Regulatory Studies

October 22, 2020 9:45AM

To Improve Content Moderation, Formalize the Collaborative Appeal

Banned social media users have long petitioned for reinstatement via informal appeals amplified by prominent supporters. They may be public campaigns, involving supportive hashtags and mass retweets, or private pleas to platform staff known to the banned user or a verified friend. While platforms provide some formal appeal mechanisms, they rarely provide opportunities to solicit support, clarification, or additional context from other users. At scale, reversing an erroneous ban often seems like a matter of making enough noise to demand a second hearing. Sonya Mann presents this phenomenon as pseudo‐​feudal, highlighting the inherent inequality of informal collaborative appeals.

When regular users run afoul of the algorithm, or are dinged by a moderator, it’s common for them to reach out to accounts with larger followings. The hope is that their pleas will be sent to the top by a sympathetic intermediary. It’s not unlike begging a duke to bring your grievance to the king. Jack Dorsey doesn’t care if some random Twitter account gets shut down, but he might lift a finger if a sufficiently prestigious bluecheck brought it to his attention. Or at least that’s the theory.

As COVID-19 has sent human moderators home and increased platforms’ reliance on algorithms, the value of informal appeals has only increased. This should be of concern to platforms as well as users. Content moderation is a top‐​down process. Unable to draw upon democratic legitimacy, moderators have increasingly attempted to legitimize their governance by adhering to procedural values. The Santa Clara Principles, a set of commitments intended to provide users with due process endorsed by Facebook, Twitter, and Youtube, requires “meaningful appeals” and “the ability to present additional information.” If, in practice, only the loudest accounts have this ability, users’ expectations of procedural fairness will be frustrated.

Last month, @MENA_conflict, a mid‐​sized conflict studies and suburban farming account operated by a former infantry marine was banned from Twitter, apparently after being mass‐​reported by would‐​be QAnon‐​hunters, before being eventually reinstated. The account received a deluge of support from journalists and prominent users, including Jack McCain. Its operator reported that he returned to “like 2500 notifications from y’all harassing twitter to reinstate me.”

Informal appeals may have saved @MENA_Conflict in this case, but they present a broader problem. Platform responsiveness to such mass appeals undermines the legitimacy of moderation as a neutral process. They are not equally accessible, often in ways that mirror off‐​platform power structures. However, these sorts of public appeals often provide moderators with otherwise inaccessible or illegible context. Knowledge of a videographers’ portfolio, past citation of a pseudonymous expert, or the jovial nature of #bitcheswhobrunch can all help inform platform decisions. Unfortunately, from the outside, it can be difficult to know whether moderators have caved to public pressure or independently determined that their initial decision was made in error.

This is not exclusively a Twitter problem, though the platform’s default openness may make the issue more visible there. A formal mechanism for the provision of additional context would improve other platforms, as well. Ford Fischer, an independent videographer, was banned from Facebook after posting footage of armed protestors before being reinstated after appealing to his followers on Twitter. He writes:

I wish I had a formal avenue to say, “I cover — not participate in – activism including that of paramilitaries in my capacity as a journalist.” Instead, I had to resort to a viral post on Twitter.

Off‐​platform calls to action have long been a popular method of informal appeal. When Rose McGowan was locked out of her Twitter account after castigating Ben Affleck for denying knowledge of Harvey Weinstein’s predatory behavior, she took to Instagram to protest the decision. However, like other sorts of informal appeals, off‐​platform pleas turn on celebrities, placing them beyond the reach of most users. In Fischer’s case, one Facebook employee even contacted him via Twitter to tell him that Facebook staffers had submitted internal appeal requests on his behalf. These may have gotten the job done but, as Fischer explains:

My qualm here is that most people do not have the extraordinary following that I do. When I had a much smaller following, I could have just as easily lost my account without the community backing to speak out.

Platforms could formalize the ad hoc process of collaborative appeals and bring it back on‐​platform by allowing banned users to tag their followers or those familiar with their case in the appeals process. Tagged users could then choose to submit written testimony or pertinent evidence that might alter moderators’ analysis of the incident or content in question. In many cases, moderators might not need that much additional information. The knowledge that a given user is a journalist merely covering a riot or that a quote deemed hateful comes from the Declaration of Independence is not difficult to convey.

