How the Internet is changing elections

Facebook, Twitter, and other social media giants have altered the way political campaigns are run. They have facilitated the emergence of new political movements, and politicized people who might once have been apathetic. They have also allowed for the easy spread of misinformation, and served as a breeding ground for extremist groups. They have made it easier for politicians to speak directly to voters—for both good and ill.

Sam Jeffers is a visiting global fellow with the Ryerson Leadership Lab (external link, opens in new window)  who was a former U.K. managing director and executive creative and strategy director of Blue State Digital (the digital strategy agency that worked with Barack Obama on his two presidential campaigns). Jeffers has led client engagements with institutions ranging from Google and the British Museum to the British Labour Party.

His new project, “Who Targets Me?” (external link, opens in new window) , tracks political advertising on social media. Working with over 12,000 volunteers, the project tracked the use of political advertising on social media services during the 2017 U.K. General Election. Since then, “Who Targets Me?” has spread to Germany, Austria, the Czech Republic, and the United States. The data collected will be used to develop a new model for electoral regulation in the digital age.

On October 18, Jeffers will visit the DMZ (opens in new window)  to lead his workshop, "Taming the Wild West of Online Politics," which will show how the Internet has changed politics, as well as the new threats it poses to democracy. In advance of his appearance, Jeffers spoke to Ryerson Today about his findings.

“Who Targets Me?” started with the U.K. election and used the participation of 12,000 volunteers. How did these volunteers aggregate the information?

By installing the software that we built. When you install the software and use Facebook as normal, we will gather up the advertising that you’re seeing and categorize that as ‘political’ and ‘non-political,’ and take our analysis from there.

The big surprise of the U.K. election was the Theresa May did poorly and Jeremy Corbyn did surprisingly well. How much of that do you think had to do with online advertising?

I don’t think it had that much to do with online advertising. What you see in political campaigns is, if you run an extremely bad one like Theresa May did, it’s going to cost you, and if you run a pretty good one, you’ll do okay. I think that’s basically what we saw. The Conservative Party relied a lot on advertising as its way of getting it message across—a negative, anti-Corbyn message—but obviously in the end, that didn’t seem to help them very much.

Was online advertising a more major part of the last U.S. election?

The problem with projects like ours is, there are lots of anecdotes but not a lot of data. The data we have only applies to late spring of this year. From what I understand, in the U.S. we’re talking about a billion dollars’ worth of Facebook advertising spent in the last campaign cycle. And if that’s just party spending, you’re also looking at all these other third parties who are very interested in the American elections; all this general widening of the political debate where it’s getting more and more extreme in search of more viral clicks and all that sort of stuff. It feels like it was much more significant in the U.S. election than it was here.

I think part of that is that campaigns are much longer, much better-resourced, much closer to the big technology companies, and the campaigns have many more people to try many more things. Even a campaign like Trump’s, which from what I understand wasn’t the best-equipped in terms of digital staff, they can still spend large sums of money and can still do a lot more than the equivalent campaign would be able to do in the U.K. or most other countries in the world.

What were some of the things you learned from launching this model in the U.K.?

There were some things we were maybe looking to confirm rather than learn. One thing we were obviously looking to confirm was this idea that the Conservative Party was going to have a big advantage in terms of its data and targeting. They hired Jim Messina of the Obama campaign to form a model of what a persuadable swing voter might look like. One of the reasons for running the project was to ask, ‘Who is that? How accurate is that model? Who are they after?’ and work out if we could track that in real time. And that was really interesting, to see where they thought their swing seats were.

Another thing we learned was how negative campaigning has really come to the U.K. for the first time. In the U.S. negative campaigning is obviously always a big part of it, and it always has been to some extent here, but you have to campaign publicly using more traditional media outlets. You have to be prepared to design and put up a poster about Tony Blair being the devil—which was a very famous ad here in the past. Now, I can create an ad and target it just at an audience that I think will be receptive to it. So on both sides of the line here during the campaign, we saw big investments in negative advertising, targeted at specific audiences. That’s new, and that is different, and it’s not something that’s an unequivocal good in the conduct of elections.

We think there’s some interesting stuff there for the future about: Is this really the way we want to go? Where we’ve got both sides trying to get their votes out, but also using negative ads to try to keep the other side’s votes and home, and using suppression tactics. That is probably the most interesting piece, because targeted advertising allows us to do that.

To what extent do you think there are positive developments? I’m giving away my biases here, but a movement like Black Lives Matter or a campaign like the one run by Bernie Sanders wouldn’t have been possible 10 years ago. The internet has allowed these ideas—which never would have been brought up in the mainstream media—to be more easily disseminated.

For every Bernie Sanders there’s a Donald Trump, and there’s an alt-right, so we can’t necessarily say that because we’ve seen the rise of some good things—my biases are potentially similar to yours—that we haven’t also seen the rise of some bad things. What the internet clearly allows is for people with similar interests to get together, and that is something that has been around now for 20 years one way or another. What’s interesting about things like Facebook is the filtering and refinement of that to the point where you really don’t escape it. If you want to see Bernie Sanders stuff in your Facebook feed, that’s what you’ll see, and it feeds you with what you want. I think that’s problematic.

What’s the future of this project?

We’ve already taken it to Germany and Austria. The big one is probably the U.S. midterms next year. We’re interested in the fact that Facebook has announced that it’s going to be publishing many more ads that it currently does, so in some respects that does some of the work that we are doing already. On the other hand, one of the things they’re starting to do is take down material that they consider to be against their terms and conditions. I think it would be interesting to have more of a record of that sort of material in the future.

We’d like to extend into more platforms—we’re particularly interested in YouTube advertising. The future is more, better technology; more countries; making it easier for the people who use our software to understand how they’re being targeted; and to start thinking about what is our advocacy role. What should people like election regulators do in response to challenges like social media advertising?

Is it too early to say what you’d like Facebook to do?

I’ve got some ideas about what I think people at Facebook should do, but the reality is that it shouldn’t be on Facebook to unilaterally do something about this. Although they’re the biggest, and although they’re doing stuff that can set an example for others, the backstop to cheating in elections is the law. It feels to me like we need to go there and train our regulators globally to understand this better and getting the data that they need to do more tracking in real time—to correct problems while they’re happening instead of waiting two years later. There’s a reactive thing going on in regulation, and they need to kind-of shake the whole thing up a bit and move quicker and do more work with data. Regulators should demand that these companies be more transparent with the way that money is spent on their platforms.

*

Sam Jeffers’ "Taming the Wild West of Online Politics" workshop takes place October 18, 6-7:30 p.m., at the DMZ.