The Facebook Whistleblower Is Heroic... And Terribly Wrong

Frances Haugen got a lot right. But a digital regulator that legitimizes Facebook's power would be the worst possible outcome.

Hi,

Welcome to BIG, a newsletter about the politics of monopoly. If you’d like to sign up, you can do so here. Or just read on…

Today I’m going to write about Facebook whistleblower Frances Haugen and her testimony before Congress. Haugen came with a specific policy agenda. Does it make sense?

An Impressive Showing

This week, Facebook whistleblower Frances Haugen testified before the Senate, after releasing a series of important documents to regulatory agencies. The document that got most of the attention was that Facebook has internal research showing that teenagers are harmed and in some cases driven to suicide by Instagram. There were others involving human trafficking and polarization, and even one revealing what looks like securities fraud and fraud on advertisers as Facebook lied about its reach and shrinking user base in key demographics.

It was an immensely slick and effective public relations campaign, and devastating to the firm’s image. Haugen offered a lot of great information, and she was compelling, articulate, composed, and authoritative. She was impressive, even if you are somewhat skeptical of her motives. Along with these documents, she also offered some a good policy ideas, like making platforms responsible for the speech they amplify through algorithms (changing the law known as Section 230 of the Communications Decency Act), as well as creating rules to move social media away from an engagement-based business model. Haugen’s goal was, in part, to simplify Facebook as a platform, to make it human scale.

But there is a huge problem with Haugen’s overall policy recommendations. And since she got a lot of attention, her ideas are getting attention as well.

Haugen is a trained designer of algorithms, and along with many naive Silicon Valley insiders turned critics, at heart does not see a danger with concentrated power. “I don’t hate Facebook,” she has said. “I love Facebook. I want to save it.” Her approach to social media is similar to what many left consumer oriented groups support, which is not to take apart a concentration of power, but to regulate it. It is, in many ways, a similar framework as Obamacare and the Dodd-Frank financial reform package, which, rather than making systemic changes to concentrated and bloated dysfunctional sectors, simply overlaid captured regulators on top of them.

In fact, Haugen’s proposal is also very similar to that of… Mark Zuckerberg. Both want to keep Facebook a dominant monopoly. Why? Both Haugen and Zuckerberg think the firm’s market power allows it to make a lot of money, and that money can be reinvested in safety systems and better site features. Haugen thinks that Facebook is a natural monopoly, as advertisers will only learn and finance one social media platform. Splitting off Facebook Blue from Instagram effectively would mean that all the ad revenue would go to Instagram. Facebook Blue, she suspects, would remain a dangerous social network, but would lack financial resources to mitigate problems. 

If the firm stays together, so goes Haugen’s story, then WhatsApp, Facebook Blue, and Instagram will all have plenty of resources to invest in safety. So what does she suggest with this dominant natural monopoly? Her recommendation is to place a separate data-specific regulatory overlay on top of Facebook and its subsidiaries to protect the public interest. This agency, according to Haugen, would allow people who are in between stints at social media firms to join the government and help make regulations on the sector. And here again she joins team Facebook, as Facebook’s Nick Clegg wrote an oped earlier this year recommending just such a regulator.

One of the consequences of forgetting about market structure and the importance of competition is that we wind up centralizing power through bad public policy choices. On the right, this centralizing pressure comes in the form of professional Beltway libertarians, who want centralized power held by Zuckerberg because they fear public institutions explicitly setting rules and prefer private regulation via monopoly firms. On the left, and Haugen falls on the left, it takes the form of calls for a digital regular to oversee - rather than break up - this monopoly power. Gene Kimmelman, who is now at the Department of Justice Antitrust Division, wants such a regulator. So does Harold Feld and advocates like Charlotte Slaiman at Public Knowledge, Karen Kornbluh at the German Marshall Fund, Obama economic advisor Jason Furman, and former FCC Chair Tom Wheeler. I like and respect all of these people, and not all of their proposals exclude stronger antitrust or other important goals, such as Haugen’s idea to simplify the business model and dial back the company’s use of algorithms. And this idea is picking up some temporary steam, with important politicians endorsing it. Here’s Senator Kirsten Gillibrand, for instance.