Providing a formal mechanism for collaborative appeals would reduce the perception of unfairness that accompanies platforms’ responsiveness to the informal process, potentially making such appeals more useful to moderators at the same time. It would provide a counterweight to mass bad‐​faith reporting a frequent misuse of platform user reporting tools. While moderators have limited resources to examine additional evidence, the mere fact that a given decision receives substantial pushback often indicates that it warrants further review.

User flagging is undoubtedly a useful and necessary feature that helps platforms catch violative content their moderators might otherwise miss; like other social media tools, however, it can be used irresponsibly. Often guided by quote tweets, users report non‐​violative content en masse merely because they disagree with it or dislike its author. Sometimes, as may have occurred in @MENA_Conflict’s case, this deluge of false reports is enough to spur platform action. While user flagging remains valuable despite this potential for abuse, the effects of its misuse could be curbed by the addition of collaborative appeal features. Given the difficulty of modifying the reporting feature to prevent bad faith flagging without undermining its usefulness, formalizing the emergent norm of meeting brigade with brigade seems like the best way forward.

Obviously, this would not eliminate inequalities between high and low follower accounts, but it would allow small accounts to access some aspects of the informal appeal currently enjoyed by larger ones. It might also enable moderation to be more responsive to context without forcing moderators to pick between competing sets of facts. Determining context, or the correct context in which to view some content, remains one of the most difficult aspects of content moderation. However, there are good reasons to privilege, or at least recognize, the meaning of speech as understood by its most immediate recipients or participants in the original conversation. They are its local, intended audience those most likely to appreciate its meaning.

Use of nonstandard English vernacular is often deemed offensive by overzealous algorithms, despite universal agreement as to its inoffensive connotations by the original speakers and listeners. Perhaps we could think of it as recognizing the “original public meaning” of speech, accepting that, as in the physical world, the internet’s little platoons often have their own dialects. Like a common law jury of one’s peers, a formal collaborative appeal mechanism could provide an understanding of local norms and conditions likely to be overlooked by external experts.

Users tagged in an ostensibly threatening post could easily be queried to provide additional information, for example, that the post’s author was merely quoting her friend’s lighthearted threat to kill her husband if he removed her plate. This low‐​hanging fruit – preventing moderation from treating conversations between friends as harassment – might not need a human in the loop at all. Machines may struggle to glean context, but they can accept a “not offensive” input from an apparent victim.

Features that allow users’ friends or participants in a specific conversation to offer testimony would provide benefits for large and small accounts alike. The informal appeal has emerged to solve a particular problem inherent to content moderation at scale and, inegalitarian as it may be, it cannot be wished away. However, by formalizing this redress mechanism, platforms can transform an elite privilege into a tool for everyone.

October 21, 2020 3:54PM

Property Taxes on the Ballot


The most famous ballot question on fiscal policy is California’s Proposition 13 passed in 1978. Voters passed the constitutional change to limit property taxes statewide by a large margin, 65 percent to 35 percent.

Howard Jarvis led the drive for Proposition 13 and made the cover of Time magazine, as shown. California property taxes are generally capped at one percent of purchase price plus an annual inflation factor. The large margin for Proposition 13 passage was impressive given that only some voters are property owners while many are renters.

This year Californians will decide whether to partly scrap the property tax limit. Proposition 15 would repeal the cap for commercial and industrial properties to raise taxes by up to $12 billion a year. These properties would be assessed at market value, not purchase price.

The teacher unions are the main funders of the campaign to pass Proposition 15. But Mark Zuckerberg and his wife, Priscilla Chan, have also contributed $7.1 million. The tax‐​hike side has substantially out‐​raised the opposition side of the ballot campaign.

My intern, Hunter Brazal, helped me tally statewide votes on tax issues over the past decade from Ballotpedia. We found that voters supported the anti‐​tax side on 60 percent of income tax measures and 76 percent of sales and excise tax measures.

What about property taxes? We found 72 statewide property tax measures on the Ballotpedia list, many of which were narrow tax cuts for specific groups. Voters supported the low‐​tax side 81 percent of the time. We don’t know how complete the Ballotpedia data is, so our results can be considered a sampling of the universe of statewide tax votes.

