clinton was established by his ability to raise over $70 million during 2015 and by his success in the national polls, in which he received increasingly more support throughout the year leading up to the first contests in Iowa and New Hampshire. On the eve of the Iowa caucuses, Sanders still trailed Clinton by more than