Three Stanford professors on the sensible regulation of Big Tech
thebrowser.com
thebrowser.com
For example: if social media companies were required to ID their users in person within 30 days of sign-up ("think of the children" IDK), the infrastructure to comply with that regulation would make it extremely difficult for a new startups to pull off, while big incumbents with compliance departments already staffed up can make it work.
Even worthwhile safety regulations have this effect. You can't just start a hazardous chemical transportation company in a weekend like you can a software company. I'd say that's working as intended, but it does still give larger players a competitive edge just for being larger/older.
For my 2¢, limiting acquisitions and mergers is the way to go here. The laws already exist, they just need to be more strictly enforced. If a company's core business grows, the company can and should grow. But we don't want a ravenous corporate blob that absorbs everything tangential to its business, whether that's for growth or defensive reasons.
Look what happened to Sears, GM, IBM, GE, HP, Sony, etc. History shows that once corporations get too big, they become unable to adapt, and fall by the wayside.
Among all these calls for "regulation" what is lacking is the broader principle that we will use to regulate not just big tech but for anyone. Google is different than Exxon but I would like the principle that regulations them to be the same. When you look at Texas or Florida trying to pass regulations around tech, they seem pretty terrifying where as California's regulation appears to be a big time nothing-burger.
Same goes for your comment about inefficiencies. That seems to presuppose a particular truth without actually presenting any evidence. Sure there can be inefficiencies from having multiple companies involved. Conversely you also can have better innovation as ideas are allowed to be tested without having to get buy in within one large massive corporation. Think for example about Xerox where their Parc laboratory was regularly churning out ground breaking research that was dying on the vine within Xerox. For what it’s worth that’s potentially the case within big tech (eg Vine getting killed within Twitter and then TikTok showing success).
> That seems to presuppose a particular truth without actually presenting any evidence.
What I have said is mostly classical economics 101. If the point is Google could innovate faster if it is split into 10 companies, Google has all the incentive to do it today or alternatively for a Google competitor to split itself into smaller companies in order to beat Google. In either case, when people are free to chose, they will over time make the most efficient choices. This is part of the reason why markets tend to allocate resources better than government folks.
If the law makes it impossible to try certain things you are killing off innovation along that dimension. Again, this does not need empirical evidence (even though there is plenty) but mostly basics of classical economics. Now, one may disagree with the markets as well and call it "superstitious belief in markets" but then I won't have any rebuttal to that claim.
> “all the regulations that apply to big company X in industry Y have to apply to big company A in industry B”
This is not what I said. The regulation can be whatever, ultimately they will be determined not by Stanford profs but lobbyist behind close doors bribing congress. Principles often are much simpler, easier to explain to people and much easier to judge regulations by looking at the principle. One simple principle could be that "pumping co2 in air is bad". We can agree to disagree but we understand the principle. We can understand if companies who produce more co2 are somehow asked to pay higher taxes.
I tried to read book excerpts from the Stanford professors and it appeared to be complete BS vending. Arguments like "Big Tech is too big" ,"Big tech has too much power" etc. simply does not make sensible principles as much as they make excellent rhetorical points.
That feels like a conclusion from flawed first principles. Google is in the game of getting maximum reward for minimum effort/cost, not innovating for innovation's sake. If the cheapest path to that reward is innovation, they'll do that (& they spend many $$ on that because of history lessons - don't get me wrong on that point). However, if the cheapest path is an alternative decision, they'll do that instead (e.g. spending many $$ paying people to work on tax loopholes). Your position ignores the fact that being the 1T gorilla in the room gives them significant advantages that, to them, are worth the disadvantage of slower innovation. Business is all about tradeoffs after all. For example, as the 1T gorilla in the room underpinning a good chunk of the US economy & employing many people, you can:
* Throw your weight around more easily with regulatory bodies & lobby for laws/government business that you would have a harder time with as a smaller company * You have a stronger reputation as being a "reliable" business ("nobody got fired for buying IBM") * You can instill regulatory roadblocks for strategic industries you wish to defend * You can tackle ever bigger problems that only exist because you're trying to grow the company ever bigger (but that don't necessarily provide societal value) * You can stifle competition through acquisitions (Instagram, DoubleClick, etc) * You can slurp up all available talent that could pose a threat * You can engage in collusion with competitors to retain a market leadership position * You can engage in collusion to control the employee market wages (see Google, Apple & others colluding to suppress Web Browser developer salaries & other cases that have reappeared as the fines these companies have paid have not outweighed the benefits of engaging in these tactics).