Will Californians defend their wallets against the teacher unions? Is the fiscal conservativism that fueled the 1978 tax revolt still alive or has it waned as state politics have moved left? Polls indicate that Proposition 13 continues to be popular, but support for hiking taxes on commercial and industrial properties has grown. Californians will deliver their verdict on November 3.

For background, see this Cato study on Proposition 13 and other state budget limits.

October 21, 2020 3:51PM

Searching for Monopolies

The Justice Department announced Tuesday that it was launching an antitrust lawsuit against Google alleging that the search giant’s deals with browser and operating system developers to make Google a default search engine amounted to anticompetitive behavior. The suit bears all the hallmarks of a political stunt—an unnecessary government intervention in the online search market that has little chance of yielding any meaningful benefit to consumers.

Oddly, the suit does not target Google’s dominance in the online advertising space, which has often been the focus of critics, but Internet searches, where it seems least plausible to claim the company enjoys anything like a monopoly. Internet users have a wide variety of easily‐​accessible options for online searches: While Google is the default search engine for most browsers and mobile operating systems in the United States, users can elect to use competitors such as Yahoo, Bing, and DuckDuckGo with almost no effort, either by manually visiting those pages, or by taking a few seconds to change their default engine settings. Though Google commands the lion’s share of search traffic, it is hard to seriously claim this is because consumers lack for choices—which would normally be a precondition of claiming a company enjoys a “monopoly.”

Just as Google itself rapidly displaced many older search engines like AltaVista and AskJeeves—which had become defaults for users in the 1990s—the company would quickly lose its dominant position if most users found that competitors yielded more relevant results, just as users routinely download and install apps that provide superior functionality to those already installed on their devices. Moreover, Google’s arrangements with operating system and browser developers are not materially different from, or more “anticompetitive” than, analogous deals for prominent product placement in stores familiar from brick‐​and‐​mortar markets. Indeed, those arrangements may produce consumer benefits by subsidizing the production of software that is free to the user. Nor is it credible to claim, as DOJ does that Google has established a stranglehold on search defaults by dint of its deep pockets: Its primary rival in competitive bids to be “default search engine” is the not‐​exactly‐​penurious Microsoft.

Some relevant evidence comes from the European Union, where antitrust regulators recently forced Google to unbundle its search and browser offerings from its Android mobile operating system. Android users are now presented with a menu of search engine choices when setting up a new device, rather than defaulting to Google. The primary result has been that Google began charging mobile manufacturers to license its operating system. Google’s share of the online search market in the EU, however, has not changed measurably at all. While it would be unsurprising if, over time, the new “choice screen” eventually shaved a few percentage points off Google’s share, there is not much evidence for the view that lazy and uninformed consumers are being systematically tricked into using a search engine they would eschew if they were only compelled, rather than merely permitted, to make an active choice.

Perhaps most disturbing, there are indications that this suit may be politically motivated. Attorney General Bill Barr reportedly overrode career antitrust attorneys who did not believe the case was sufficiently strong in order to rush a complaint out the doors. Rep. Jim Jordan (R-OH) responded to the suit by declaring on Twitter that “Big Tech” was “out to get conservatives,” and the Attorney General would not let them “get away with it,” though the suit itself is not explicitly concerned with the company’s supposed political leanings. Eleven state attorneys general have joined the Justice Department suit, all of them Republican.

In short, it is impossible to divorce this action from its political context: Pundits and elected officials on the right, not least among them the current president, routinely bash Silicon Valley by alleging (on the basis of, to put it charitably, highly dubious evidence) that tech companies systematically wage a vendetta against conservatives. Now, weeks before a national election, the attorney general has taken a very public swing at one of the primary villains of this narrative, employing arguments that seem conspicuously at odds with Republicans’ traditionally far more restrained view of antitrust. If this weak suit has been rushed to court in order to punish a company that many on the right perceive as harboring a “liberal bias,” it would constitute not just witless economic policy, but an egregious abuse of power.

Disclosure: While I try to not know which individuals or companies in my policy area may also be Cato Institute donors, my understanding is that Google has supported Cato in the past, and (for all I know) may still do so. For context, less than 3 percent of Cato’s total revenue comes from corporations.

October 16, 2020 1:14PM

Maryland Lawmakers’ Task Force Urges Repeal of Police “Bill of Rights”

A working group of Maryland lawmakers appointed by House of Delegates Speaker Adrienne Jones yesterday recommended that the state repeal its first‐​in‐​the‐​nation Law Enforcement Officers’ Bill of Rights, a law that I and others have inveighed against for years.

