Samuel Corum / Getty / NIH / The Atlantic

Over the past six months, it has seemed like every news story is about the coronavirus, or President Donald Trump’s failed response to it. It’s been a challenge for even crucial issues like racial justice and police violence to break into the media agenda for more than a few days at a time.

Although the events of 2020 present unique challenges to the media—and to humanity as a whole—the narrowing of the media agenda, a phenomenon I call “agenda compression,” dates back to the 2016 election. Former CBS President Les Moonves famously observed of Trump’s candidacy: "It may not be good for America, but it's damn good for CBS.” News networks rode the twists and turns of Trump’s candidacy to record ratings and ad sales, and the spotlight has not turned away from Trump for a second since he won the presidency.

When Barack Obama was president, his name appeared in roughly one in 10 news stories published by major U.S. news sources. During his first term, Trump has appeared in roughly one in four stories.

(Throughout this article, I’ll be using Media Cloud’s “US Top Online News 2017” collection—a set of 32 sources that includes websites for major newspapers, cable and broadcast news networks, magazines, and online-native publications—as a proxy for all U.S. news sources. These sources were the most popular U.S. news sites in 2017 according to data from comScore, Activate, and Alexa, and include The New York Times, Fox News, Breitbart News, and The Atlantic.)

Domestic politics is almost always the leading topic in U.S. media, followed by stories about finance, international relations, and technology. But the unusually intense focus on U.S. politics, and especially the presidency, since 2016 has had major consequences for the rest of the media ecosystem.

In 2014, 7 percent of stories were about sports. By 2019, that percentage had dropped to 5.2. (In 2020, with most sports on hold because of the pandemic, the share is 4.4 through mid-July.) This drop in attention might not be a call for alarm in and of itself—the future of our republic is probably not dependent on ensuring a constant level of NFL coverage—but this change in media focus is indicative of larger shifts taking place.

Media Cloud indexes stories from tens of thousands of publications and adds metadata to each story to assist media researchers. Over the course of a year, stories are sorted into more than 400 topical bins. Some bins are far more populated than others—there are many more stories about “finance” than “flowers and plants.” One shorthand for thinking about the diversity of topics represented in the media is to consider how many topic bins are required to represent 50 percent of stories. In 2013, Media Cloud sorted stories into 443 bins, from “politics and government” to “commuting” (that year’s least popular). The 37 most popular bins contained 50 percent of 2013’s stories.

Let’s contrast that with 2019, when the scandal over Trump’s pressure on Ukrainian President Volodymyr Zelensky began to dominate news coverage. Media Cloud sorted stories into 434 bins that year, but 50 percent of the stories were contained in 24 bins. Picking news stories at random in 2013, you had a significantly higher chance of encountering a story that was about a less popular topic than if you conducted the same experiment in 2019.

News diversity has fallen since 2016’s presidential election, and it’s fallen further in 2020, as stories about the pandemic have taken over the news globally. Fourteen Media Cloud topics—two of them explicitly focused on public health and disease—are all that’s needed to represent 50 percent of the stories published in 2020 thus far. The year isn’t over yet, and topical diversity may increase, but right now, 2020 looks to have a more compressed media agenda than we’ve seen in the past four years of Trump compression. This makes sense: The coronavirus is the only story that’s been able to claim more media attention than Trump himself, and the focus on the pandemic is pushing out any number of other stories that might have garnered attention in a less fraught moment.

The reasons for compression in 2020 are easily apparent. Not only does the coronavirus, which touches the lives of everyone on the planet, demand coverage at the expense of other topics, but many events that normally would be covered have been delayed or canceled. With sports and Hollywood largely on hold, large categories of stories are absent. But the compression that predates this year, going back to 2016, demands a broader explanation.

Nonstop coverage of Trump has been aided and abetted by media that have become more and more dependent on real-time analytic tools such as Chartbeat to determine what readers will click on. The fault is not purely that of editors, readers, analytics tools, or the president—social-media platforms play an important role as well. Opaque and unauditable algorithms govern which stories platforms such as Facebook amplify. Stories that provoke comments or other reactions are likelier to spread than those that simply inform; items about politics and about personal health decisions both meet this criterion. The topical-diversity statistics from Media Cloud suggest that not only are these stories becoming more common, but other stories are becoming less common.

An additional factor that contributes to agenda compression is the fiscal straits that local-news organizations are in. With decreasing revenue, local newspapers have fewer reporters producing less content, which means that key local stories go uncovered and issues that may rise to national discussion might not be reported on in the first place.

What does the compression of the media agenda mean for our society? It’s harder for a novel story to break through the stranglehold that Trump and the coronavirus jointly hold. The stories that do break through are really, really big.