etc etc. These examples aren't hypothetical. They're easily observed by just passively observing business history over your own lifetime or reading up on it a little. For example, look at the widespread wage suppression and non-competes tech companies have engaged in over the years.
> Again, this does not need empirical evidence (even though there is plenty) but mostly basics of classical economics.
I challenge this statement. To me your theory and my empirical evidence are in direct conflict. The scientific method encourages you to question your theory or to back it up with concrete evidence that the world works the way you are modelling it. Your argument is kind of like the high school physics problem of "let's ignore friction and wind resistance" except worse because to me it's more akin to arguing for Aristotle's bodies of motion when you already have a lot of experimental evidence showing the model is incomplete/wrong.
> I tried to read book excerpts from the Stanford professors and it appeared to be complete BS vending. Arguments like "Big Tech is too big" ,"Big tech has too much power" etc. simply does not make sensible principles as much as they make excellent rhetorical points.
I agree with you there but I think you're not presenting the arguments in the best light (independent of whether or not these profs make them). Typically the stronger versions of the arguments go something like: * Big tech is acquiring companies before they can compete. This both hurts innovation & the US economy in the long term because these big tech companies are extremely efficient at their own growth, but that's not necessarily yielding the broader benefits needed by the economy for productivity/innovation. * Big tech media companies aren't responsible stewards for their platforms. They've grabbed the highly lucrative side of the business (ads) while pushing costs as negative externalities (misinformation, manipulate & malicious ads, scams, etc) through intentionally or incompetent policies and practices. Indeed, maybe less consolidation in the media space across the board would be better (not just in the Internet) so that issues are more diffuse rather than being hyper concentrated. * Big tech companies are colluding with each other in ways that are counterproductive to the industry and the broader economy, indirectly raising costs through their domination of the industries they compete in. There are numerous instances of such collusion and in the past we've punished other companies trying to do similar things in other industries.
I think "too big" or "too much power" is used as short-hand for these (& many other) specific problems that can't be fully enumerated in interviews. If the professors fail to make a more cogent argument in the books as you've claimed, then that's a valid critique, but I wouldn't base that assumption just from excerpts.
> Your position ignores the fact that being the 1T gorilla in the room gives them significant advantages that, to them, are worth the disadvantage of slower innovation.
You are absolutely right. You are talking about regulatory capture here and larger companies are in better position for regulatory capture. What you are proposing however is more regulation to prevent this very situation. Very likely this will lead to even more regulatory capture. We have seen this with health insurance, utilities and telecom companies where government steps in to prevent these companies from making too much of profit but then grants them monopolies causing stagnation of innovation. The correct solution IMO here is that less power in the hands of government so there are less avenues for regulatory capture.
But let us look at the facts. Is Apple, Google or Facebook today too large because of regulatory capture or innovation? Compared to any industry I think tech in USA is growing due to the latter. (Exceptions like Visa, Mastercard, Intuit exist, and I do believe government has failed to be a custodian of markets here).
> Your argument is kind of like the high school physics problem of "let's ignore friction and wind resistance" except worse because to me it's more akin to arguing for Aristotle's bodies of motion when you already have a lot of experimental evidence showing the model is incomplete/wrong.