As I wrote in 2015:

Maryland was the first state to pass a LEOBR, in 1972, and by now many states have followed, invariably after lobbying from police unions and associations. Often the bills are sponsored by Republicans, who seem to forget their normal skepticism of public employees as an interest group when uniformed services are involved.

Despite minor revisions following the Freddie Gray episode, the Maryland law retains its most objectionable features, as I observed in a piece this summer, including a five‐​business‐​day “get your story straight” period in which a department cannot question officers after an incident; rules prescribing that the process of investigation and discipline be reserved in general to fellow officers; and a lid on the release of information in which the public is legitimately interested.

As lawmakers observed at a late September hearing, repeal would still leave in place numerous other mechanisms by which police officers charged with misconduct could resist investigation and discipline, including state civil service rules, union contract provisions, and Supreme Court doctrines giving public employees constitutional rights to contest some dismissals.

Definitions vary, but per one 2015 roundup, at least a dozen other states as of then had enacted bills similar to Maryland’s into law, including California, Delaware, Florida, Illinois, Kentucky, Louisiana, Minnesota, Nevada, New Mexico, Rhode Island, Virginia, West Virginia, and Wisconsin. All states with such a law, and the equivalent laws sometimes enacted for correctional officers, should follow Maryland’s lead in considering repeal. And the truly terrible idea of making everything worse by imposing LEOBR rules from shore to shore, a perennial measure championed in the U.S. Congress by such as Sens. Mitch McConnell (R‐​Ky.) and Joe Biden (D‐​Del.), should be buried on the ocean floor, never to rise again.

October 15, 2020 3:09PM

Accusations of Social Media “Election Interference” Put Online Speech at Risk

social media logos

Earlier this week, the New York Post published articles containing information about alleged emails between Hunter Biden, the son of Democratic presidential nominee Joe Biden, and employees at Chinese and Ukrainian energy firms. Twitter and Facebook both took steps to limit the spread of the articles, prompting accusations of “election interference.” Prominent Republican lawmakers took to social media to condemn Twitter’s and Facebook’s decisions. These accusations and condemnations reveal a misunderstanding of policy that could result in dramatic changes to online speech.

According to Twitter, the company restricted access to the New York Post’s articles because it violated the company’s policies against spreading personal and private information (such as email addresses and phone numbers) and hacked materials. Twitter cited the same policy when it prohibited users from sharing 269GB of leaked police files. Twitter users who click on links to the two Post articles face a click‐​though “this link may be unsafe” warning. The articles in question include such information in images of the leaked emails. Those accusing Twitter of a double standard because the company allows users to share the recent New York Times article based on the president’s leaked tax documents neglect the fact that the New York Times did not publish images of the documents. Although consistent with Twitter’s policies, the decision to block the spread of the Post’s articles on Twitter absent an explanation or context was criticized by Twitter CEO Jack Dorsey.

According to a Facebook spokesperson, Facebook’s decision to restrict the spread of the Post’s Hunter Biden articles is “part of [Facebook’s] standard process to reduce the spread of misinformation.” Compared to Twitter’s response, Facebook’s was less clear.

Whatever one thinks about Twitter’s and Facebook’s decisions in this case the decisions were legal and consistent with Section 230 of the Communications Decency Act. Much of the online commentary surrounding restrictions on the New York Post (head over to #Section230 on Twitter to take a look for yourself) makes reference to a non‐​existent “publisher” v. “platform” distinction in the law.

In brief, Section 230 states that interactive computer services (such as Twitter, the New York Time’s comments section, Amazon, etc.) cannot — with some very limited exceptions — be considered the publisher of the vast majority of third‐​party content. Twitter is not the publisher of your tweets, but it is the publisher of its own content, such as the warning that appears when users click on the two New York Post article links. Section 230 applies to “platforms” and “publishers,” and does not prevent social media sites from fact‐​checking, removing, or limiting access to links.

