The Digital Trade Agenda is a relatively new phenomenon. A decade ago, when the USTR went abroad to twist the arms of America's trading partners, the only "digital" part of the agenda was obligations to spy on users and to swiftly remove materials claimed to have violated US media monopolies' copyright. But as the tech sector grew more concentrated, they were able to seize a greater share America's trade priorities.

4/

One person who had a front-row seat for this transformation was Wendy Li, a PhD candidate in #sociology at #UWisconsin, who served in the USTR's office from 2015-17, and who leveraged her contacts among officials and lobbyists (and ex-lobbyists turned officials and vice-versa) to produce a fascinating, ethnographic account of a very specific form of #RegulatoryCapture.

5/

That account appears in "Regulatory Capture’s Third Face of Power," in *Socio-Economic Review*. The article is paywalled, but if you access it via this link, you can bypass the paywall:

https://pluralistic.net/wendi-li-reg-capture

Li's paper starts with a taxonomy of types of regulatory capture, drawn from the literature. The first kind - the "first face of power" - is when an industry wins some battle over a given policy, triumphing over the public interest.

6/

Li notes that defining "public interest" is sometimes tricky, which is true, but still, there are some obvious examples of this kind of capture.

My "favorite" example of horrible regulatory capture is from 2019, when Dow Chemical - working through the West Virginia Manufacturers Association - convinced the state of #WestVirginia to relax the limits on how much toxic runoff from chemical processing could be present in the state's drinking water.

7/

Dow argued that the national safe levels reflected a different kind of person from the typical West Virginian. Specifically, Dow argued, the people of West Virginia were much fatter than other Americans, so their bodies could absorb more poison without sickening. And besides, Dow concluded, West Virginians drink beer, not water, so poisoning their drinking water wouldn't affect them:

https://washingtonmonthly.com/2019/03/14/the-real-elitists-looking-down-on-trump-voters/

8/

This isn't even a little ambiguous. Dow's pleading wasn't just absurd on its face - it was also scientifically bankrupt - there's no evidence that being overweight makes you less susceptible to carcinogens. And yet, the state regulator bought it. Why? Well, maybe because chemical processing is WV's largest industry, and Dow is the largest chemical company in the state. Regulatory capture, in other words.

9/

The second kind of regulatory capture is the #RevolvingDoor: when an executive from industry rotates into a role in government, where they are expected to guard the public interest from their former employers. There's some of this in every presidential administration - think of Obama's ex-#Morganstanley and ex-#GoldmanSachs finance officials.

10/

But while Obama and other "normal" pols sketched their corruption with a fine-tipped pen, making the overall shape hard to discern, Trump scrawled large, crude, unmissable figures with a fisted Sharpie. Remember #ScottPruitt, the disgraced Trump #EPA who wanted to abolish the EPA? Pruitt was was such a colossal asshole that even the lobbyists who'd been bribing him with free housing actually *evicted him*:

https://www.cnn.com/2018/04/06/politics/pruitt-trump/index.html

11/

After Pruitt resigned in the midst of chaotic scandal, he was succeeded by his deputy, #AndrewWheeler - a former coal lobbyist:

https://www.nytimes.com/2018/07/05/climate/scott-pruitt-epa-trump.html

That's the "second face of power." What's the third? It's taking over the *shape* of the debate, getting to define its axioms. Think of the reflexive idea that government projects are "wasteful" and "inefficient."

12/

Once all players internalize this idea, the debate shifts from "what should the public sector do?" to "which private-sector entity should the government pay to do this?" Anyone who says, "Wait, why doesn't the *government* just do this?" just gets blank stares.

We can see this in the cramped and inadequate debate over the #SVB #bailout; apologists for the bailout insist that it was necessary.

13/

After all, if SVB's depositors had been forced to take a haircut, every large depositor in America would pile into Morganstanley, making it so "too big to fail" that it could tank the nation.

This is probably true - but only if you discount the possibility of establishing a *public* bank. Public banks are hardly a radical idea: America had nationwide public banking through the postal service until 1966:

https://pluralistic.net/2023/04/15/socialism-for-the-rich/#rugged-individualism-for-the-poor

14/

Li summarizes: "the first face of power is measured through the winner of the game, and the second face of power can be understood as the referee. The third face of power is the field, the rulebook, and agreement that there is even a game at all."

It's the creation of this third face that Li's paper dissects - the creation of "Type I" ideas that form the unquestioned assumptions for all other debate. Sociologist call these ideas "#schemas."

