KEY POINTS FROM THIS ARTICLE
— Two factors that historically have been helpful in forecasting presidential elections — the power of incumbency and the state of the economy — may be less useful in this year’s election, both because of long-term changes in American politics and the current public health crisis.
— A forecasting model based on evaluations of the incumbent president may be a better fit for this election.
— This incumbent accountability model makes Donald Trump an underdog, but gives him about a 30% chance of winning based on his current approval rating.
— However, he needs to improve his approval rating significantly to better position himself for a second term, based on history.
A different kind of election
The 2020 presidential election presents forecasters with unique challenges. First and foremost, the election is taking place in the midst of one of the most severe crises that the United States has faced in the past hundred years — the coronavirus pandemic. Not only has the pandemic already killed over 150,000 Americans, with thousands more likely to die before Election Day, but it has also produced the most severe economic downturn since the Great Depression with the unemployment rate reaching double-digits and annualized GDP growth declining by a staggering 33% in the second quarter of 2020.
President Trump was already facing a stiff battle for reelection before the coronavirus pandemic hit the country — consistently trailing the Democratic frontrunner, Joe Biden, in almost every national poll and most swing state polls in January and February. Over the past few months, however, Trump’s situation has become increasingly dire with his approval rating falling from the mid to the low forties and his deficit against Biden in national polls rising from an average of 5-6 points to an average of 8-10 points.
My “time for change” election forecasting model relies on three predictors to forecast the outcomes of presidential elections: The incumbent president’s approval rating in late June or early July, the change in real GDP in the second quarter of the election year, and a dummy variable based on whether a first-term incumbent is running for reelection. This time for change factor reflects the fact that first-term incumbents like President Trump generally enjoy a significant advantage even after controlling for their approval ratings and economic conditions.
There are good reasons to expect that two of these predictors — the change in real GDP in the second quarter and the time for change dummy variable — will not perform as they normally do in 2020. Although the U.S. economy is currently experiencing a severe downturn, with real GDP plunging by an almost unprecedented amount in the second quarter, voters do not appear to hold the incumbent president responsible for this. That is undoubtedly because the recession was deliberately induced in order to try to control the spread of the deadly coronavirus. Thus, despite a massive rise in unemployment and decline in real GDP, President Trump’s approval ratings on handling the economy have generally remained positive.
It is also the case that over the past 10-20 years, assessments of economic conditions have become increasingly divided along party lines, with supporters of the president’s party consistently rating the state of the economy much more positively than supporters of the opposition party. Thus, immediately after Donald Trump took office, Republicans’ assessments of economic conditions became much more positive while Democrats’ assessments of economic conditions became much more negative. This growing partisan divide in how voters view economic conditions means that economic trends may have less impact on election results than in the past.
Increasing partisan polarization may also have the effect of reducing the electoral advantage of being a first-term incumbent. That advantage was based mainly on the ability of such incumbents to appeal to voters across party lines who might be reluctant to replace a president after only one term in office. However, rising partisan polarization means that voters today are much less willing to support a candidate from the opposing party for any reason. Thus, Gary Jacobson has shown that the advantage of incumbency in U.S. House elections has declined sharply in recent elections. I would expect this to be true in presidential elections as well, especially in the case of an extremely divisive incumbent like Donald Trump who appears to have no interest in appealing to voters beyond his own party’s base.
A simplified incumbent accountability model
Given the concerns described above, for the 2020 presidential election, I am using one predictor to forecast the results: the incumbent president’s net approval rating in late June. In addition, I have made two other modifications of my usual presidential forecasting model: using the electoral vote rather than the popular vote as the dependent variable and limiting the analysis to contests with a running incumbent. Along with the current forecast, I will present conditional forecasts based on the president’s net approval rating in late October.
The reason for using the electoral vote rather than the popular vote as the dependent variable is simply that the electoral vote is what determines the winner of the election, and two of the last five presidential elections have produced Electoral College misfires — outcomes in which the winner of the popular vote lost the electoral vote. Despite this fact, there is a very strong relationship between the national popular vote and the electoral vote. Across all 18 presidential elections since World War II, the correlation between popular vote and the electoral vote is .97. The magnitude of the 2016 misfire was quite extraordinary.
Table 1: Correlations of net presidential approval with incumbent party electoral vote
Note: No late October approval rating available for 1948 election.