Fair enough. After all, all theory must be put to test. We can easily see my logic in practice by taking example of utility companies like PE&E, Oil and Gas industry, health insurance industry or almost every industry in Europe which is heavily regulated to prevent them from making "too much profit" or is restricted in terms of how much they can grow. Contrast nearly any industry on planet earth today with USA's tech industry and USA's tech industry comes on top in terms of wealth creation, better workplace standards, innovation and in some cases mind boggling ambitious companies like SpaceX. Not only American tech companies are engines of growth and some of the best places to work for, they are also cheap for their consumers and massive force for good at global scale. (This is not to say they don't have any dark side but when you put all numbers together they come up on top.)
What is not broken should not be fixed.
> * Big tech is acquiring companies before they can compete. This both hurts innovation & the US economy in the long term because these big tech companies are extremely efficient at their own growth, but that's not necessarily yielding the broader benefits needed by the economy for productivity/innovation
Yes, that is a good principle that these people can talk about rather than "too big" argument. Once these principles are clear, I think it is much easier to show them as well known economic fallacies.
> I think "too big" or "too much power" is used as short-hand for these
Yes, and that is precisely what I find dangerous. We must be debating those very principles and not the question of whether we need more regulation or not.
For example on this very HN thread hoe many people are debating those precise principles ? Almost none.
Isn't this offset somewhat by increased competition forcing those operations to be as efficient as practical?
Large, vertically integrated companies can afford to comply with more stringent regulation.
There are many good examples here. First one that comes to mind is the auto dealers. Telsa without dealers offers far better experience where as car dealers are normally despised. Having more car dealers have not really improved dealership experience for us. Resources wise it is extremely wasteful.
> Large, vertically integrated companies can afford to comply with more stringent regulation.
While this is true, they do not tend to get large overnight. It is a slow process and a learning curve for many. With higher regulations you end up creating significant friction at certain key levels. Some companies that would have gotten large wont grow large if you apply higher regulation to some companies.
No, in a free market, it's winner take all in the race to become too big to fail, regulate, or compete against, so the mega-corp(s) can sit back and collect rents. We don't want free markets, we want competitive markets.
Car dealerships aren't a good example because they're often franchises. That's like saying it's "competition" if McDonalds Corporate ship fries directly to consumers, bypassing its own (franchised) McDonalds locations. Competition is between McDonalds, Burger King, and your local hamburger stand.
The only way I've ever read of/seen for maintaining something approximating a free market is significant government regulation and intervention.
Significant regulation of large economic entities is probably a very good thing for the overall economy in the long term.
The subsidiary fragmentation itself is uninteresting from a regulatory perspective as long as the regulatory environment is carefully crafted to pierce through intentional corporate governance obfuscation.
I can't remember people being denied service for their political views until just recently.
Another odd thing is that a company would give a rip about their employees political views.
These are strange times.
The idea of not discriminating based on politics or other overt characteristics, as a widespread cultural phenomenon, was relatively new and short-lived.
Edit: the above is perhaps a bit of an incomplete thought. I'd generally argue that picking wartime era hysteria is a little misleading, that happens to basically every country. On the other hand, there's a reason why people always jump on the "myth" of the First Amendment when it's mentioned in discussions about private companies: Americans often confuse the constitutional principle with the general principle, and they do this because Americans are largely supportive of the general principle.
My point is that, as far as I know, the kind of freedoms and political neutrality the West enjoyed between 1980s to 2010s were a short-lived anomaly. Implied in that was an observation that we're on a downslope now and I agree with 'OneEyedRobot that we're not enjoying much of that neutrality today.
In the early 1970s? I'd like to see some citations of that.
Probably wasn't as noticable when they were turning far more people away because of their race, religion, or gender.
For common carriers that has been a longstanding policy, for example, telephone companies can't deny you service for things you say over the phone.
The fewer the viable alternatives to a service are, the more that makes sense.
The big shift is the exact opposite, that you would allow companies that run services with network effects to choose what messages are sent over their services while at the same time avoiding liability for the messages that are sent.
If you try to say revolution is a political viewpoint, I guess you could make that point, but it seems entirely reasonable that asking for select people to be killed is crossing a meaningful line. Especially in a democracy.
