Originally published at Project-Syndicate | January 7, 2022
Notwithstanding the lasting shock of the January 6, 2021, attack of the US Capitol, the Democratic Party can take comfort in the broader demographic trends. Not only was the 2020 presidential election an administrative triumph; record-high turnout showed that the real problem has always been barriers to voting.
AUSTIN – With the anniversary of the January 6 riot now over, let’s focus on the big picture.
The great anomaly of the 2020 US presidential election was that Joe Biden won the national popular vote by seven million votes, yet came within 43,000 (in three close states) of losing the Electoral College, and thus the election. In California alone, Biden had five million more votes than he needed, and in New York, another two million.
So far this century, only Barack Obama has won decisive victories in both the popular vote and the Electoral College. In 2000 and 2016, the popular-vote winner lost the election. In 2004, the result turned on a single state: Ohio. This anomaly is not only persistent but constitutional, which makes it practically unsolvable.
Nevertheless, the 2020 election was a triumph for democracy. Turnout, as a proportion of eligible voters, was higher than in any election since 1900 (when the franchise was limited to males, almost all white). The COVID-19 pandemic forced local election administrators to innovate, and they did so with expanded voting by mail, early-voting days, 24-hour voting, and drive-in voting. More than 100 million ballots were cast before Election Day. In the end, Donald Trump’s final count was 11 million higher than it was in 2016, and Biden’s exceeded Hillary Clinton’s 2016 total by 15 million.
Low turnout in America is usually blamed on voter apathy, but 2020 proved that the real problem has always been barriers to voting. In previous elections, polling places were scarce, the ballots long and complex, and the whole process a slow one, with queues often stretching for hours. Many people lack the time, the patience, or the physical stamina to wait.
The system also discouraged any change in voting patterns, because local election boards allocated machines and poll workers according to past turnout. So there were never enough machines for new voters whenever turnout surged, anywhere at all, for any reason. The 2020 election was thus a great unintended experiment in blowing up the barriers to voting – and it worked.
Those now crying fraud cite the vast increase in turnout as evidence. In fact, the growth in turnout in so-called swing states was no greater than in states where the outcome was not in question. One exception was Arizona, where turnout grew by 30%. But once you adjust for Arizona’s rapid population growth, the proportionate increase is similar to California, where turnout fraud would have been pointless. In any event, the Arizona vote was administered by Republican officials.
Nor do the vote counts look suspicious. Votes are recorded and reported by county, and not merely at the state level. Any tampering with vote counts would have had to happen in specific counties. And because the 2020 election had a close precedent in 2016, strange changes in county voting patterns should be easy to spot.
An analysis of the county-by-county results by me and three colleagues compared the five swing states (Georgia, Arizona, Wisconsin, Pennsylvania, and Michigan) to five states that were foregone conclusions – California, New York, New Jersey, Ohio, and Texas. We did notice a few oddities. Along the Mexican border in Texas, for example, there was a sharp swing in outcomes toward Trump, clearly due to the prosperity brought by federal spending on the border. But these few counties are extremely small. Elsewhere in Texas, two large counties showed strong swings toward Biden, and the same was true of two large counties in Georgia. Those outcomes can be traced to voter mobilization and demographic change. Otherwise, the analysis shows that swing-state and non-swing-state shifts, in either direction, are not distinguishable in the data.
Why did Biden win? The simple answer can be found in the polling data. Compared to 2016, Trump did better with women, Blacks, and Hispanics, but he lost ground with white men, who shifted about five percentage points toward Biden. This shift was driven mainly by men who had voted for Obama in 2008 and 2012 but chose Trump over Clinton in 2016. Their return made the difference in three close states – Michigan, Pennsylvania, and Wisconsin – that had been decisive in 2016. Apart from being close, the swing states weren’t special; the overall shift to Biden was a bit larger in other states, including California, Texas, and New Jersey.
There is a great irony in how US presidential elections now play out. The states with the greatest growth in income inequality since the early 1990s – including California, New York, Connecticut, New Jersey, and Massachusetts – invariably vote Democratic. And the states where inequality has grown less largely (though not entirely) vote Republican. This pattern has been clear for decades, and it grows stronger with each presidential election.
What explains it? It is not about attitudes toward inequality – most people don’t know (or care) about inequality levels in their home state (which we computed for our study). Rather, it is that the Democratic Party has become a coalition of two major groups representing the tails of the distribution: high-income urban professionals and low-income minorities. The Republican strongholds are in exurbs, small towns, and the countryside, in the middle of the income scale. Republicans thus dominate where inequality is lower, and Democrats where it is higher. It is a simple, consistent, and compelling pattern.
The implications of this pattern are unfolding across the South and Southwest, where minority populations (especially Hispanics) are growing rapidly, and where cities are gradually coming into a controlling position against the towns and countryside. That is why Arizona and Georgia flipped in 2020, and why Nevada went to the Democrats a few years back.
In Texas, with 38 Electoral College votes – more than Pennsylvania and Michigan together – there has been an inexorable three-point swing toward the Democrats every four years: Obama got 40% in 2012, Clinton got 43% in 2016, and Biden got 46% in 2020.
Republican legislatures, especially in southern and southwestern states, have done the math and are terrified. That is why they have worked to reverse the great ballot-access experiments of 2020. The GOP’s unspoken watchword is: Get American voters back into long lines (without drinking water)! The point is to discourage as many as possible from voting at all.
If Congress now fails to protect voting rights, that strategy may work for a while, especially in the low-turnout mid-term elections this year. And the Democrats may falter for other reasons in 2024. But voter suppression can’t save the Republicans. Voting is a habit, and habits are hard to break. The writing is on the wall.
James K. Galbraith: Professor of Government and Chair in Government/Business Relations at the University of Texas at Austin, is a former staff economist for the House Banking Committee and a former executive director of the Joint Economic Committee of Congress. From 1993-97, he served as chief technical adviser for macroeconomic reform to China’s State Planning Commission. He is the author of Inequality: What Everyone Needs to Know and Welcome to the Poisoned Chalice: The Destruction of Greece and the Future of Europe.
