What do strong local newspapers do? Well, past research has shown they increase voter turnout, reduce government corruption, make cities financially healthier, make citizens more knowledgable about politics and more likely to engage with local government, force local TV to raise its game, encourage split-ticket (and thus less uniformly partisan) voting, make elected officials more responsive and efficient, and bake the most delicious apple pies. Okay, not that last one.

Local newspapers are basically little machines that spit out healthier democracies. And the best part is that you get to reap the benefits of all those positive outcomes even if you don’t read them yourself. (On behalf of newspaper readers everywhere: You’re welcome.)

Now a new paper suggests that weakened newspapers hurt communities in a different way: by reducing the number of options voters have to choose from.

The paper in Urban Affairs Review, by Cleveland State’s Meghan Rubado and the University of Texas’ Jay Jennings, examines what happens to mayoral elections in cities where staffing at the local daily gets cut. Here’s the abstract, emphases mine:

Newspapers have faced extreme challenges in recent years due to declining circulation and advertising revenue. This has resulted in newspaper closures, staff cuts, and dramatic changes to the ways many newspapers cover local government, among other topics.

This article argues that the loss of professional expertise in coverage of local government has negative consequences for the quality of city politics because citizens become less informed about local policies and elections.

We test our theory using an original data set that matches 11 local newspapers in California to the municipalities they cover. The data show that cities served by newspapers with relatively sharp declines in newsroom staffing had, on average, significantly reduced political competition in mayoral races.

We also find suggestive evidence that lower staffing levels are associated with lower voter turnout.

This isn’t the first paper to connect newspapers’ health to the number of candidates for local office. One paper I’ve long been partial to used the natural experiment of the Cincinnati Post’s closure in 2007 to determine that, in the places where the paper had had the most readers, “fewer candidates ran for municipal office…incumbents became more likely to win re-election, and voter turnout and campaign spending fell.”

But most research in this area has focused on newspapers closing, and longtime Nieman Lab readers know that closures aren’t the big problem (yet). It’s that nearly all American local newspapers have been cut and cut and cut to lilliputian dimensions. The Cleveland Plain Dealer hasn’t closed, but its newsroom is 1/13th the size it used to be. That’s gonna leave a mark (on your local democracy).

Rubado and Jennings tackle this by looking at newspapers that report staff numbers to ASNE’s annual census and that are in California, where local election data is more robustly reported than elsewhere. After limiting their dataset to newspapers that were clearly the paper of record in a community and setting up controls for demographic and other factors, they ended up with 11 newspapers, 46 municipalities, and 246 distinct mayoral elections to analyze.

So imagine a 10,000-circulation daily newspaper, and imagine it at two staffing levels: a robust 18-person newsroom or a skimpy 3-person one. According to Rubado and Jennings’ findings, that newspaper’s city would likely have about 1 additional mayoral candidate running if it was covered by that strong 18-person newsroom than by those three poor overworked souls.

One extra candidate may not seem like much — but logically speaking, it’s the difference between an uncontested reelection and a real race, or between a two-person race and a three-way battle that gives voters 50 percent more options. It’s not meaningless. (Separately, Rubado and Jennings found that the bigger the newsroom, the smaller the chance that a mayoral election will feature only an incumbent running unopposed.)

A strong newsroom was also associated with a smaller margin of victory for the eventual winner. Back to that imaginary 10,000-circ newspaper: With 5 newsroom staffers, the predicted margin of victory in a typical mayor’s race is about 50 percent. With 15 newsroom staffers, it drops to 33 percent. With 20, it drops to 24 percent.

From the paper:

We believe these findings have important implications. Newspapers have long been thought of as essential to the delicate fabric of democracy. In a well-functioning system, citizens need to be actively engaged in their government and aware of decisions made by their elected representatives. Newspapers are a means of citizen engagement, and this study provides evidence of the importance of this link.

When this link is not strong — when newspapers lose professional capacity to fulfill this role — we see signs of waning engagement in the form of reduced voter turnout. We further see indications that as newspapers reduce staffing, political competition suffers in the form of reduced mayoral challenger candidates. We reason that when newspapers have reduced staffing dedicated to local government, mayors are less likely to be held accountable through provision of information about their decisions and behavior.

For this sort of analysis, there’s always the question of causality. Yes, the most obvious story would be “when newspapers shrink, local political engagement goes down.” But it could also be the reverse: “When local politics become more boring, newspapers are more likely to shrink.” Or both trends could be driven by some third force — say, “When an area’s economy is in decline, political engagement goes down and newspapers are more likely to shrink.”

Rubado and Jennings guard against that by including a lag in their number crunching. They look at newspaper declines for one year and then political participation measures for the following year, which limits the degree to which causality could flow upstream.

You can find more work by Rubado here and Jennings here. Rubado primarily researches local governments, so most of her work is less closely tied to Nieman Labby topics, but as a political scientist who looks at the role of political knowledge, Jennings has a number of papers that might be of interest. I’ll highlight this new one, with Nicolas Anspach and Vin Arceneaux, which found that people can actually gain political knowledge from News Feed scrolling without actually tapping through to read any of the articles — but the tradeoff is that they end up being too confident about how much they actually know.