Ever since the Corona-Virus pandemic hit America, much of the average American’s life has changed. In most states you can’t go the gym, dine in, or go to work (hence 22 million unemployment claims since the end of March). In addition, for the first time in quite a while, the President’s numbers have changed. First, his approval numbers bumped up. Then, they fell back down to Earth. But, as this was happening, a particular occurrence was being seen in election surveys – for the first time ever the President’s election numbers were under-polling his approval numbers. In the past it had usually been the opposite.
So, what should be made of this? More importantly for the Trump camp and its supporters, should they be worried? The easy answer is maybe. The longer answer is yes.
Undeniably, a lot of opinions of the President are tied to his response to the COVID – 19 pandemic and the economy tanking due to government fiat. Fortunately, for the President, he remains competitive in swing states and most polling outside of partisan pollsters, such as PPP (D), show him leading or slightly trailing in key battlegrounds. It is worth repeating we don’t have a single national election but 50 individual elections for President. In this case, the polling shows Trump is in trouble but competitive.
It is primarily in the national polls we have seen the President’s approval ratings be higher than his head to head numbers against former Vice President Joe Biden. Per the latest numbers from Realclearpolitics, the President’s approval rating sits at 46 percent but he only polls around 42 percent against Joe Biden. In poll after poll, we have seen this phenomenon play out. In a recent Economist survey, the President’s approval rating sat at 45 percent but he only polled 43 percent support against Biden. A Harvard-Harris survey found the President’s approval rating sitting at 49 percent but he lost to Biden 47 percent to 53 percent. Most recently, a WSJ survey found the President at 46 percent approval but losing to Biden 49 percent to 42 percent.
These numbers are eye-catching and suggest a couple factors could be at play, especially since digging into the cross-tabs of the surveys we don’t see a lot of reason why there is this discrepancy. Yes, in each survey, people worry about the pandemic and trust the former VP over Trump but they also trust Trump over Biden to fix the economy. Minor fluctuations could be at play in the numbers but a whopping four percent difference in the WSJ survey is outside the margin of error and suggests something else is going on (especially since state polls are not reflecting this dynamic).
Based on past electoral experience and common sense, one can come to three possible conclusions. The first is a reflection of Kentucky’s 2015 gubernatorial election, another is simply the polls are wrong and the third is the “Shy Trump Voter Effect.”
Considering what is happening now the 2015 Kentucky gubernatorial election is not to say Kentucky is a microcosm of the nation. Rather, the circumstances largely match up. In the 2015 gubernatorial election, you had an unpopular then Republican candidate in Matt Bevin running for Governor against an entrenched and well known Democratic Attorney General, Jack Conway. Most of the polling of the contest showed Conway leading (until close to the end) but a large number of voters remaining undecided. Yet, when Election Day came Bevin vastly outperformed his polling average by around seven points. Now, to be fair, Bevin was not an incumbent, unlike Trump, but Bevin had held local elected office and had name ID (he ran against Mitch McConnell in the 2014 GOP Senate primary). It is possible many undecided voters are leaning towards Trump but have ultimately not decided to support the President. However, if true, they will likely come home in the end after three debates and the President spending his now over $200 million on the election.
The second is simply the polls are wrong. Their samples are either wrong, they are not reaching Trump voters or something else. We saw most national polls miss the mark in 2016 by an average of two to three points. This didn’t mean Trump won the national vote but it means he came closer than many individual polls showed. What the biggest misses were in 2016 was the state polls. Many pollsters mirrored the national numbers and if the same is true this year it means Trump is in much better shape than any of us currently suspect. While not Presidential numbers, it should be noted a number of pollsters had egg on their faces after the 2018 election in Iowa, Ohio, and Florida, in particular, after they showed Democratic wins, only to see Republicans emerge triumphant in a disaster of a national environment.
The final possibility is the emergence of the “Shy Trump Voter Effect.” The theory goes something like this. Pollsters reach out to voters, and because peer pressure, judgement, media bias or some other reason, these voters either say they are undecided or will vote for somebody other than Trump when in reality they secretly support Trump. Initially, a conservative conspiracy theory during the 2016 campaign, the theory was given more credence after Trump’s victory and subsequent statistical studies. If this is true, whether by itself or some combination of possibility two where pollsters don’t correct for their own bias, it could explain why more voters approve of the President than say they will vote for him.
Even if any of these possibilities are true, it is safe to say it is better to be ahead than behind and right now polls show Trump continuing to trail Biden. The same polls show Trump in striking difference and pollsters do admit they do not know what turnout will look like and Trump supporters are far more motivated than Biden voters (though a vote is a vote, enthusiastic or not). Still, right now, with the pandemic raging, a perceived shortage of medical equipment, tests and supplies, and Governors blaming the Fed for everything the President is taking a hit. Likely, if the economy quickly comes back, proof emerges the virus emerged in China and they lied, and the worst death rates of the virus do not occur the President will benefit. If not, well, undecided voters will not be Trump voters like they were in 2016. Turning out one’s base only goes so far.