He makes some valid points, but what doesn't quite add up is that not all of these "monopolies" are built on network effects. Google search, for instance, doesn't benefit from network effects. Gmail doesn't benefit from network effects, nor does anything else that Google does (successfully). Of course Google benefits from its size in many ways but that's not the same kind of self feeding dynamic as a network effect.
Amazon became big by selling books. Did publishers really say, hey, let's sell to the biggest online book store only? Would it be any more inconvenient for buyers to buy from a different online book store just because it was not the biggest? I don't think so. There is no network effect, just plain old economies of scale and the power of brands.
Myspace was the biggest. Friendster was the biggest. Both in areas that clearly do benefit massively from network effects.
So, I agree with the author on some points, but I think he plays down the "being better" part a little too much and he ignores substitution effects.
Did we read the same article? The author definitely payed much less attention to "network effects" than most articles on this subject usually do, explaining that it is quite common for monopolies to be created for new technologies, as well as giving several other reasons, some of which you reiterated.
It appears to me that he explains the emergence of these monopolies almost exclusively with network effects:
"Internet industries develop pretty much like any other industry that depends on a network: A single firm can dominate the market if the product becomes more valuable to each user as the number of users rises. Such networks have a natural tendency to grow, and that growth leads to dominance."
and
"Still, in a land where at least two mega-colas and two brands of diaper can duke it out indefinitely, why are there so many single-firm information markets? The explanation would seem to lie in the famous American preference for convenience. With networks, size brings convenience."
Did you find any other explanation in the article?
I think this is the meaty quote: "Apart from brief periods of openness created by new inventions or antitrust breakups, every medium, starting with the telegraph, has eventually proved to be a case study in monopoly. In fact, many of those firms are still around, if not quite as powerful as they once were, including AT&T, Paramount and NBC."
Also, your latter quote seems to me to be about size & convenience, not networks. He's using "network" in the sense "on the internet", not in the sense of "network effects". My basic point is that the article has a lot of good things to say, and says relatively little about network effects. So if people read your comment first, they're not going to read the article because they can think "oh yeah network effects, been there done that"
I do agree that the article has other things to say as well. It's not a bad article. But I'm pretty sure that the second quote is supposed to mean that network effects cause larger online services to be more convenient for users. I don't agree with that in general. It's true only in some specific cases where integration of several services makes things more seamless. Your quote doesn't give a reason _why_ monopolies develop. It just states _that_ they do. Also, there is nothing about how they can unravel pretty quickly.
I didn't really agree with this idea initially, but the argument was persuasive. Certainly I don't think any of them have the kind of monopoly that AT&T had, but Facebook might be close as they continue sinking their Facebook Connect hooks in.
The trouble is that monopolies are not good or bad per se. Google and Facebook are still mostly innovating and adding value. At some point the landscape starts to change, and that's when they get scared and start getting anti-competitive and the balances start to tip.
The tricky thing is determining when this is. Certainly the government is ill-equipped to figure it out, because it's all sound bites and popular opinion by an apathetic voter base. So the rationale for why a monopoly should be broken up gets sanded down to a slogan about monopolies being bad, and pretty soon politicians are jumping on the monopoly-busting bandwagon without any kind of legitimate rationale.
This is why, I really value the old-school intellectual movement that Chomsky represents of standing outside and speaking truth to power. Even if a lot of his analysis is flawed and ideas ultimately wrong, the honesty with which he pursues them is admirable and we need many more like him to keep America honest. The rich won't do it (as a group they are too fearful), the powerful won't do it, and government is certainly too awash in lobbyists and bureaucracy to have any hope.
Remember that the great monopoly to be broken up by anti-trust law was in fact a government-mandated one: AT&T. Without the government banning competition, AT&T would have been very much different in 1974 (the time of the divestiture).
It became illegal because competitor operators were a threat to AT&T. The government made a deal with them by giving them legal protection, getting some political control in return. Even after the divestiture, the Baby Bells had preferential treatment and received automatic mobile licenses when the technology appeared.
Secondly, it's kind of silly to think about web monopolies when users can freely change operators. Google, Facebook etc have to work hard to maintain their users.
A natural monopoly doesn't need legal protections, it will continue to exist regardless. The very reason competition was banned was because it was possible.