In August 2014, the story of Michael Brown’s killing in Ferguson, Missouri, dominated news coverage. Brown’s name, “Black Lives Matter,” “protest,” or “uprising” appeared in 7.4 percent of stories the week immediately after his death and in 11.3 percent the following week.

The uprisings that followed George Floyd’s killing by a Minneapolis police officer in May of this year captured an unprecedented amount of media attention: 21.5 percent of stories in the week after Floyd’s death mentioned his name, “Black Lives Matter,” or “uprising.” The following week, 32.7 percent of stories included one or more of these terms. Attention to Floyd briefly outpaced attention toTrump, and challenged attention to the pandemic (these events coincided with a temporary decrease in new coronavirus cases in the U.S.).

Brown’s death was a turning point for reporting on police violence against people of color in the United States. Floyd’s death may be a similarly significant turning point in discussions of systemic racism in America. Media coverage of these events is part of what allows a tragedy to become a transformational moment.

My research on Brown and subsequent coverage of Black deaths at the hands of police suggests that a high-visibility media event shapes subsequent media coverage, making it more likely that subsequent deaths will be widely covered. Even so, there’s a finite amount of attention in the news cycle, and there is often fierce competition for that attention. Most deaths of Black people at the hands of police still receive little or no coverage.

Breonna Taylor, a 26-year old Black EMT, was shot and killed in her apartment on March 13 by Louisville, Kentucky, police officers who woke her and her boyfriend while searching their residence with a “no knock” warrant. Only one of the three officers involved in Taylor’s killing has been fired, and no criminal charges have been pressed against the officers. Protests continue in Louisville nightly. But Taylor’s death (which preceded Floyd’s, but did not receive widespread attention until after his death) hasn’t appeared in more than 2.5 percent of stories in a week’s news coverage. In a less compressed news environment, her death and the Louisville protests might have become a major national story.

Another significant story crowded out of the news agenda is the 2020 presidential race. In a normal election year, we would expect to see a wealth of stories comparing the platforms and election strategies of the two major-party candidates. But 2020 is unusual: The front-runner is not just the incumbent president, who shows up in roughly 10 percent of all U.S. news stories anyway; he’s Trump, who captures roughly a quarter of the media spotlight. Add in coronavirus coverage, and it’s not entirely surprising that Joe Biden has appeared in only 5 percent of news stories since he became the presumptive nominee in March. By contrast, at this point in the 2016 race, Hillary Clinton was appearing in 9 percent of media stories and Trump in 13 percent of stories.

At a virtual event in late June, Biden quipped, “The more that Donald Trump is out, the worse he does. I think it is wonderful that he goes out.” If Biden’s appeal to voters is primarily that he’s not Trump, allowing the president to own attention with his visible mishandling of the coronavirus crisis is a promising campaign strategy.

But here’s another possibility: There may simply not be enough media attention to go around to permit the wall-to-wall coverage Americans have come to expect of presidential campaigns. Every four years, thousands of reporters flock to report on a pair of weeklong political infomercials, the Democratic and Republican National Conventions. Perhaps we wouldn’t lose much if convention coverage failed to displace stories about the coronavirus response this summer: Since 1968, even “contested” conventions have merely reaffirmed the results of the primaries. And in 2020, with neither candidate attending an in-person convention, it’s unclear what the coverage will even look like.

This year may be a good chance for the United States to consider whether we really need an 18-month-long presidential election. Elections in Japan are limited to 17 days of campaigning for seats in its upper legislative house and 12 days for the lower house. In many parliamentary democracies, snap elections mean that campaigns rarely run much longer than a few weeks. With the need to focus on the coronavirus and the Trump administration’s response to it, 2020 may demonstrate an uncomfortable truth: The months of coverage of conventions and debates is more helpful in providing advertising revenue to the news industry than it is in informing citizens about the political decisions they are asked to make.

Even if losing wall-to-wall coverage of the 2020 presidential race turns out to be a blessing, the compression of the media agenda should worry us as citizens in a democracy. The protests about Floyd’s killing are a concrete demonstration that protest, media coverage, and social change have an interlocking and synergistic relationship. When important stories, such as the genocide of Uighur Muslims in China, go undercovered, the potential for protest and change are reduced.

One step news outlets could take to combat the compression is to regularly monitor the diversity of topics in their news coverage. Newsrooms that track the performance of their stories using tools like Chartbeat and adjust their coverage accordingly are making a rational decision in the short term. Tracking topical diversity over time may show that this is an unwise decision: A pack of news outlets chasing the same set of stories will capture clicks, but will ultimately lose what makes them special and competitive in a market where another hot take is always a click away.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.