Some “Big Tech” critics decided not to focus on Section 230 and instead focus on election interference. The conservative outlet The Federalist issued a statement making this claim, as did many others. According to those making the “election interference” claim, the New York Post articles are embarrassing to Joe Biden, and Twitter’s and Facebook’s actions constitute a pro‐​Biden interference in the 2020 presidential election. Conservative pundits are not the only ones making this kind of claim. Senator Joshua Hawley (R-MO) wrote to Dorsey asking him to appear at a hearing titled “Digital Platforms and Election Interference.” Sen. Ted Cruz (R-TX) wrote to Dorsey accusing Twitter of trying to influence the upcoming election. Later he accused Twitter of election interference and supported the Senate Judiciary Committee issuing a subpoena to Dorsey, which is expected to happen this coming Tuesday.

It is one thing for conservative pundits to accuse a private company of interfering in an election. In today’s political climate it is expected. What should send chills down the spine of everyone who values the freedom of speech and the freedom of association is the sight of two of the most powerful politicians in the country making the same accusation and insisting that Twitter’s CEO appear before a hearing and hand over documents related to how Twitter conducts its business.

To portray how Twitter and Facebook handled the New York Post articles as “election interference” has significant implications. Twitter and Facebook limited access to an article that is potentially embarrassing to a political candidate. If such actions can be considered “election interference,” should every content moderation action by a private company taken against any politician or candidate be considered interference? If The Wall Street Journal rejects an op‐​ed written by the Green Party’s presidential candidate is not that also “election interference”? When a music hall owner decides to allow the Trump campaign, but not the Biden campaign, to host a rally is that not “election interference”?

“Election interference” is a term that ought to mean something useful. Unfortunately, conservative commentators seem intent on warping the term so that it means little more than, “moderating content.”

So‐​called “Big Tech” and content moderation will continue to make headlines next year regardless of who wins the presidential election next month. While conservative commentators and activists are convinced that “Big Tech” is engaged in an anti‐​conservative crusade, they should consider that the political left has its own complaints. Bipartisan anger towards Big Tech could result in Section 230 reform or other legislation that puts the freedom of speech and freedom of association at risk. As lawmakers continue to criticize the most prominent social media companies we should remember that attempts to regulate online speech could have disastrous consequences.

October 14, 2020 3:13PM

Judge Barrett’s Record On Civil Procedure—And On Employment Law

Attorney James Wagstaffe writes for Law360 about Judge Amy Coney Barrett’s surprisingly extensive rulings on issues of civil procedure. A couple of big themes:

On personal jurisdiction, like her mentor Antonin Scalia, Barrett is right in line with the (to me, welcome) modern developments in which the high court has pushed back against state courts’ assertion of “long‐​arm” power over out‐​of‐​state defendants. For example, “Judge Barrett regularly cites Walden v. Fiore, where the Supreme Court, in an opinion authored by Justice Clarence Thomas, held that it is the contacts of the defendant — not the plaintiff — that determine the existence of personal jurisdiction.” (The Court was unanimous in that ruling.) While state courts can properly take jurisdiction over an out‐​of‐​state enterprise that has directly targeted the state for substantial business activity, it will not do to identify just any old way, no matter how incidental or indirect, by which the business has had an effect on or benefited from the state’s market. (More: James Beck takes issue with a decision that Barrett joined but did not write, in Mussat v. IQVIA, Inc., on the topic of personal jurisdiction in class actions).

On the kind of injury needed to confer standing to sue, Barrett has vigorously applied the Supreme Court’s influential 2016 decision in Spokeo v. Robins, which required as a prerequisite to sue that a plaintiff have suffered an actual concrete and particularized injury from the defendant’s conduct, as opposed to, say, tripping it up for some regulatory infraction that occasioned no such injury. The lack of such harm, in fact, deprives the court of jurisdiction altogether over the matter. In one of numerous injury‐​standing cases, she ruled that a “blind plaintiff in an Americans with Disabilities Act case, Carello v. Aurora Policemen Credit Union, suing a credit union for not having a text‐​aloud reader lacked standing since he was ineligible for membership” in the organization he had sued.

