How Loud Does the Watchdog Bark?
A Reconsideration of Losing Local Journalism, News Nonprofits, and Political Corruption
About a year and a half ago, which seems like a century ago, Sanghoon Kim-Leffingwell and I finally submitted our working paper about the links between declines in local news and political corruption in the U.S. And NOW, it’s finally, finally peer-reviewed and published, in a very, very good journal, the International Journal of Press/Politics (ungated version here): and I’m stoked about it - for a few reasons, the most important being that we think we have some strong support that nonprofit news does indeed make democracy more robust and keep politicians accountable.
There were a few assumptions about declines in local journalism that kept bugging me, ones that seemed on point, but also lacking data.:
When we (we=the future of news cognoscenti, concerned members of the public, etc.) say that local news is declining and one consequence is increased political corruption, well, we might have anecdotes to suggest that journalism indeed does a great job helping to hold public officials accountable, but what do we have, specifically in the US context, that really, really shows a demonstrable link between declines in local journalism and increased political corruption?
Moreover, we are pumping so much money into local news nonprofits right now, with one important justification being that the commercial market cannot support public interest/accountability journalism in any sort of traditional ad-supported or even subscriber-supported way. There are now calls to add
$0.75–1.75B To Fund Not-For-Profit Local News (as estimated by Boston Consulting Group)
BUT, do we have any indication, at a systemic sort of level, that news nonprofits are beneficial to democracy? (beyond cases of excellent journalism and in some cases, improved community relations between journalists and the public)… there are lots of older, cross-sectional massive data dump survey studies about free press and news circulation and democracy/government accountability, but most of this hasn’t been updated since… basically the birth of the contemporary social media / smartphone/digital universe of news and information.
As we argue in this paper, all previous efforts have pretty much failed to dig into this link between declines in newspaper journalism, the rise of nonprofit news, and increases in political corruption in the United States looking at the US, itself, as a case study. We think we’ve given this a solid try and… a start to a more solid answer, using federal prosecutions for public corruption as the yardstick ( defined by DOJ as crimes involving the abuse of public trust by federal, state, and local public officials). Let me explain what we asked and answered, and then how/why we followed this approach.
Are declines in local newspaper employment and circulation associated with changes in prosecutions for public corruption?
Sort of. Sheer declines in the number of employed journalists relative to public corruption really don’t seem to matter much. BUT, consistent with all of the studies from years of study on the pre-internet role of journalism in democracy, the circulation of newspapers really does seem to matter, such that the more circulation, the more prosecutions for public corruption.
Are efforts to supplement watchdog journalism with nonprofit journalism mitigating some of the damages to public accountability?
YES. YES!!! We look at this in two ways: first, just what happens when you introduce a news nonprofit (and after some lag) to the prosecutions for political corruption. We then look at funding for these news nonprofits. Funding matters: the more nonprofit $$$, the more prosecutions for public corruption. In this case, we are super thankful to the Institute for Nonprofit News for sharing their membership list with us and for the Media Impact Funders database for listing inflows and outflows of foundation dollars.
Let me explain the approach and our thinking. There is a complicated logical problem with measuring corruption: we cannot measure the corruption we do not see.
In other words, we cannot measure the fact there is more corruption occurring that politicians/public officials engage in when there are fewer people watching because no one is there to actually record that it is happening. Does this makes sense? We can’t see what we can’t see: we can only see the corruption (accusations) when public officials DO get caught. So actually, in a healthy democracy, we want to see prosecutions for public corruption robust and active, not diminished. Evidence of the check is the only evidence we have of the system working. Basically, in cases when democracy isn’t working well and journalism isn’t helping serve as a check against the powerful, we would expect to see fewer politicians/public officials behaving badly not more, in part because the visibility would be blocked as the journalist watchdogs are hindered in their transparency efforts, thus undermining the presumptive system of accountability.
So how do we operationalize this? The only way to do so is to ask whether prosecutions for public corruption declined rather than increased, because this would suggest that the FBI /enforcers of punishment/sanction for badly behaved public officials were somehow hindered in their pursuits because of a lack of information or pressure coming from watchdog journalists.