15/

Li describes two ways that the tech industry changed the schemas used in trade negotiations. First, schemas are changed through "knowledge production" - creating reports and data.

Second, schemas are embedded through "recursive institutional reproduction" - a bit of unfortunately opaque academic jargon that is roughly equivalent to what activists call "#PolicyLaundering."

16/

That's when an industry can't get its way in its home country, so it leans on trade reps to include that policy in a treaty or trade deal, which transforms it into an obligation at home.

In tech policy, the Ur-example of this is the #DMCA, a 1998 digital copyright law that has profoundly changed the way we relate to everything from online services to our coffee makers.

17

For instance, requiring a separate, negotiated copyright license for every transitory copy made by RAM, or a network buffer, or drive cache:

https://www.wired.com/1996/01/white-paper/

Gore laughed Lehman out of the room and told him to hit the road. So Lehman did, scurrying over to Geneva, where he turned his batshit ideas into the #WIPOCopyrightTreaty (#WCT) and the #WIPOPerformancesAndPhonogramsTreaty (#WPPT).

19/

Then he raced back to DC where he told Congress that they had to get on board with those UN treaties. In 1998, Congress passed the DMCA, turning a failed regulatory policy into a federal law that endures to this day.

That's "policy laundering." Lehman couldn't get his ideas though the US government, so he rammed them through a UN agency, converting his proposal into an obligation, which Congress duly assumed.

20/

The Digital Trade Agenda triumphed by both knowledge production and recursive institutional reproduction (AKA policy laundering). Under Obama, trade officials created the Digital Trade Working Group in consultation with industry, through the US Chamber of Commerce. This group worked with the US International Trade Commission (USITC) - a quasi-governmental research body - to produce copious reports, testimony and data in support of a focus on "digital trade."

21/

In particular, they inflated the value of digital trade to US officials, convincing them that getting wins for the digital industry would have an outsized impact on the US economy. This is reflected in the terms of the #TransPacificPartnership, a trade deal that was negotiated in the utmost secrecy, in hotels all over the world surrounded by armed guards, where neither the press nor activists were welcome.

22/

TPP represented a kind of farcical wishlist for America's corporate giants, including the tech sector, and it looked like a done deal - until Trump. Trump unilaterally withdrew from TPP, so the tech industry's reps simply tacked around TPP. They took everything they'd wanted to get out of TPP and crammed it into the #USMCA, Trump's rewrite of #NAFTA. This makes perfect sense - corporate America's priority was TPP's policies, not TPP itself.

23/

Li's paper doesn't just document this shift, she also gives us interviews with (anonymized) officials and lobbyists who speak frankly about how this happened behind the scenes. For example, a former Commerce official turned tech lobbyist describes how he lobbies his former coworkers:

24/

"Sometimes, [meetings are like] hey, let’s grab lunch, let’s grab coffee, and catch up. And half of it is about our kids, and half of it is about this [work related issue]. We’ll have a formal meeting [with government officials], but obviously we chitchat before and after. Because we’re human. So, a lot of it is just normal human interaction, right?"

25/

This social coziness lets lobbyists position themselves as "stakeholders," which legitimizes - and even requires - their participation in policymaking. As a trade negotiator says,

26/

"So to get your handle on a problem, you’ve got to pull the right people together, and you’ve got to sift through all the various ideas, so we obviously have a lot of regular interaction with companies [. . .] I spend a lot of time with the companies trying to understand their business model, try ing to understand how they interact with the governments in different countries, and then of course, socializing it within the building."

27/

Once lobbyists are "stakeholders," they get to define not just what position the US takes - they get to define which positions can even be considered. As a trade negotiator says:

28/

"[Lobbyists don't] spout talking points. They’re not giving us draft text because we haven’t gotten to the text phase yet. The way these meetings go is, generally we provide an update on what is happening and what approach we’re taking. The remainder is usually devoted to companies talking about their particular interests, and inquiring as to whether and how their issues are being addressed in that forum."

That's not just winning the game - it's defining the rules.

29/

Li's paper is a fascinating tour of the sausage-factory and a close examination of the gunk that litters the factory floor. That said, I think there are areas where she drops policies and fights into neat categories that are much messier. For example, Li contrasts the rules in TPP with the rules in ACTA, the #AntiCounterfeitingTradeAgreement, a failed international treaty from 2010.

