Q & A: Fact Funding

Hamilton: considering the future of journalism. Megan Morr

The field of journalism isn't looking so hot right now. With revenues in sharp decline, papers large and small are scrambling to cut costs, often laying off news staff. Media economist James Hamilton, Charles S. Sydnor Professor of public policy studies and a professor of political science and economics, is in his first year as director of Duke's DeWitt Wallace Center for Media and Democracy. In a new strategic plan, developed this past fall, Hamilton, the author of All the News That's Fit to Sell: How the Market Transforms Information Into News, worked to develop several initiatives that will allow the center—and its students—to play an important role in shaping the future of the field.

You're very concerned about the preservation of the watchdog role of the media.

When you look at where news markets are today, I think that it's pretty clear that the biggest market failure lies in the threat to accountability journalism. People have four different types of information demands: producer, consumer, entertainment, and voter. The first three work pretty well, because people seek them out. For voter information, the fact that you're not really the decider, that your single vote will not determine the outcome of the election, means that many people remain rationally ignorant about the details of politics. That means papers often don't have a profit incentive to engage in significant watchdog or accountability journalism.

Is this a new trend?

You used to be able to do well and do good at the same time. When families or individuals owned newspapers, they got a psychic return to being good community citizens. When your profit rate was 35 percent and you said, "Okay, I'll engage in some accountability journalism, and that might make it 30 percent," that was still feasible. Then, in a world of Craigslist, you see the collapse of the advertising revenue. What's most at risk is this expensive information that's not in high demand.

One possible solution the center is looking into is nonprofit ownership of media outlets. Can you explain how this works?

More nonprofits are beginning to try to provide information that's then taken up in the commercial media. There's ProPublica, an experiment funded by the Sandler Foundation. The Sandler family set aside $10 million a year for three years to create a national investigative reporting team led by Paul Steiger from The Wall Street Journal; he now has a staff of twenty-five reporters, and they're breaking investigative pieces and partnering with commercial outlets like 60 Minutes in order for the stories to get broader play. For media outlets, it's free reporting.

The Knight Foundation is trying to convince community foundations that they should start subsidizing provision of information in their own communities. The idea is that community foundations care about education. Guess what? All the newspapers just fired their education reporters as a staff cutback. So the Knight Foundation is trying to get community foundations, through a matching grant program, to figure out if there is a way they could pay for the creation of information about schools in their areas.

The idea that there might be more nonprofits owning or creating and providing or subsidizing information is something we're going to do a conference on in May at Duke. It really will be focused on examining the hurdles to nonprofit ownership.

By hurdles, do you mean finding funding?

On the newspaper side, we see they're very skittish about taking money from anybody. But the idea that the advertising model is diminishing is making some of them reconsider this notion that they wouldn't take money from a foundation or a nonprofit.

There's a saying on the Internet that information wants to be free. That's true once it's created, but somebody needs to pay for the story. In the old days, people created stories if subscribers were willing to pay to read them, or if the attention of readers could be sold to advertisers.

The other [possible] incentives are, "I will create this story if it changes what you think about"—that's what nonprofits or foundations do. Or, "I will create this story
if I can get your vote," and that's what a political campaign does with something like my.barackobama.com or change.gov.

What about other potential ownership arrangements? Last spring, The Chronicle of Higher Education ran an essay suggesting that elite universities might bankroll newspapers. Can you see that model working?

There are restrictions on nonprofit ownership of commercial enterprises, but there are exceptions for educational institutions. Nelson Poynter, when he owned The St. Petersburg Times, said, "I've never met my great-grandchildren, and I might not like them," so he decided to give away his newspaper. He created the Poynter Institute, a nonprofit educational institute, which now owns The St. Petersburg Times, and they demand a lower profit rate than for-profit shareholders would.

There are examples of public radio stations owned by universities, so it could be an option. Yale has started an environmental Web magazine, Yale 360, that's very good. One thing that makes people skittish is, what happens when the university is the big player in town? Who's going to watch the watcher?

One other thing you talk about in your strategic plan is the tension between privacy and Web advertising.

We'd like to spark a debate about privacy and monetizing the attention of people who are interested in hard news. When I talk to my students in my media-policy class, I say, "Would you be willing to trade off some of your privacy for a better deal?" and most of them say no. And then I say, "Who has a Harris Teeter [grocery store] card?" and they raise their hands. Then I say, "You've done it. You've traded off getting to that lower price by giving them information about who you are and what your purchase patterns are."

Right now Google and Yahoo know a lot about each of us: what we surf, what we search for. But when you search Google and end up reading something at The News & Observer, right now you and I would see the same mortgage ad. If you allowed the company to use more about what they know about you, and share it across companies to target ads to you, that could raise the advertising rates.

Which could, in turn, fund more news reporting?

The question is, are people who say they are concerned about the future of journalism willing to let Google and Yahoo behaviorally target them? It's obviously a tradeoff, but people don't really see it that way right now. We'd like to generate that discussion. The people who say, "I really care about the future of journalism," are often privacy purists. If in order to get the information created, you need to give something up, and that something is part of your privacy that gets a better targeted ad, is it worth it?

In order to fill in the public-affairs reporting gap, you've talked about something called computational journalism.

That's the idea of taking publicly accessible data and combining it with algorithms and knowledge about social science to basically create information that's a supplement or a substitute for current watchdog coverage.

This is something we've talked about at Duke. The Knight Foundation believes in our idea of computational journalism strongly enough that they allowed us to redefine [a faculty position] so that we now have a search going on for the Knight Chair in computational journalism.

Can you give me an example of how this works?

There's an organization called MAPLight.org that has a product they call the "influence explorer." They pull data about campaign contributions and pair them with earmarks data and Congressional offices' press releases about projects and look for correlations. That could be a tip: "Oh, look at that pattern. Let's try to investigate it more."

On a local level, there's a great site called Everyblock, funded by the Knight Foundation, that takes public data streams from about eleven different cities and gives you, say, crime data. It allows you to type in your address and then see the crimes around your area, but it doesn't give you context or analysis over time.

But you could imagine the following: You type in your address, and it will say, "Over the last two weeks there have been three breaking-and-enterings in your area. Police have charged one person with one of those crimes. The News & Observer and The Herald-Sun covered two of those crimes, and here's what they have to say about them. The police have not increased patrols in your area, and this is a higher level of crime than we would normally expect, given what has happened last year, and here's what it looks like where those crimes took place."

All of that could be done by algorithm, but it's not today because the industry doesn't have the ability to engage in R&D in an era when their stock price is plummeting and they're firing significant numbers of people. I see academia potentially playing a role to try to develop scalable open-source algorithms that would allow you to supplement reporting—giving journalists tips—or to substitute for reporting, creating stories themselves.