openDemocracy investigates: How does commercial interference affect editorial decisions?

by Mary Fitzgerald
Oct 30, 2018 in Investigative Journalism

This is a guest post from Mary Fitzgerald, the editor-in-chief of openDemocracy. openDemocracy is a not-for-profit global website that publishes independent, public interest journalism. 

“A free press is essential to a healthy democracy. There is a purpose of journalism, and it is not just to entertain. It is not to pander to political power, big corporations and rich men. Newspapers have what amounts in the end to a constitutional duty to tell their readers the truth.”

So wrote Peter Oborne, the UK Telegraph’s former chief political commentator, in a blistering 3,000 word resignation letter on openDemocracy earlier this year.

Oborne had uncovered that the Telegraph, one of Britain’s leading newspapers, had suppressed negative coverage of the international banking giant HSBC in order to protect its advertising revenues. (HSBC is a major advertiser with the Telegraph). He also alleged that this type of editorial ‘protection’ extended to a number of other major advertising clients.

The story, which broke right in the middle of the International Consortium of Investigative Journalists’ HSBC-Swissleaks revelations, made headlines worldwide and prompted a stream of similar (confidential) allegations from journalists from other outlets. Most were unwilling to speak publicly, for fear of losing their jobs.

But now openDemocracy is developing a major project called Media UnSpun. We’ll investigate commercial interference in editorial decisions – without asking journalists to risk their livelihoods. We plan to pilot this project in the US, the UK and at least one non-western country, which we’ll decide on the basis of the strongest leads we can gather.

Our starting point will be traditional journalistic methods: We’ll talk to reporters working across the industry about where they believe such interference exists and investigate their claims. Then, we’ll use data and media monitoring algorithms to factcheck their allegations. Journalists won’t have to go public and can tip anonymously if they wish.

How will this work? To give a purely hypothetical example: a journalist who works for The New York Times tells us that her paper is strongly averse to publishing negative stories about Shell, a major advertiser. A big oil spill occurs at an offshore Shell refinery, so we use our algorithm to scan all the NYT’s coverage and pull out relevant data about how that outlet covers the story – and then we compare that with selected competitors that don’t carry advertising or sponsorship from Shell.

Data for comparison will include how many stories are written, how prominently those stories are displayed in print and online, how they are pushed on social media, newsletters and through other promotional mechanisms, as well as indicators about editorial emphasis such as headlines, and length and prominence of quotes of the relevant parties (ie. who is quoted first and/or at more length, the Shell PR representative or a member of the community affected?)

Using this data to help factcheck allegations of editorial bias is now possible with media monitoring algorithms.

A fascinating example of the potential of such algorithms was provided by the UK Media Standards Trust’s Election Unspun project, which pulled out data about media coverage of the UK election earlier this year. We are now working with the with the team behind the Election Unspun project to raise funds for a ‘test-case’ six-month pilot to analyze UK media coverage of climate change in the run up to the Paris climate summit in November-December, the most important meeting of world leaders on this issue in half a decade.

As with our Telegraph expose, it is very likely that media outlets implicated will seek to challenge or discredit some of our most significant findings – arguing, for example, that the absence of a story does not prove that it has been spiked or suppressed. This is why it’s so important to use strong first-hand evidence to lead us to the data.

Even though the Telegraph issued a swift and full denial, we had a highly credible source, and the outpouring of evidence after the story broke – from leaked internal memos in the press to confidential allegations put directly to us – put its credibility beyond doubt. The story also had an internal impact on the Telegraph. Even as they issued denials, several sources confirmed that Telegraph executives were less willing to spike stories unfavorable to advertisers in the wake of the scandal, and the paper publicly committed to new guidelines for employees to reinforce the independence of editorial from advertising.

Of course, the risk with one major story is that it has a short-term impact and then it’s back to business as usual. To fully understand the scope of this problem – and to have a sustained and longer-term impact if and where we find it exists – we need to be scrutinizing and reporting on media behavior over a longer period of time.

This is why we’re embarking on a major, multi-year project. Our ultimate goal is to stimulate and sustain a wider and global public debate about the role of the media: who owns and controls it, who it should serve, how it should be funded – and how to keep it fearlessly independent.

If you agree with us that this is an important debate, and you either want to be involved in this project or have a lead you think we should investigate, write to info@openDemocracy citing Media UnSpun in the subject line.

Main image CC-licensed via Flickr by Diego Martínez Castañeda.