Sen. Kirsten Gillibrand @gillibrandny

.@FrancesHaugen is right – we need a dedicated regulatory agency to hold Facebook and other Big Tech companies accountable for how their algorithms push misinformation and how our data is used and misused for their profit. We need a Data Protection Agency.

Still, I think a digital regulator is a terrible idea, and I’m going to explain why. Start with the most basic red flag, which is that Facebook’s top leadership wants it too. I’m not someone who reflexively reacts to an opponent, but there’s a reason that this regulated monopoly framework, undergirded by a digital regulator, is shared by both Haugen and Zuckerberg. On philosophical grounds, neither fears centralized power in the hands of well-meaning Silicon Valley executives. Haugen also believes it is inevitable that such a concentration will exist because of the dynamics of advertising markets.

Yet the concentration of power in the hands of a small group is the fundamental political and economic problem with Facebook. We have never allowed one man to set rules for communication networks that structure the information ecosystem of billions of people. But that is the situation we’re in. We have to radically decentralize this power. But a regulatory overlay in some ways would worsen the problem, because it would explicitly fuse political control with market power over speech and it would legitimize the dominant monopoly position of Facebook. (Common carriers, for instance, have an antitrust exemption from FTC rules). The right is suspicious of such a regulator because they are afraid of what Biden and the left would do with it. But I suspect that suspicion isn’t out of place on the left either. If you are a Democrat, imagine, for instance, if Trump were able to pick a regulator for social media, to negotiate with Zuckerberg on how to run global discourse. Better not to have such concentrated power in the first place!

There are a lot of wheels. So why isn’t there a wheel regulator?

If you’re thinking about building out a new regulatory apparatus for privacy and data, what is privacy? What is data? These may seem like simple questions, but they are not. There are deep questions about data in, say, derivatives markets, but that stuff is rightly handled by the Commodities Futures Trading Commission, not a data regulator. Similarly, there are data and privacy questions pretty much everywhere you look, from automobiles to labor markets to the distribution and retailing of office supplies (the FTC put a data firewall when Staples bought a major distributor of office supplies). Data and privacy is everywhere and in everything. Calling for a ‘data’ or ‘privacy’ regulator is like saying ‘hey there are a lot of wheels these days, we should have a wheel regulator.’ But the regulatory model for cars versus trains and airplanes and wheelbarrows is totally different, even though they all have wheels. It’s the same for privacy and data, they aren’t sufficiently coherent to embed in one regulatory agency.

On a practical level, it’s important to recognize that Facebook exists today because of the failure of regulators to take action on privacy, on market power, and on basic rule of law problems such as advertising fraud. We already have agencies like the Federal Trade Commission, which is supposed to police and research unfair practices and work on privacy questions, and the Federal Communication Commission, which does the same in the telecom space. Why would a new regulatory agency be any better? In fact, a new regulator is simply an institutional excuse not to improve the FTC or FCC.

Any new regulator would take years to set up, and it of course would be staffed by people from those agencies, who will be unchastened about their own earlier failures. (We didn’t have the right authority before, or the right institutional setup, they would say. In reality, Facebook shouldn’t have been allowed to buy Instagram or WhatsApp.) Haugen’s twist on this idea is even worse; she wants to staff this new digital regulator with well-meaning Silicon Valley insiders who created the problem in the first place. (Indeed, a cynic might point out that the only difference between Haugen and Zuckerberg is they disagree on which naive Silicon Valley insider should wield monopoly power.) I don’t think any rationale for a new agency makes sense. Let’s fix what’s broken among our regulators, and not ignore it by creating a new one with all the weird turf battles that’ll invite.

