How Inequality, Not Polling, Predicted the 2020 Election

By Karen Petrou

Perhaps nothing is as startling about the 2020 election as the bad calls pollsters made up to the minute votes were counted.  One might have thought all the mistakes that led to similar 2016 gaffes were corrected – pollsters certainly said so – but they weren’t and the reason why is sad, but simple.  The political-science models on which polling is premised are, like monetary-policy models and so much conventional wisdom, predicated on the vibrant U.S. middle class that once was but is no more.  As we showed early on the economic inequality blog, economic inequality breeds not just acute political polarization, but also a strongly right-leaning shift in voter sentiment.  No wonder – American voters denied the iconic promise of modest economic security and inter-generational mobility are angry.  The more they see prosperity enjoyed by only a few and often a progressive few at that, the angrier they get.  Add in COVID, and this is a witch’s brew of economic despair, social anger, political polarization, and national instability.

Why Americans Are So Angry

Much in this blog details the hollowing out of the U.S. middle class, the high cost of ultra-low interest rates, millennial despair, the stunning spurt of U.S. income and wealth inequality since 2016, and how COVID took away what little most families hoped to have.  Racial economic inequality is awful but U.S. economic inequality knows no racial or ethnic bounds. 

Although the Fed’s most recent Survey of Consumer Finances sought to reassure Americans that the Fed’s vaunted “good place” finally arrived after 2016 thanks to ultra-accommodative policy, the survey’s data actually show how much more unequal America became.  It is true that LMI households gained a bit – the Fed takes pride in a 182% wealth bump for this group, but this windfall amounted only to $200, bringing bottom-quartile median net worth up to $310 by 2019.  The top ten percent’s median net worth then was $2.6 million.

There is of course a middle class the Fed thinks did better from 2016 to 2019.  However, the middle class is not only hollowed out as income and wealth inexorably rose upward, but also struggling even to get medical care for themselves and their kids.  A 2020 Cleveland Fed study finds that whatever middle-class gains have come since 2020 are due to two-income families and all of these income gains disappear when health care and housing are taken into account.

Inequality and Political Polarization

The U.S. has just had two devastating financial crises with the economic-inequality impact briefly summarized above.  A study of financial crises from 1780 to 2014 in advanced economies shows just what financial crises do to political consensus.  It finds that “far-right” parties tend to increase their vote shares by thirty percent within five years of a financial crisis, with the study defining far right by factors such as anti-democratic and pro-ethnicity or pro-nationalist ideologies.  Ten years after a crisis, voting totals tend to revert back to country norms, although this is a weak effect across all the times and nations studied yet to show itself in many post-crisis advanced economies.  Political protests and violent riots – 2020 in the U.S.? – also pick up after financial crises, with none of these effects found following business-cycle recessions.

This result is unsurprising given how much political belief has come to color economic perception.  A Federal Reserve Bank of New York study found that survey respondents in Republican-majority U.S. counties said they were economically better off immediately after the 2016 election even though material economic facts did not vary much before or after the November election.  While the U.S. economy was already on the mend at least in aggregate terms, Republicans did not recognize this until their candidate won.  Democrats largely remained disgruntled even though their circumstances also did not change materially immediately before or after 2016.

Given current economic circumstances, one might have thought – Democratic leaders surely did – that voters would have chosen candidates espousing far more fiscal stimulus.  A majority did, but millions also voted for Mr. Trump because they feared that a lockdown would take away whatever economic advantage they think they still have.  Although much remains to be known about the 2020 vote, initial analysis shows that hollowed-out middle class largely voted for Donald Trump.  High- and low-income voters agreed on a progressive agenda, but the middle class – once a strong progressive and stabilizing political force, voted in record numbers instead to reject candidates – even those they supported in 2018 – out of fear that Democrats would defund the police and socialize the economy. 


When Barack Obama was president, he said that, “[A] dangerous and growing inequality and lack of upward mobility has jeopardized middle-class America’s basic bargain — that if you work hard, you have a chance to get ahead.  I believe this is the defining challenge of our time.”  He knew what pollsters missed, but had the luxury of running for president in 2008, when the financial crisis had barely begun and again in 2012 when fiscal stimulus propped up the economy, financial markets were fragile, and U.S. income and wealth inequality had yet to reach the heights that Mr. Trump exploited to considerable advantage in 2016 and – with almost as much success – in 2020 when income and wealth inequality was far worse than it had been just four years before.

It’s too easy to say that COVID saved the Biden campaign.  Even if this proves to be the case, a pandemic is a high price to pay.  By definition, the middle class is at least fifty percent of a nation.  If it’s angry, then politics is polarized.  So far, this has favored populists.  Progressives may think it will come to favor them, but this has yet to happen.  History isn’t on their side.

Leave a comment