Michael Spence and David W. Brady
MILAN/PALO ALTO – In the US presidential election this year, President-elect Joe Biden received 79.8 million (51%) votes, Donald Trump received 73.8 million (47.2%), and the remaining candidates received 2.5 million (1.7%). Though votes are still being counted in California, New York, and Illinois, this year’s turnout has reached a record high for the post-war period.
Yet, owing to America’s Electoral College system, which allocates more electors per person to states with smaller populations, the outcome of this election was much closer than the popular vote would suggest. Though Biden ultimately won a significant number of battleground states, he did so by very thin margins, effectively reversing the outcomes that gave Trump his victory in 2016. In Pennsylvania, Michigan, and Wisconsin, just enough moderate Republicans, Democrats, and independents shifted toward Biden to make the difference.
A county-level view of voting patterns in 2016 and 2020 shows a deep and consistent partisan divide. The axes represent the Democratic vote minus the Republican vote, expressed as a percentage of the total vote in 2016 and 2020. A dot on a negative number signifies a Republican victory in that county. Overall, the points are highly concentrated around the 45º line, along which the margin of winning or losing was the same in both election cycles.
Generally, Democrats gained in counties that they won or only narrowly lost in 2016. Similarly, in counties that Republicans won by a substantial margin in 2016 (of which there are many), the data tend to show an even larger margin of victory for Republicans in 2020.
The crucial area is in the blue circle, where the margin of victory or loss in 2016 was quite small. It is here that Democrats made gains and, in some cases, won over just enough voters to flip the county in their favor.
The red outline comprises mostly Democratic setbacks. In these counties, Democrats lost outright in 2020, or won by a substantially smaller margin than in 2016, thereby reducing their total at the state level. These outliers are mostly Hispanic-majority counties in South Florida – specifically Miami-Dade – and in Texas along the border with Mexico. These negative swings for Democrats were significant factors in Trump’s victories in those two states.
In the battleground states, both Biden and Trump benefited from higher overall turnout, but Biden clearly benefited more. Relative to pre-election polls, including YouGov polls that we analyzed, neither Biden’s narrow win in these states nor the huge overall turnout was a surprise. By contrast, the polls did not fully anticipate the strong, materially relevant support for Trump in certain segments of the Hispanic community. Moreover, many commentators assumed that a huge electoral turnout would favor the Democrats, leading some to speak of a potential “blue wave.” In the event, there was a very large turnout for both sides.
In terms of absolute numbers nationwide, many more people identify as Democrats than as Republicans, but a significant share of voters identify as independents.
In battleground states, however, it is Republicans who have the advantage (with independents remaining a substantial group).
As in 2016, the 2020 election was decided by razor-thin margins in battleground states. But the bigger story this year was the massive turnout on both sides. Does this mean that both sides had unusually attractive candidates? Not exactly. While 80% of Trump voters told pollsters they were specifically voting for the candidate himself, 56% of Biden voters described their decision as a vote against Trump.
A better explanation, then, is that voters turned out in record numbers on both sides because they emphatically did not want the opposing party to win. Moreover, this “negative partisanship” is symmetric. Democrats and a few moderate Republicans and independents did not want another four years of Trump, and most Republican voters appear not to have wanted to pass the baton back to the mistrusted “coastal” elites, globalists, “biased” liberal media, and Washington insiders.
With two runoff elections still to be held in Georgia, the final Senate result is not yet decided. Democrats would need to win both races in order to control the chamber (with 50 seats and Vice President-elect Kamala Harris providing the tie-breaking vote). As matters stand, this seems unlikely.
The Senate shares some of the characteristics of the Electoral College in a more extreme form. The number of senators (two for each state) per eligible voter is roughly inversely proportional to the population of the state. This is by design, to ensure that the interests of less populous states are effectively represented in Washington, DC. Because the nine states with populations greater than ten million account for 51% of the US population, these states would effectively control the legislative agenda if Senate seats were allotted in proportion to population.
In the face of a deteriorating public-health situation and significant economic challenges, the wise course would be for both parties to pursue some of the existing opportunities for bipartisan initiatives, such as those recently outlined by former World Bank President Robert B. Zoellick.
But the wise course is not necessarily the one that will be followed. The most glaring area of bipartisan agreement nowadays is the American public’s broad lack of trust and confidence in institutions, which has led to persistent policy paralysis.
The US has just held an election in which candidates from both major parties won seats in Congress. It is up to elected officials on both sides of the partisan divide to ensure that the country will win, too. And it is up to voters to hold them accountable if they fail.
Michael Spence, a Nobel laureate in economics, is Professor of Economics at New York University’s Stern School of Business and Senior Fellow at the Hoover Institution. David W. Brady is Professor of Political Science and Leadership Values at Stanford University and Senior Fellow at the Hoover Institution.
Copyright: Project Syndicate, 2020.