30/

Li characterizes ACTA as being anti-tech because it imposed copyright liability on tech companies, which would have raised their costs by forcing them to police their users' speech, items for sale and uploads for copyright infringement. But that's not quite right: ACTA was *much* broader. First, because "counterfeiting" doesn't mean what you think it does: in an international trade agreement, counterfeiting concerns itself with all kinds of totally legitimate activities.

31/

For example, Apple engraves microscopic Apple logos on every part in an iPhone; no user ever sees these parts. But Apple uses the presence of an Apple trademark on these tiny components to lodge trademark claims with US border officials in order to block the importation of parts harvested from dead iPhones, as part of the company's war on repair:

https://pluralistic.net/2022/05/30/80-lbs/#malicious-compliance

32/

Likewise, companies like Rolex and Cartier have national subsidiaries in countries all over the world with the exclusive license to sell their goods in each country. These companies then claim that, say, an official Mexican Rolex watch becomes a counterfeit Rolex the minute it crosses the US border, because Rolex Mexico doesn't have the right to use Rolex International's trademarks outside of Mexico.

33/

Asking tech companies to police "counterfeits" isn't just about stopping knockoffs - it's about letting multinational corporations control *all* secondary markets for their goods, giving them total control over repair and used goods.

34/

Beyond that: creating an affirmative duty for platforms to police their users' uploads and speech for copyright infringement is one of those things that not only won't prevent copyright infringement (beating filters is easy for dedicated copyright infringers), but it will also compromise users' speech (because filters are rife with false positives).

35/

And it will hand eternal dominance to the largest tech firms (both #Youtube and #Facebook support mandatory filters, because they've spent hundreds of millions on them, and know that their small rivals can't).

36/

ACTA wasn't a way to "punish" tech to make life better for media companies - it was a way to shift some of the oligarchic control of both tech and media around, while shoring up its dominance. Yes, parts of the tech sector hated ACTA, but it died because millions of *people* campaigned against it.

37/

And of course, ACTA got policy-laundered into law in 2019, when the EU adopted the #DigitalSingleMarketDirective and created a filtering mandate, ignoring the largest petition in EU history and the people who marched in 50 cities. That was recursive institutional reproduction in action all right.

Likewise, TPP can't be understood as the tech sector sidelining the entertainment companies - because both of them rallied for the parts of TPP that feathered *all* their nests.

38/

For example, the entertainment sector and the tech sector both love rules against reverse-engineers (like Section 1201 of the DMCA), which make it a felony to unlock your books, music, games and videos from the store that sold them to you and take them with you to another player.

Tech loves this because it gets them lock-in - if you break up with Amazon, you have to kiss your Kindle and Audible books goodbye.

39/

Media loves it because it gives them control - DRM stops you from recording Christmas movies between Feb and Dec, when they come free with your streaming service, and that means you have to pay-per-view them in December, when you want to watch them.

In other words, the Big Tech and Big Content's policy fights aren't so much about which policies we get - they're about who gets to profit from them.

40/

They both want the same stuff - no taxes, no unions, no minimum wage, no consumer rights, no privacy - but they each want to hoard the benefits from that stuff.

Both tech and media love "IP" - not in the sense of "copyright" or "trademark," but in the sense of "any law that lets me control the conduct of my competitors, critics and customers":

https://locusmag.com/2020/09/cory-doctorow-ip/

41/

In USMCA, it wasn't just the "Digital Trade Agenda" that made it into the final agreement - it was mandatory DRM laws, massive copyright extensions, and the evisceration of fair use and its equivalents in Mexico and Canada:

https://pluralistic.net/2020/08/01/set-healthy-boundaries/#la-ley

There's another important factor missing from Li's analysis of the rise of the Digital Trade Agenda: #monopoly.

42/

Tech used to be composed of hundreds of competing firms that hated each other's guts and were incapable of working together. The entertainment industry, by contrast, was already hugely consolidated and able to lobby effectively as a body.

That was hugely important in the Napster Wars, when international copyright proposals like the Database Right and the Broadcast Treaty were popping up at the UN and in country-to-country trade deals.

43/

While the tech industry was competing to give users a better deal, Big Content was able to solve the #CollectiveActionProblem and come up with a common lobbying position, getting nearly identical (and absolutely ghastly) tech bills introduced in dozens of state legislatures at once:

https://web.archive.org/web/20030425210736/https://www.eff.org/IP/DMCA/states/200304_sdmca_eff_analysis.php

44/