There is another wrinkle/complexity, which is how to actually measure what constitutes journalism in a community and whether or not it declines. We defined for the purposes of our article the geographical unit as the 94 different federal judicial districts, where the FBI had purview over such prosecutions for public corruption. But how do we assess declines in journalism? I have a larger beef about declarations of “news deserts” that you can read about here in Political Communication, but in this case, I believe what we worry most about is the loss of sustained local newspaper coverage, which has long provided the bulk of original news content in a community.
Scholars like neat measurements, so in the past, this sort of question has typically been considered from the perspective of “newspapers open” and “newspapers closed” (a 1, 0 binary choice if you’ve been following my other summer writing). But to me, open/shut is sort of a limited measure, and whether or not a newspaper exists tells you little about either its resources or its pick up in the community. While there are folks working to figure out the best way to measure these declines in employment using annual Editor and Publisher guides, for now, we look at newspaper employment as a labor count - literally as the annualized estimates from the Bureau of Labor Statistics for news industry employment-publishing at the county level.
OK, but what about reach and availability? Well, enter the formidable and super useful UNC News Deserts dataset maintained by the Center for Innovation and Sustainability in Local Media - and thank you thank you for keeping this dataset open upon request to researchers and for keeping a community going. This data set gives circulation (online/offline as appropriate/measured).
So, then, with these news provision questions divided into supply (employment) and demand (circulation), what kind of political corruption do we mean anyway (and how do we measure it?)? I think the most problematic are public officials behaving badly with public money - what public officials do in their own time probably does impact politics and civic life in many ways, but to me, the worst offense is abusing public trust by abusing taxpayer dollars or the power of the office. For this reason, and for the enduring and the FBI’s counts of prosecutions for public corruption’s somewhat steadfast resistance to politicking at the system level, Sanghoon and I chose to use these counts as our proxy for political corruption (note: these are not convictions, just prosecutions of suspected corruption).
But what about nonprofits? And nonprofit news investments? Here is where we get both lucky and a little mushy at the same time. INN, or the Institute for Nonprofit News, is the industry standard/leader in bringing together a collective of nonprofit news outlets focused on public accountability journalism, broadly described - this can include accountability journalism that specifically covers politics and government as well as journalism focused on education and the environment and beyond, and the unit of specific geographic focus for each outlet can vary.
From the perspective of public accountability, we make the case that accountability journalism does not simply have to be limited to investigative journalism — that’s too narrow - rather, investigative journalism that discloses political wrongdoing is rare in local news, but local news provides a form of “scrutiny” that is related to ordinary news reporting, whereby the press helps inform and critique the performance of public institutions and performance of public officials. So to that end, INN’s membership list, their geographic distribution, and the length of time any one INN member is in a particular judicial district is essentially a proxy for “this place has people that care about journalism and are doing it for the right/best reasons” - with the presumption that having a top notch environmental nonprofit covering Alaska (let’s say) is going to strike fear in the heart of public officials and help overall efforts system wide at government accountability (you can only prosecute what you can spot as problematic).
Overall, we find strong support that nonprofit watchdog journalism can serve an important, supplemental role in shoring up government accountability that can help mitigate the impact of the declines of local newspapers. Furthermore, we also find that philanthropic funding for journalism matters: a well-funded nonprofit watchdog outlet is associated with greater prosecutions for public corruption, Commercial logics undermine the incentive for for-profit news media to continue to invest in watchdog journalism; the support for our questions about the link between nonprofit news provision and funding point to the efficacy and the importance of finding revenue streams to support journalism that is poorly supported by the market outside of commercial logics.
Look, we don’t want to overstate the application of our work as causal: there many endogenous and exogenous explanations that might account for some sort of causal link between declines in newspapers, increases in public corruption. {n.b. I’d like to think my econometrics training this summer can help create better models for causality in these sorts of cases, but this paper was written before BSE. }
And it’s likely that press/politics relations are more complicated and interconnected: questions about the ways in which FBI or SEC investigations might feed journalists who might in turn help the FBI and SEC unearth more evidence which in turn…. and there are just questions about what underlying conditions such as local political cultures and traditions of good governance might shape the norms of conduct for political officials. (I think one addition would be to figure out how to account for SEC prosecutions/indictment as well, although that’s about questions of unfair advantages for unfair gain, and maybe less so say, direct bribes but… moving on)
But here we are - this is a start to actually finding some actionable data and insights about the future of journalism and nonprofit interventions in helping journalism keep democracy functioning and civic life robust.