Read the whole piece here. Meanwhile, I’ve got a new piece in The Dispatch on Barrett’s record in workplace and employment discrimination cases:

According to Sens. Sherrod Brown (D‐​Ohio) and Sen. Elizabeth Warren (D‐​Massachusetts), Supreme Court nominee Amy Coney Barrett is “anti‐​worker”.…

As a libertarian, it wouldn’t bother me if Barrett were an ardent advocate of freedom of contract and property rights.. … [In fact, however,] Barrett has hewn carefully to the precedent and guidance handed down by the U.S. Supreme Court, just as you’d expect from an accomplished appeals court judge…

But that’s how it tends to go when pressure‐​group ideologues compile tidy checklists of cases meant to provide ammunition against judicial nominees. Much, even most of the day‐​to‐​day work of judges consists of relatively routine and technical issues in which emotion plays little role. The role of groups like Alliance for Justice is to jam this work into a “which side are you on” framework based on the notion that the only thing of interest in a case is which side won.

Speaking of those tidy checklists, Ken White deftly dissects one such story making the rounds, about how Barrett (along with every other judge to rule on the case) wouldn’t let an Illinois highway worker sue even though a racial epithet had been directed at him. In doing so, she applied, as the law directed, the Supreme Court’s formula spelling out the requirements of a hostile environment claim. Ken White concludes: “There are plenty of reasons to oppose Barrett without lying or misinforming people about the law.”

October 13, 2020 4:20PM

Joe Biden’s Tired Old Infrastructure Plan

The infrastructure plan recently released by the Biden campaign is a collection of tired ideas that have consistently failed in the past. Too much of the plan is based on last year’s groupthink and not enough of the plan recognizes the new realities that have emerged from the pandemic.

A large part of the plan is based on getting people out of their cars and onto transit and bicycles. American cities have been trying to do this for the last fifty years, spending $1.5 trillion subsidizing transit, and it hasn’t worked anywhere. The plan calls for connecting low‐​income workers to jobs by building more transit, yet people can reach far more jobs by automobile than by transit while auto ownership, not transit subsidies, are the key to getting people out of poverty.

The plan is based on assumptions about transportation dollar and environmental costs that are fundamentally wrong. Transit, the plan says, saves money while cars impose a burden on low‐​income people and produce too many greenhouse gas emissions. In fact, when subsidies are included, American transit systems spend five times as much moving a passenger one mile than the average automobile. Ignoring subsidies, average transit fares are still more than the average cost of driving per passenger mile. Transit also uses more energy and emits more greenhouse gases per passenger mile.

The only token acknowledgement of the changes brought about by the pandemic is the use of the word “resilient” in the plan. But the planners seem to think that transit is resilient as it proposes “expanded public transit systems, giving more Americans an affordable, efficient way to get around without their cars.” In fact, as Hurricane Katrina, the Camp Fire, and other natural disasters have shown, highways and private motor vehicles are far more resilient than mass transit.

Just look at the current pandemic: transit agencies are in financial crises, but the highways are there when we need them, 24/7. As of August, transit ridership was still down by 65 percent, while driving was down only 12 percent. None of this is taken into account by the anti‐​auto parts of the plan.

The plan calls for electrifying cars and trains in order to reduce greenhouse gas emissions. It fails to note that most of the electricity in the country comes from burning fossil fuels so electrification doesn’t reduce greenhouse gases.

Of course, planners want to shift to renewable energy, which will cost trillions of dollars. Adding the transportation system to the electrical grid will massively increase the burden and cost of doing so. California is already suffering rolling blackouts due to its emphasis on renewable but unreliable energy; just think how bad it would be if the demand for electricity were doubled.

The plan mentions self‐​driving cars, but suggests that the government needs to spend billions on “smart cities” to make those cars work. Yet the government can’t even coordinate traffic signals in most cities; how is it going to operate and maintain so‐​called smart infrastructure? Biden apparently never got the memo that virtually all of the companies developing autonomous vehicles are designing those vehicles to rely on existing infrastructure and not to need any smart improvements in that infrastructure.

Finally, it is worth noting that the very first item in the plan is to “create good, union jobs.” Unions are among the biggest backers of the Biden campaign and this emphasis shows that Biden is just catering to his constituency. While that’s not a surprise, many of the problems with crumbling infrastructure emphasized in the plan can be traced to the high cost of union labor and union rules that makes it difficult to keep infrastructure in a state of good repair.

In the end, the Biden plan shows that government can be counted on to pick losers, not winners, and to prefer obsolete technologies and solutions that fail to recognize the changes society has made in recent months and years. Implementing this plan will waste trillions of dollars on infrastructure we don’t need and that we won’t be able to afford to maintain. This will cripple our cities and reduce the productivity of our economy.