There is enough legal and philosophical body of work around that question. But either ways it is not a meaningful reason for "regulation". All platforms already have restrictions for such things and very likely government regulations will force platforms to allow for "kill all X" talks but not "kill all Y" talks making the platform very toxic.
For example, American company's response to Jan 6 appears to be far more mature and useful to American society than that of American federal government and FBI.
The whole situation is crying out for some sort of civic process. If we live part of our lives in cyberspace, we need a voice in how those spaces are governed. Something “By the people, for the people.” We don’t let corporations run towns in the real world and do this sort of thing. It’s entirely consistent to want to bring democratic values online too.
We already have a framework for physical networking services that does this: Common Carrier. It doesn’t and shouldn’t apply to companies other than “dumb pipe” utilities. No company which can or does do any moderation of content should be governed by CC laws.
You seem to think forcing all companies on the internet to maximize democratic values is somehow either possible or a good idea. I can’t see how that could be implemented without destroying every company’s ability to build a brand, to enforce their own contractual terms.
Example: you local church/temple website that accepts with user-generated comments would be forced to allow and keep every comment submitted, no matter how offensive to the members or the faith the church practices. Non-believers would be able to deface the service in a way that trespassing laws prevent IRL. This compelled democratic values maximization concept doesn’t exist in the offline world so you would have to build the logical/rational foundations for doing it from scratch. Obviously the church/temple example is contrived and those are far from Big Tech companies, but if you assert that some services are required to retain all offensive content and others aren’t, then you need to develop a litmus test, which I argue is futile.
And from where I stand, demanding a civic process is really strange considering most civic processes in the USA are deteriorating while Big Tech is doing the opposite.
As an example, in Germany nudity is much less of a taboo than in the US. Facebook has a blanket ban on nudity for American cultural reasons. A nudist event in Berlin can be argued for and against in Berlin by the communities involved. They can talk about it in council, protest, write letters, vote or get politically involved. On Facebook people can’t effectively do any of those things. They just enforce America’s squeamishness from on high. If Facebook and Twitter want to be an online version of the public square, people need the right to govern our own communities online as we do offline.
I like Reddit’s moves in this direction with community moderation of subreddits. I’m not a free speech absolutist - I like that entire subreddits can still be banned if they become too toxic (as we saw recently). But this makes people really angry because Reddit’s internal governance is entirely unaccountable. An unelected monarch is ruling over a bunch of small community run spaces. The next healthy step is for the monarch to be replaced with a board of directors or something - who are themselves accountable to the users. But like all kings throughout time, reddit has no incentive to do that.
And on Facebook, Twitter and YouTube it’s much worse. At scale, the editorial choices made by their content algorithms are inordinately powerful, invisible and actively destructive to society. I don’t know the right way to reign in this stuff. But I’m worried that if we don’t, society at large won’t survive the partisanship, extremism and conspiracy thinking that these editorial algorithms en masse create.
I agree there’s something deeply wrong in Washington. I wish us technologists were making things better instead of making things worse.
A local website with a few hundred users shouldn't be held to the same standards as one that has made itself a dominant factor in public discourse.
I don’t agree that FB should be a content neutral platform. Just because there are network effects and social media is common do not mean they are “the” town square. They are more like a private mall where lots of people are allowed, but some people have been banned for trespassing.
https://en.wikipedia.org/wiki/Rob_Reich
https://profiles.stanford.edu/jeremy-weinstein
https://en.wikipedia.org/wiki/Mehran_Sahami
Contrast this with what Senator McGovern wrote after he retired to run a small hotel business:
https://www.wsj.com/articles/SB10001424052970203406404578070...
"In retrospect, I wish I had known more about the hazards and difficulties of such a business, especially during a recession of the kind that hit New England just as I was acquiring the inn's 43-year leasehold. I also wish that during the years I was in public office, I had had this firsthand experience about the difficulties business people face every day. That knowledge would have made me a better U.S. senator and a more understanding presidential contender. Today we are much closer to a general acknowledgment that government must encourage business to expand and grow. Bill Clinton, Paul Tsongas, Bob Kerrey and others have, I believe, changed the debate of our party. We intuitively know that to create job opportunities we need entrepreneurs who will risk their capital against an expected payoff. Too often, however, public policy does not consider whether we are choking off those opportunities."
