Polls, pundits and even bookmakers have already elected President Hillary Clinton. However, at this point in the Game of Oval Office Thrones, models are better oracles than polls – and the models do not like her odds.
Hillary a Slam Dunk, Right?
We are still twenty months out from the US presidential election and ten months from the first primary, but the electoral season is in full swing. Already many pundits are speculating about who will and will not take the White House in November of 2016. At the moment, Hillary Clinton is the consensus favorite. Most cite her polling numbers, popular appeal to women, and formidable war chest.
This ‘Clinton-think’, though, has weakened over the last few months as the media has focused its attention on her from “email-gate” to the questionable financials of the Clinton foundation to other varied issues. Still, few believe that any on the Democratic side will challenge her for the party nomination.
Alternatively, the Republican field is quite broad and especially interesting to media and political analysts. Most ask the simple question: which candidate will be able to survive the primary wood chipper and best appeal to the average, or median voter, in the general election? Which one has the political chops to see it through?
Ultimately, many believe that a Clinton victory is inevitable given her strong lead in the polls. However, does such heightened attention around individual candidate characteristics and political variables, like funding, really matter? Are these the things we should be looking at this far out?
Beware of False Positives and Misleading Polls
In our experience, the answer is no; we shouldn’t. Horse race polls this far out often produce ‘false positives’. Indeed, while polls are good predictors of electoral outcomes a few weeks from election day, they are extremely error prone at this stage of the game (see table below).
This, however, still doesn’t stop pundits and politicos alike from using the horse race polls. Bookmaking site Paddy Power has pretty much crowned her as the winner. Such prognostications often confuse more than help. So what should we use to assess relative electoral odds at this stage of the game?
Fundamentals Over Polls
We like to look at what we call ‘political fundamentals’—factors like the economy, incumbency, and approval ratings. This is nothing new. There is a large academic and empirical tradition in employing such models to good effect.
Such variables are structural in nature and do a good job of defining the conditions of a given election. Simply put, they help us answer the simple question: do the political fundamentals favor the government or opposition candidate?
To assess the 2016 US political fundamentals, we look at two simple variables: (1) whether the incumbent is running and (2) the government’s approval rating. I have aggregated more than 450 elections from 35 countries into one database.
As an aside, we aggregate across election types and countries because it gives us a large sample size to work with and such numbers tend to wash out the idiosyncrasies. If we were to focus only on US elections, we would only have a sample size of 16!
So what do we find?
Incumbency is King!
The party in power running an incumbent has an almost three-fold (2.64) greater chance of winning than when they run a successor! Think Obama (2012) or Bush (2004) as examples of incumbents of the party in power, versus McCain (2008) and Gore (2000) as successors to the party in power.
The incumbency advantage becomes especially distinct when taking into consideration the approval rating of the government, or party in power. Specifically, according to our model, an incumbent has a 78% chance of winning when the government’s approval rating is at 45 points, while a successor only has a 14% chance at the same approval level!
So what does our exercise mean for 2016?
2016 Looks like a Republican Year
First and foremost, 2016 looks to be a Republican year. The Dems, of course, are not running an incumbent—Obama’s two terms are up! This is a rough starting place for the Dems.
Additionally, the average government approval rating at the time of this article is at or around 45 points. Such levels put the Democrats’ chances of winning at a lowly 14%!
Now, given the likelihood of an improved economy, approval ratings may tick up. But they would have to pass the 55 point mark to give the Democrats even odds of keeping the White House. This, in my mind, is extremely unlikely, given the equally important trend of approval ratings declining over time.
However, I always take such models with a grain of salt. At this stage, we use models more as a general marker rather than a definitive statement about an electoral outcome. Indeed, maybe the Dems aren’t as bad as 14%; other factors probably will play a role which might help their odds, such as Clinton’s name recognition, funding, “get out the vote efforts”, etc.
Dems and Clinton Only at 30% Chance of Winning the Presidency
Even so, Dems still look to have a significantly lower than 50/50 chance of taking the White House. I would say 30/70.
Some will argue that Clinton is special; that her chances are significantly better than 30% because – given her popularity and status as a ‘legacy’ candidate – she seems more like an incumbent. Such cases are extremely rare. I prefer to rely on averages.
That said, for the sake of argument, let’s say she is more like an incumbent (somewhere between an incumbent and a successor). Let’s run that hypothesis through our model. At an approval rating of 45 points, Clinton’s chances still are less than 50/50 at 46% (or 78%+14%/2).
Now, of course, the fundamentals can always change, and with them the odds for the Dems. But don’t be surprised when a Republican takes the oath of office in January 2017. I won’t be!