Major newspapers in the U.S. largely ignore economic signals most relevant to the welfare of lower- and middle-income households, according to new research, a dynamic that raises fundamental concerns about whether citizens are getting the information they need to accurately gauge how the economy is working for them.
In an article published April 12 in American Political Science Review, researchers found evidence that economics reporting “strongly and disproportionately” tracks the fortunes of the country’s richest households, likely because newspapers focus their coverage disproportionately on aggregate measures of economic performance, using data that, while plentiful, doesn’t track the financial fortunes of the vast majority of Americans.
Economics coverage with a positive tone was correlated with growth in the incomes of the top 20% of earners, the researchers found, but wasn’t linked to growth for earners in lower quintiles. The result of that skewed focus could be an electorate that can’t effectively assess what economic performance means for them or how implemented policies aligned with their interests.
“To the extent that voters’ perceptions are shaped by the media . the ‘economy’ on which most voters have been voting has, in an important sense, not been theirs,” the researchers wrote.
Underlying dynamics have been shifting in recent decades, transforming the relationship between what’s good for “the economy” as a whole and actual income gains for the average Americans. While aggregate indicators such as aggregate growth and employment tracked the incomes of the non-rich populace relatively closely through around 1980, research has shown that subsequent economic expansions distributed gains and losses in increasingly unequal and counterintuitive ways.
Since 1985, for example, a 1 percentage point drop in gross domestic product growth has been correlated with a 4.55% drop in the incomes of the top 0.1% but a 1.08% increase in the income of the median earner, the researchers wrote, citing the findings ofa 2014 study. Similar findings hold for unemployment, they added: A 1 percentage point rise in the male unemployment rate is associated with an average income loss of 6.87% for the top 0.1% but only a 1.77% loss at the median.
The richest households have also gained far more during expansions than the rest of the country, the researchers noted, further evidence that, “It is the very rich whose fortunes rise fastest and fall most steeply with the business cycle.” These trends may represent a few of the pathways driving rising income inequality in the U.S., which saw the highest-earning 20% of households rake in 52% of total household income in 2018 — up from 43% in 1968, according to Pew Research.
The phenomena may also have structural implications for the way economic trends are picked up in the news media, according to corresponding author Alan Jacobs, professor at The University of British Columbia. That’s both because journalists have easier access to aggregate economic data and because most of them, like much of the populace, assume that what’s good for the economy is probably good for everyone participating in it.
“The idea that a rising tide lifts all boats, or that generally we’re all doing better when the economy is growing or all doing worse when the economy shrinks, is a prevailing notion among journalists,” Jacobs told The Academic Times, citing sociological research on news orgamzatlons. That idea encourages reporters to highlight aggregate statistics, an instinct Jacobs said is understandable but tends to leave out crucial information about who’s actually receiving gains and losses.
“It’s not that journalists aren’t aware that there are winners and losers in the economy, it’s not that journalists aren’t aware that there are distributional issues about who’s getting what,” he said. “But those are much more complicated to deal with. . It’s much more time- and labor- and information-intensive to try to figure out. “
The researchers set out to determine whether the tone of economic news reporting tracked with periods of broad economic gains or losses, or whether the tone of the news moved in periods where those gains or losses were concentrated in the hands of the highest income earners.
To do so, they adopted a measure of economic news tone which used a dictionary with 6,016 English words coded for their positive or negative connotations. A computer program counted the numbers of positive and negative words in each article, ultimately yielding a tone score that adjusted for the amount of neutral content.
The program examined a sample of nearly 2.5 million articles from 32 high-circulation U.S. newspapers, performing a time series analysis of their economics coverage. Articles about foreign economies and articles featured in the business section — and therefore likelier to focus on stock market news — were excluded from the sample.
The researchers regressed tone on the growth rates of each U.S. income quintile, finding that across all modeling specifications, a positive news tone was linked with income growth among the top 20% of earners. “It is the only quintile for which this — or anything even close to it — is true,” they wrote.
Depending on the specific regression model they studied, a one standard-deviation difference in income growth for the top quintile — approximately 0.03% — was associated with 14% to 21% ofa standard-deviation difference in news tone. By comparison, they noted, an economic recession is on average linked with a drop in news tone of 43% ofa standard deviation.
Most of the link between news tone and top-quintile income growth is actually driven by income growth within the top decile of earners, according to further analysis. A change of one standard deviation in income growth rates among the top 1% of earners is correlated with a change in news tone equivalent to 11% of one standard deviation. No evidence suggested a consistent association between news tone and income growth for low- or middle- income earners, they discovered.
Overall, the researchers wrote, the findings suggest that newspapers are painting a picture of the economy that over-represents the experiences of the richest households “to a stunning degree.”
That bias seems to stem from the news media’s tendency to cover the business cycle, according to the research, but redirecting some of the media’s focus away from aggregate statistics and toward factors impacting distribution may be a challenge, Jacobs said. GDP and employment data is released regularly throughout the year, but distributional indicators are rarer and published much less frequently.
“Right now, the official government-generated indicators that tell us [about distributional patterns] are released on quite a long delay, and so they’re not very useful for the typical economic news article,” he said. The government should do more to collect and disseminate information about how economic gains and losses are being shared, Jacobs said; but even so, he added, newsrooms may have to do more of their own data collection in order not to rely on outmoded government sources.
Overall, Jacobs said journalists should be more critical of the lines they often draw between positive aggregate economic developments and outcomes for people at various spots along the socioeconomic spectrum. “When a new GDP number or new unemployment numbers are released, I think that journalists ought to be a lot slower to trumpet that as economic recovery, or an upturn in the economy — which is a frame that already implies that it’s going to have a broad positive impact on American households,” Jacobs said.
Instead, he added, they should keep in mind that aggregate data are “very indirect proxies for welfare” rather than hard evidence that things are getting better or worse for everyone.
The article “Whose news? Class-biased economic reporting in the United States, ” published April 12 in rhe American Political Science Review, was authored by Alan M. Jacobs, University of British Columbia; J. Scott Matthews, Memorial University of Newfoundland; Timothy Hicks, University College London; and Eric Merkley, University of Toronto.