Clinton v Bush in 2016 Could Be the Most Political Science-y Presidential Election of All Time

​Originally published in the Mischiefs of Faction blog.

By Jonathan M. Ladd

As Jeb Bush considers whether to run for the 2016 Republican presidential nomination, and Hillary Clinton remains the most likely Democratic nominee, analysts have already begun speculating about what it would mean to have a Bush v.Clinton election in 2016. Why are political dynasties so successful in the United States? Is it further evidence that the United States is really an oligarchy? While these questions are certainly important, as a political scientist, this sparks another thought in me. A Bush v. Clinton 2016 race could be the most political science-y presidential election of all time! If ever there is ever an election where the fundamentals are vastly more important than details of the candidates and campaigns, this will be it!

There is a stereotype that political scientists think presidential election results are entirely determined by national conditions. These conditions lead the parties to perform similarly regardless of whom they nominate or what happens in the campaign. While few political scientist would go that far, there is some truth to this. The consensus in the field is that structural factors matter much more than is usually depicted in political journalism. The variables political science most often puts forth are national economic performance in the election year, how long the current party has held the presidency, whether the nation is involved in a costly foreign war, and the division of party identification in the mass public. Beyond these, it is hard to find anything else that is correlated with election results, yet not caused by them. For instance, issue preferences and assessments of candidates personalities are strongly correlated with vote choice, but largely because they are rationalizations of pre-existing political orientations.

This is not to say that candidate attributes never matter. But we only have good evidence of this in a few instances. One type of circumstance is when the demographics of the candidate is very novel. In 1928, the first Catholic major party nominee Al Smith did better than expected among Catholic voters. And in 2008, some plausibly argue that the first African American nominee Barack Obama performed worse because of his race than the disastrous economic conditions would predict. Even the classic 1960 book, The American Voter, which is usually cited for its argument about the importance of long term party identification, admits that candidates can matter. One of the themes of the book is that in 1952 and 1956, Dwight Eisenhower's tremendous personal popularity based on saving western civilization from the Nazis allowed him to overcome the Democrats' advantage in long term party identification. Campaign resources also can matter when the discrepancy between the campaigns is huge. At one point Larry Bartels estimated that a 2-to-1 advantage in general election spending, which the Republican Party enjoyed from 1964 to 1972, produced a three-percentage-point increase in vote share. But short of a dramatic discrepancy in resources, almost all campaign efforts by the two sides cancel each other out.

If Hillary Clinton is the Democratic nominee and Jeb Bush the Republican nominee, none of these special situations where candidates and campaigns matter will be present. It will be practically the exact opposite. Rather than standing out from their parties' typical nominees, at this point Bushes and Clintons are synonymous with their respective parties. A 2016 Clinton v Bush race would mean that since 1980, 8 of 10 presidential elections had a Bush or Clinton on a major party ticket and 6 of 10 had a Bush or Clinton at the top of a major ticket. This is more prior national campaign exposure linking nominees' names to their parties than in any previous presidential election.

(Historical Side Note: The closest analogue might be 1968, when Nixon had previously been on a national ticket 3 times and Humphrey once. FDR was elected 4 times, and was vice-presidential nominee in 1920, but none of his opponents' names had been on a national ticket before. Of course his fifth cousin Teddy had been president from 1901 to 1909, but representing the opposite party. Going all the way back to a very different media era, Andrew Jackson and John Quincy Adams ran against each other twice in a row in 1824 and 1828. Adams' father was widely known for his role in the revolutionary era. But when John Adams was president 1797-1801, he represented the Federalist Party, which was defunct by the 1820s, when his son ran essentially without a clear party affiliation.)


One of the reasons the structural circumstances of the parties usually matter so much is that those who have weak enough party attachments to be persuadable pay only loose attention to politics. Once a candidate gets a  major party nomination, they are mainly known to these voters as someone who holds said nomination. That is by far the most salient thing about them.They are assumed to hold the policy positions and economic management aptitude of their party.

Yet in this case, it is not a matter of party reputations overwhelming candidate reputations so much as the two already being inseparable. The actions of Clintons and Bushes have helped create the parties' national reputations over the past several decades at the same time as the families' long prominent partisan activities have adhered those parties' reputations to the families. If they are nominated in 2016, they might as well be named Jeb Republican and Hillary Democrat. As a result, structural party circumstances that political scientist love to talk about will loom large and it will be harder than ever for some new campaign appeal to distract voters from that.

The two ways that this could be wrong is if Hillary's gender has an impact, or if one side obtains a huge financial advantage. While possible, I think both are unlikely. The reason I am skeptical that women would support Hillary in greater numbers (or men in smaller numbers) than they would for another Democrat is that women are historically not disproportionately likely to favor so called "women's issues." Women in recent decades have disproportionately supported Democrats, but that is because women have more liberal views than men on social welfare spending, defense policy, crime policy, and racial policy. Women and men have on average the same views on things like women's role in the family and abortion (see here and here). Gender differences in party preference are not the result of gender-specific political appeals. Thus, I expect Hillary to have the same advantage among women (and disadvantage among men) that Democrats typically do, because that is based on the foundational policy commitments of the parties.


With recent Supreme Court decisions loosening campaign finance regulations, it is difficult to speculate about how much spending will favor each side in 2016. It is possible that money from the wealthy and corporations will produce a Republican advantage greater than we've seen in recent years. All I will say is that, if there is any nominee who minimizes the chances of that happening, it is Hillary. The Clinton family's long-developed network of donors and fundraisers, along with their reputation as members of the more pro-business wing of the Party, will likely give her as much money as any Democrat could hope for in 2016.

If Jeb and Hillary are nominated in 2016, you'll be able to ignore the day-to-day blather of the campaign more than ever, knowing that voters will be deciding based on structural assessments of the two parties. For better or worse, those hoping to get a respite next cycle from political scientists feeling superior to campaign journalists may be disappointed.