Sources: Gallup Poll for 1948-2012; FiveThirtyEight average for 2016.
For this forecast, I am only using elections like 2020 with a running incumbent. The reason for doing this is that the connection between the incumbent’s job approval rating and the election outcome is much closer in these contests than in those without a running incumbent. As the data displayed in Table 1 show, this is true for both the late June approval rating and the late October approval rating. When an incumbent is running for a second term, the election is largely a referendum on that incumbent’s performance during the previous four years. When there is an open seat election, the public’s appraisal of the incumbent’s performance matters, but not nearly as much.
I have argued that the extraordinary circumstances under which the 2020 presidential election is taking place along with deepening partisan polarization mean that objective economic conditions, while terrible by any standard, are unlikely to have much effect on the outcome of the election. That is because it is the coronavirus pandemic that seems to be driving voters’ assessments of President Trump’s performance. Thus, while assessments of Trump’s handling of the economy have continued to be more positive than negative, assessments of his handling of the pandemic as well as assessments of his overall job performance have turned decisively negative since April.
Table 2: Regression analysis of Trump job approval on Trump pandemic approval and Trump economy approval
Source: RealClearPolitics data on 40 national polls completed from April 1-July 15, 2020.
Some evidence for the central role of the pandemic in shaping assessments of President Trump’s overall performance can be seen in Table 2. This table displays the results of a regression analysis of the president’s overall job approval rating on ratings of his handling of the pandemic and ratings of his handling of the economy in 40 national polls conducted between the beginning of April and the middle of July.
The results displayed in Table 2 show that assessments of Trump’s overall performance were highly dependent on assessments of his handling of the coronavirus pandemic. Thus, a one-point rise or fall in net pandemic approval was associated with a .72-point rise or fall in net job approval. In contrast, assessments of Trump’s handling of the economy had no effect at all on his overall approval rating. These findings suggest that Trump’s almost single-minded focus on the economy as the key to his chances of reelection was highly questionable. Voters have been far more concerned with controlling the pandemic, and they have become increasingly dissatisfied with his performance in that area.
Forecasting Trump’s electoral vote
Table 3 displays the results of a regression analysis of incumbent electoral votes on late June approval rating as well as a forecast of the number of electoral votes that Donald Trump will receive in the 2020 presidential election based on an approval rating of -15 percentage points at the end of June. The table also includes the estimated probability of a Trump victory based on this forecast and the standard error of estimate of the regression equation.
Table 3: Regression analysis of incumbent’s electoral vote with late June net approval rating, 1948-2012
Source: Electoral votes from uselectionatlas.org; presidential approval from Gallup Poll; Trump late June approval from FiveThirtyEight average.
The results displayed in Table 3 show that for every one-point increase or decrease in net approval in late June, an incumbent president can expect to gain or lose an additional 4.7 electoral votes. The estimated intercept of the equation indicates that an incumbent with a net approval rating of zero — meaning equal approval and disapproval — would have a decent chance of winning a second term with a predicted 290 electoral votes. That is partly because there is still considerable time between late June and Election Day and, as we shall see, incumbents can sometimes increase their approval rating over these four months.
With a late June net approval rating of -15, this model predicts that Donald Trump will lose the electoral vote to Joe Biden by a decisive margin of 319 to 219. However, because the model has an adjusted R2 of only .64 and has a standard error of estimate of about 92 electoral votes, the model gives Trump about a 30% chance of winning the election. In other words, at this point, according to this simple incumbent accountability model, Joe Biden is a solid but not overwhelming favorite to win the 2020 presidential election.
Table 4: Regression analysis of incumbent candidate electoral vote share with late October net approval rating, 1948-2012
Source: Electoral vote data from uselectionatlas.org; presidential approval from Gallup Poll.
We can also estimate a model based on the incumbent president’s net approval rating in late October. Table 4 displays the estimates for this model. Not surprisingly, the model is considerably more accurate than the one based on late June approval ratings. It has an adjusted R2 of .82 and a standard error of estimate of only about 69 electoral votes. Another important difference is that an incumbent with a net approval rating of zero is now given less than a 50% chance of winning reelection. Such an incumbent would be predicted to win only 262 electoral votes, slightly below the 270 required for victory. According to this equation, a gain or loss of one percentage point in net approval is predicted to produce an increase or decrease of about 5.6 electoral votes.