Monopoly Power is Upstream from Censorship

Haugen’s right that we should simplify the business model of Facebook by getting rid of algorithmic amplification. To do that, one could ban surveillance advertising, or go after algorithms directly, or both. But it’s also critical to foster competition in this sector by breaking apart Facebook, because without doing so, there will be no incentive for Zuckerberg to do anything but focus on corrupting the new regulator. Currently, there is no market pressure on Zuckerberg to change because advertisers have nowhere else to go. Despite the remarkable series of articles, and the outage, for instance, not a single advertiser has spoken up in any meaningful about Facebook this week. Think about that. And this is why.

Still, there’s a useful question I hear nonstop. Would competition really make a difference? Feld calls this the ‘starfish problem;’ break up a firm like Facebook, and you’ll just get a bunch of equally bad mini-Facebooks. The starfish theory doesn’t make sense, because at the very least, in a world with competitive networks, advertisers could apply pressure by taking their dollars from one network to another.

If there were competition in social media, what would happen? There is a very simple answer. Platforms would compete by differentiating themselves based on safety and privacy. That sounds a bit far-fetched until you realize that that is exactly how Facebook competed with MySpace. Facebook promised it was the safer alternative to the creepy MySpace. Once Facebook bought up all its competition in 2014, it began aggressive surveillance. Privacy and safety is a quality improvement, and firms mostly improve their products if they are forced to by competition. This basic framework, of rules for market behavior around data/ads, plus break-ups, is known as ‘regulated competition.’ It’s very similar to having a bunch of car companies who compete with one another over quality, price, and service, but also seat belt mandates and stop signs that regulate the market to encourage safety.

In other words, we have to eliminate Facebook’s toxic business model, not regulate it, because regulating something serves to legitimize it. It’s not about stopping the collection of data and manipulation, it’s about stopping the practices that make that collection of data and manipulation profitable in the first place. Get it at the source. Monopoly and market structure are upstream from the data, misinformation, censorship and privacy problems. 

Here’s the solution. First, we need to ratchet back Section 230 as Haugen suggests (or even further), which would simplify the business model. Second, we need to break up Facebook, as well as the other big tech firms. Third, we need to ban surveillance advertising. (There are other rules to consider, such as interoperability mandates, but that should happen after the monopoly power is smashed.) The institutional home to do this properly is the Federal Trade Commission, which already has legal authority, research capacity, buildings, and a staff. Already, new FTC Chair Lina Khan is moving in this direction; she has an antitrust suit against Facebook, and has had the FTC combine its consumer protection/privacy apparatus with their antitrust apparatus, ensuring that policymakers are looking at the whole picture.

The FTC now has good leadership, but needs more authority and support from Congress, as well as changes to funding to let the FTC pay its lawyers as much as lawyers at other government agencies are paid. What we don’t need is to legitimize social media monopolies with a separate agency that will be easily captured by the industry it regulates.

Thanks for reading. Send me tips on weird monopolies, stories I’ve missed, or comments by clicking on the title of this newsletter. And if you liked this issue of BIG, you can sign up here for more issues of BIG, a newsletter on how to restore fair commerce, innovation and democracy. If you really liked it, read my book, Goliath: The 100-Year War Between Monopoly Power and Democracy.

cheers,

Matt Stoller

P.S. I’d like to address two other points I couldn’t fit in this essay. First, there’s always the ‘what about TikTok’ question when it comes to competition. But TikTok isn’t a real competitor to Facebook. They vaguely compete in some areas, but TikTok is more broadcast than social, and users didn’t switch to TikTok when Facebook went down for six hours, they switched to Snapchat. '

Second, I do not think Haugen’s analysis of the ad market is correct. We have historically always had multi-channel advertising campaigns put together by ad firms (like those fictionalized in Mad Men)until the very recent vertical integration of the ad business. There’s no reason we couldn’t have an intermediary tier of ad buyers once again if we stopped allowing consolidation. To put it differently, Facebook already lets advertisers buy on multiple networks, it’s just that it controls every one of those networks. If you buy ads through Facebook, they place those ads on Facebook, Instagram, WhatsApp, and something called Facebook Advertising Network. If Facebook’s ad buying tech were split off as a separate company, then that would create an immediate intermediary layer that can buy on multiple networks.