I do think something must be done, but the notion that it should be the government taking action scares me. I believe, platonically that is the right way for government to act, to take necessary actions to counter predatory behavior.
What scares me is that we chanced into this. Zuck made some brilliant movies but he was the same guy at 22 that thought that younger people were smarter. Notice he wasn't young enough to say that at 32. The people in government do not understand tech. They don't understand how it works, they don't understand where it is going, and they don't understand its ethos. Government right now is dominated by ideologues and demagogues. I'm very worried that any government action is going to end up being something that only FB, Google, etc can do anything about.
Why does the government need to act? Why spend so much time complaining about it on HN? We've been doing that for a decade.
Effectively if you piss off one of these companies, they can exclude you from entire swathes of tech, media or the economy. In such situations many people will choose to self censor. Personally I don’t believe that is healthy; it leads to situations like the US 2016 election when all the polls were well in favor of one candidate and then the other won.
Another thing about corporate power in general is that it is not the case that the corporation uses its power proportionately representing either its employees or shareholders. The corporation uses its power in accordance with its officers’ desires. In this way corporations are the ultimate anti-democratic institutions as they concentrate the economic power generated by many diverse people and centralize it to one or a few people. Hence Bezos, Zuckerberg, etc.
And the combination of skill, hard work and luck that led to their leadership of these corporations doesn’t in any way relate to their values or judgment and the ways that they will use their power.
I also don't have any idea what that has to do with Big Tech. Big Tech did and continues to discriminate against right-wing/alt-right views* which you say leads to self-censorship, and that somehow led to an electoral outcome where the right-wing/alt-right candidate won; So Big Tech's corporate power is so immense & influential that it led to...the opposite outcome?
I'm sure there's an argument there somewhere about the dangers of excessive corporate power but I can't make sense of your connection between Big Tech corporate power and your example of the 2016 election of the most anti-Big-Tech US president ever.
* I'm assuming you're referring to right-wing/alt-right discrimination because I don't recall many significant examples in which Big Tech excluded people/partners from "entire swathes of tech, media or the economy" for left-wing views, but feel free to correct me!
And your Big Tech self-censorship narrative has no discernable substance. Unusual levels of self-censorship would be reflected in unusual levels of polling error, but the actual polling error wasn't unusually high in 2016 [1].
Election polls aren't even conducted by Big Tech AFAIK and they don't have access to personally identifiable poll responses. Pretty hand-wavy and baseless to suggest that people were somehow afraid of Big Tech punishment based on anonymous third party polling. The Big Tech censorship issue centers almost entirely on user content submitted voluntarily on Big Tech platforms.
[1] https://fivethirtyeight.com/features/the-polls-werent-great-...
Cartel behavior does not involve monopolies. Vertical integration of supply chains does not require a monopoly to be a problem. Markets are defined in a way that makes it look like on paper the leader doesn't have a dominant position simply because they don't have 100% market share, but it hides that market share doesn't tell the whole story. Facebook should have never been allowed to buy their most promising competitors in Instagram and WhatsApp, for example.
Intentionally confusing the market for phones with App stores is an example of a problem that 20th century antitrust never had to deal with. Just because two products look the same doesn't mean they compete with each other if there are other reasons, such as identity politics or vendor lock-in, which cause people to not treat them as replacements for each other.
All these things are issues which government should legitimately be regulating, and at one point did regulate. But there is no will left to do that anymore, and the poor quality of our politicians is partly to blame for it.
I'd say that the issues with app stores are a pretty direct parallel to Company Towns[0], whose issues have been known and addressed for more than a century.
>[Democracy] doesn't mean that there's some alternative way to accomplish that underlying task
>[Democracy is] about a fair process for refereeing in an ongoing way, the contestation of citizens’ own choices and preferences.