We can use the estimates in Table 4 to produce conditional forecasts of incumbents’ electoral votes based on their net approval rating in late October. Those predictions, along with the estimated probabilities of victories for the incumbents, are displayed in Table 5. In this table, I present conditional forecasts based on net approval ratings ranging from -20 to +5. This range would seem to encompass the plausible upper and lower limits of Donald Trump’s net approval rating in late October because his actual net approval has very rarely fallen below -20 or risen above zero during his presidency.
Table 5: Conditional forecasts of Trump electoral vote and probability of victory based on late October net approval rating
Source: Predictions based on regression equation in Table 4.
The conditional forecasts displayed in Table 5 indicate that if Donald Trump’s approval rating remains unchanged in late October from where it was in late June, there is a strong likelihood that he would be defeated in an Electoral College landslide. Under this scenario, Joe Biden would be expected to receive 361 electoral votes to only 177 for President Trump, and the president would have only a 9% chance of winning.
Based on these results, in order to have a reasonable chance of winning the election, the president would have to raise his net approval rating to much closer to neutral. A net approval rating of zero would result in a prediction of 262 electoral votes for Trump to 276 for Biden and would give the president a 46% chance of winning.
Table 6: Change in net presidential approval for running incumbents between late June and late October, 1956-2012
Note: No late October approval rating is available for the 1948 election.
Source: Gallup Poll
What are the chances, then, of President Trump substantially raising his net approval rating between late June and late October? Table 6 displays the June and October net approval ratings for the 10 most recent incumbent presidents seeking reelection. According to these data, six of these 10 presidents experienced an improvement in their net approval rating over these four months. The average increase in net approval for these six presidents was just under seven points. However, only one incumbent, Richard Nixon in 1972, saw his net approval rating improve by a double-digit margin. Moreover, rising approval ratings have not been universal. Three incumbents saw their net approval rating decline between late June and late October.
President Trump’s approval ratings have been remarkably stable throughout his nearly four years in office. According to the FiveThirtyEight average, the only time his net approval rating reached -5 or better in the past three and a half years was for a brief period in late March when he received a small boost in popularity during the early days of the coronavirus pandemic in the United States. Based on this history, it seems unlikely that the president will be able to raise his approval rating substantially by late October.
In 2016, Donald Trump won a majority of votes in the Electoral College while losing the national poplar vote by more than two percentage points. The huge discrepancy between this popular and electoral vote margins was due to his very narrow victories in several swing states including Michigan, Wisconsin, Pennsylvania, and Florida. However, based on the results presented in this article, Trump’s chances of repeating this feat in 2020 appear to be slim.
When an incumbent president is running for a second term, the election is always largely a referendum on the president’s record during his first term. Normally, an important component of that record is the performance of the U.S. economy, especially during the first half of the election year. In 2020, however, due to the devastating impact of the coronavirus pandemic on American society and on the economy, it appears likely that the presidential election will turn much more on the president’s handling of that pandemic. By the summer of 2020, the public’s assessment of the president’s handling of the pandemic had turned decidedly negative.
It also appears unlikely that President Trump will enjoy the electoral advantage that normally accrues to first-term incumbents. Partisan polarization has drastically reduced the ability of incumbent office-holders at all levels to appeal to voters across party lines. Moreover, unlike previous incumbents, Trump has made little effort to expand his base of support during his time in office.
Based on these considerations, I have presented a simple incumbent referendum model for forecasting the outcome of the 2020 electoral vote. The president’s net approval rating of -15 percent in late June yields a forecast of a decisive 319-219 vote victory by Joe Biden in the Electoral College. Yet the model still gives Trump about a 30% chance of winning the election due to uncertainty about what will transpire between June and November. However, if Trump’s approval rating remains at -15 in late October, the model predicts an even more overwhelming defeat with 361 electoral votes for Joe Biden to only 177 for the president. At that point, Trump would have only a 9% chance of winning the election according to the model.
Alan I. Abramowitz is the Alben W. Barkley Professor of Political Science at Emory University and a senior columnist with Sabato’s Crystal Ball. His latest book, The Great Alignment: Race, Party Transformation, and the Rise of Donald Trump, was released in 2018 by Yale University Press.