Covid-19 | Economics | Politics

In 2020, is it still “the economy, Stupid”?

The axiom that James Carville made famous during President Bill Clinton’s successful 1992 run for the White House—that the the U.S. economy would dictate whom voters would ultimately elect—has remained a guiding principle in politics ever since. The coronavirus pandemic has upended all expectations for 2020, but the importance of the economy and healthcare for American voters remains. A week before election day, APM Research Lab has compiled various economic indicators to make some sense of the current moment.

by GABRIEL CORTES | Oct. 27, 2020

Eight years ago, when President Barack Obama was running for reelection, one of the most persistent issues that dogged his campaign was the question of unemployment. The jobless rate among U.S. workers had remained stubbornly high after hitting a 10% peak in October 2010, and while the Great Recession had officially ended in 2009, Obama had not seen an unemployment rate lower than 8% during his first term.

Years after the financial collapse of 2008, the aftershocks from those events continued reverberating across the economy.

And then, in September, weeks before the 2012 election and four years after Lehman Brothers had folded, unemployment finally fell below 8%, and Obama was reelected.

The unemployment rate was then (and continues to be) one of—if not the most—ubiquitous markers by which the American voters judge the economy. Before the pandemic, the U.S. experienced the longest economic expansion in the country’s recorded history. Unemployment continued its downward trajectory from the Obama administration, dropping to 3.5% at the end of 2019: a low not seen since Neil Armstrong had walked on the moon 50 years before.

However, as Covid-19 began spreading across the U.S. at the beginning of 2020 and various state and local governments instituted widespread lockdowns in March and April, the unemployment rate skyrocketed to almost 15%. But as spring turned to summer and businesses began reopening, joblessness fell almost as quickly as it had risen; by the end of September, it was 7.9%, more than double what it had been nine months earlier.

The number of people looking for work has also fallen. In February 2020, the employment-to-population ratio (the number of people working or actively looking for work) was almost 61%, according to the Bureau of Labor Statistics, but in April, it fell almost 10 points. It’s since recovered somewhat, but it’s still lower than 60%, meaning that there are fewer people working or looking for work now than there were at the beginning of the year.

Another important metric in the political zeitgeist is median weekly income. On the surface, this measure would seem to have had as good a 12-year trajectory as the unemployment rate had. At the beginning of 2009, nominal figures published by the Bureau of Labor Statistics show that median weekly income was slightly more than $700, and that in the third quarter of 2020 it was more than $1000.

However, when APM Research Lab adjusted those data for inflation, they revealed a more mixed picture. The graph on the left shows the nominal (unadjusted) data published by the Bureau of Labor Statistics. The graph on the right shows our inflation-adjusted figures and a much more modest increase for U.S. workers: slightly more than 10%.

As with the unemployment rate, the denominator for median earnings has also shrunk, according to the Bureau of Labor Statistics. The number of workers earning wages dropped more than 11% in the first half of 2020, and while it rebounded somewhat in the third quarter, there are still about 7.6 million fewer people earning money now than there were at the end of 2019.


The national unemployment rate and median weekly income are helpful long-term indicators, but since the late spring, the Census Bureau has collected data about U.S. households during the pandemic in its new Pulse Survey. The topics are wide-ranging, but they include a question about people who are behind at least one month on their rent. Since late April, about 15% of Americans have consistently been delinquent on their rent.

An even more interesting trend comes from the second iteration of the Pulse Survey, which the bureau has circulated since August. The updated survey asks Americans how confident they are that they can pay their usual expenditures. On average, more than 50% of Americans said they had some difficulty paying those bills.


The last indicator comes from our ongoing collaboration with Marketplace about the pandemic economy. Since July, we’ve kept tabs on the number of people receiving unemployment benefits across the country, along with a few other indicators. The result is a comprehensive map that includes the expanded unemployment benefits passed in March through the CARES Act, along with a few other indicators.

While first-time unemployment claims by and large receded after reaching their maximum in April, the total number of people receiving benefits is more than four times higher than it was before the arrival of Covid-19.

As of the second week of October, more than 21.5 million people were receiving some kind of unemployment insurance. See the differences across states in the map below, and read the full story at Marketplace.

Note: “Current unemployment claims” is calculated from the Labor Department’s weekly report on Unemployment Insurance (UI) Claims. It is the sum (for the most recent week reported) of: “Advance State Claims—Insured Unemployment” + “Pandemic Unemployment Assistance (PUA)—Continued Claims” + “Pandemic Emergency Unemployment Compensation (PEUC)—Claims Filed”.

The number of “current unemployment claims” may be higher than the number of people actually collecting benefits in a given week. In some cases, a claim may be reported as “insured” or “continued” before the claim has been approved for payment. There have been inconsistencies and delays in state data collection and reporting. Claims for regular state UI are reported one week ahead of claims for federal PUA and PEUC.

The average unemployment benefit is from August 2020 (the most recent data available). Apartment List data are the three-month average for the second quarter of 2020.

Thoughts? Questions?

We want to hear from you! Leave us a message below.

…or email us at info@apmresearchlab.org