Like they said, would you prefer the Pope to censor your tech? Would you wear unregulated clothing? Of course not. "[W]e need to break people out of a kind of binary mindset where markets are good and markets have all the solutions" and people's needs must be decided democratically.
So this vicious circle is hard to break if we aren’t will to break with old ways.
But they do, over time. This is what competition does. You may start out in a fresh market fully intent on doing a good and honest job, but then competitors arrive. First, they force you to innovate and improve your offering. Then, when improvements start yielding diminishing returns, you're forced to either become exploitative and dishonest, or be outcompeted.
That may have been true 20 years ago. The people in government understand current tech and its ethos very well, because it's the same ethos as in every other industry: profit over all.
What surprises me is to see people still thinking our industry has any trace of the original "hacker ethos" left in it. It doesn't. Big Tech companies are just run-of-the-mill megacorporations, with all pathologies that come with megacorps run by suits, differing only by the flavor of the products they sell. Startup world is just another Wall Street, with a sprinkling of tech sugar on top of the dollars being gambled.
Governments understand this well. Whether it's Google or GM or Monsanto that's pulling business shenanigans, it's all the same business shenanigans. Hacker ethos? That ended when programming started to pay well.
How do you think anyone reigned in the tobacco industry? Everyone in the industry just went, "well I hate coffers full of money, let's stop working here" ?
That's what governmental authority exists for. The idea that the foxes polices themselves while they have all the keys to the henhouse is comical.
Exactly. "The tragedy of the commons" is a well-known terms that means, "given the keys to the henhouse, the foxes are not able to police themselves well enough to keep themselves from starving" - and this doesn't even begin to consider the interests of hens.
What is tech's ethos?
Worse is the likely objective/outcome: big tech and gov't. joining forces.
A real question is, in areas where network effects are very strong, how do we avoid monopolies forming?
Almost like saying: how do we keep gravity from eventually creating a few giant blackholes?
Granted, we can't change physical laws. But if we could, we would clearly need some sort of anti-gravity that kicks in just passed the point of a black hole.
So I guess there needs to be some benefit of not being a monopoly.
And all that Congress and rest of the bunnies "activities" is just foam and smoke.
Why ? Becouse US as a whole makes money from current state. Corporations are just part of US controlling rest of the world.
Not so evil so far and always will be someone on the top. Except maybe South America think USA is that bad capitalist from XVIII+ century...
Still, USA have a lot of influence via tech and tech is used by plain peoples. So when something is bad in USA it have influence on rest of us. Eg. attempts to deny repairing phones... Pure idiocy but who cares in Congress or elsewhere ??
And, sadly, you - USA - will do not change much until forced by someone proposing (and having some power to enforce) some sanity in tech and other parts of life.
It's not easy to be World capital country :)
as seen from past antitrust actions, I expect much of "big tech" to be thoroughly dismantled.
alphabet will be chopped up along business lines. Each business will have to develop its own independent ad system.
facebook will be forced to spin off their acqs and separate completely from them.
amazon and AWS will be split, obviously.
dunno about apple. at least split their entertainment businesses from their hardware businesses.
This will be a great opportunity for startups who might otherwise never would have a chance.
In a sense I'd say that they are now a branch of government.
> the government really doesn't like a challenge to its power.
But the government as a whole hasn't really made up its mind. There are a few junior congresspeople who are vocal about monopoly problems with Big Tech. But senior Democrat congresspeople from Big Tech states have not uttered a word against Big Tech. Even senior Republicans, and republicans have been at the receiving end of Big Tech censorship, are loath to antitrust as they have started conflating checks on Big Tech monopoly with checks on free enterprise. In short, government will has not coalesced around antitrust action against Big Tech.
I don't know if Big Tech can be stopped by the government. A possibility I see is that the stock market bubble pops, Fed money printing loses its ability to inflate tech stocks, leading to a decline in Big Tech power.
It's probably better to view the 'government' as a coalition of self-interested groups. It's never really monolithic.