Associated Press
- Countrywide measures to contain the coronavirus outbreak have caused an economic toll, but economists say traditional economic measures can't accurately measure in real time the extent to which it has plunged.
- Steve Cicala, an economist at the University of Chicago, has released data that reveals a picture of the economy using a more unorthodox measure: electricity usage, which has plunged as thousands of workers find themselves without a job.
- "People may be at home consuming electricity but [when laid off] there's more leisure consumption than work consumption," Cicala explained. He's been studying electricity usage as an economic measure since 2012, when he noticed it dropped in reaction to the 2008 Great Recession.
- Cicala found that electricity consumption fell 8% in the first three weeks of the 2020 coronavirus-driven economic plunge, a faster and deeper drop in electricity usage than the 18 months of the Great Recession.
- It's been hailed as a measure by economists like Justin Wolfers, who used the data as a basis for economic analysis in the New York Times. "We need economic indicators that can tell us what's happening across the whole economy, day by day," Wolfers tweeted.
- You can read more about the new measure in Wolfers' NYT article.
- Visit Business Insider's homepage for more stories.
The US economy is most certainly in freefall as the coronavirus outbreak has sent people home, forced businesses to shutter and prompted waves of layoffs.
But the nation's usual array of monthly and quarterly economic indicators, built over months of research, are at a time-lag disadvantage when trying to measure the full extent of what's happening to the economy in real time.
As a result, a slightly more unorthodox measure has gained ground: electricity usage.
Steve Cicala, an economist at the University of Chicago, first thought of electricity usage back when he was doing economic research in 2012, where he realized that power outage data had provided a quicker picture of US economic activity during the Great Recession than the monthly unemployment data released weeks or months later.
Now, as the unemployment rate ticks up each day at a rate more typically seen measuring year-by-year changes, electricity usage has allowed Cicala to come up with a measure that gives a grim but timely view of just how bad the economic downturn has already become.
"Over the first year and a half of the Great Recession, electricity consumption fell 5.5%," Cicala told Business Insider. "We're at about 8% right now.... keeping in mind that parts of the country haven't yet instituted measure to contain the outbreak." That's a faster and deeper plunge in less than a month of this recession, compared to the first 18 months of decline in the first 18 months of the 2008 Great Recession.
Cicala's method isn't entirely unprecedented. John Fernald, an economist at the Federal Reserve Bank of San Francisco, has previously used electricity usage data as a means to help verify economic data coming out of China - a fairly standard method used by China analysts in the 1990s, and one that has proven to track independent economic activity like China's level of imports very closely.
Out of 8 non-GDP indicators used to track China's economic activity, electricity-usage proved to be the most accurate individual indicator of the data, Fernald said. And although Fernald ultimately prefers using an index of indicators like the New York Fed's weekly economic index to avoid idiosyncrasies swinging the data, he still finds electricity usage data to be one of the best single economic measures - especially in a case like the current economic downturn.
"Here, the world's changing so fast that the existing data like March's unemployment numbers is so stale," Fernald said. "Finding timely indicators is really valuable."
Cicala's measure has also been hailed by economists like Justin Wolfers, who wrote a piece in the New York Times about why electricity usage might be another way of viewing the downturn.
"The economy has never changed this fast before. The unemployment rate is moving as much every two days as it typically moves in a year," Wolfers tweeted. "And guess what? This new indicator works. Recall the days after Lehman collapsed in Sept '08. Electricity use immediately cratered. But it took months -- and many revisions -- for the official stats like GDP or payrolls to pick up the carnage." He added a graph of the electricity usage data.
And the speed at which the data can be released could also allow it to signal when the downturn's over, according to Wolfers.
"It is likely that these electricity data - which can be produced with a lag of about a day and are available for different regions of the country - will also provide the first clear signal of when the current free fall will end," he wrote in the New York Times.
Differences in data compared to 2008
To be sure, as Cicala acknowledges, there's still some noise crowding the measure because the coronavirus outbreak has prompted a wave of companies to shift to remote work. Variable factors like electricity-powered heating or air-conditioning in the workplace versus at home could then cloud the data.
"If people's homes have heating and cooling that uses natural gas or oil, and it's powered by electricity in the workplace, there's some reshuffling, as it looks like a big fall in electricity," Cicala noted. "That's important to think about when we want to transfer the results of what we're seeing over to the real economy. We want to think about the fact that this is a stay-at-home shock rather than a financial crisis."
But the two economic downturns have some underlying similarities that still make using electricity-usage a strong economic measure.
"People may be at home consuming electricity but [when laid off] there's more leisure consumption than work consumption," Cicala said. "At home, they're looking at online ads to apply for new jobs rather than doing their regular work."
And guess what? This new indicator works.
Recall the days after Lehman collapsed in Sept '08. Electricity use immediately cratered. But it took months -- and many revisions -- for the official stats like GDP or payrolls to pick up the carnage. pic.twitter.com/rdT4y2H2Z6
- Justin Wolfers (@JustinWolfers) April 8, 2020