This system is "tolerated," simply because there is nothing wrong with rightfully earned monopolies. Monopolies turn bad the moment they gain special treatment from the government. That's where things get ugly.
There's nothing stopping another search engine from springing up (and they do, of course). There's nothing stopping search engine XYZ from topping Google except for the aggregate of our voluntary decisions. This is not a bad monopoly.
Monopolies can do a great deal of anticompetitive things without the help of the government. In the past antisocial pricing that was the primary concern. While these firms are different in that they do not charge users directly, their revenues are directly related to the size of their user base and other familiar tactics can be employed: For example, using their user-base to squeeze competitors in vertical industries, and employing defensive tactics that ensure direct competitors can never gain traction (see contacts spat).
If the monopoly is in an industry where the barrier of entry is very high, then yes, they could raise prices to antisocial levels and make exorbitant profits; However if the monopoly was in an industry where the barrier of entry is very low, the monopoly doesn't have that much power to raise prices to antisocial levels; If they do other competitors will jump in. Imagine, through low prices and effective logistics you've achieved a monopoly in producing wheat. You have a monopoly because no one can afford to sell as low as you do, not even the poor developing country farmer; Now imagine that you got arrogant and raised your prices by 350%, you'd start to lose the monopoly because other competitors can now enter the market. That's why you wouldn't raise prices even if you have a monopoly, if your industry has a low barrier of entry. No matter what you cannot employ defensive tactics to ensure your competitors (other farmers) can never gain traction, if your prices are too high.
Now, whether search or social networks are high barrier of entry industries, is another question. (I tend to think so)
Let's say in a town of 200 people you open the first fruit shop. Turns out people there love fruits but just never knew it. Everyday they only come to your shop to buy fruits. That's a monopoly already. However, you'd only stay a monopoly if your prices stay competitive. If your prices are too expensive and your margins are too high, other opportunists will jump in to open their own fruit shop, and there's nothing you can do to stop them. You can try to drop your prices, but that's it, and now you're worse off, your prices are lower and there's a competitor. You'd rather have prices to be low all the time and have all the business to yourself.
I think that your example does get a point across, however there are a couple of problems with it:
1. I was hoping for a real world example (if its possible then there should be many examples, right?). 2. You imply that the fruit shop owner has access to the most efficient supply chain of all the other potential, would be fruit merchants and thus the only way he can loose his primate is if he raises the prices too much. How about people who start competing for various reasons like: I don't like the guy, or I don't have a better idea for business, etc. 3. Defining monopoly too locally has its own problems: Grocery store has a monopoly over selling groceries on a block, if there are two groceries, we can say that each has a monopoly over their part of the block, etc. See that's not really a monopoly.
The point is that in free, easy to enter markets monopolies are impossible - by the virtue of free and easy to enter market. What monopoly really is - domination of a single entity over a regulated (possibly by the monopoly itself) and expensive to enter market. Examples: http://en.wikipedia.org/wiki/Monopoly#Examples_of_legal_.28a...
By your definition, Amazon is not really a monopoly either, its got to compete with all these brick and mortar book stores...
My definition: a firm is a monopoly in a market when it is the only firm trading in that market.
The grocery shops on two blocks: not really, their markets overlap.
To look for real examples, pick any country town with 1 petrol station, or 1 book store, or 1 supermarket, or 1 fruit shop. You're never going to get these mentioned in a Notable Monopolies list on wikipedia.
Amazon is waaaay from monopoly. It competes with eBay, specialty web stores,...
And also - entering the field of WEB commerce is not a hard to get into market per say. Getting into oil market or telco marker, even getting Wal-Mart out of the way - now these are some hard undertakings.
In a town with 1 fruit shop - you can always go to nearby town if you don't like your fruit merchant (out of spite for example). But when you arrive to these nearby towns - and everywhere there is basically the same fruit retailer, with the same problem that bothered you with the first one. That's when you encountered a true monopoly.
I'm from ex-socialist republic and mostly everything was monopolized (many things still are) by the government or government owned enterprises. If you're from US, you wouldn't have seen stuff like that since AFAIK you guys are really good at breaking stuff like that apart.
So while I admit that your reasoning is solid, I view the term "monopoly" as inherently bad, since lets face it most (even all?) businesses start squeezing their customers when they get the chance.
Seriously, drive 100km to the next town to buy fruit when you have a reasonably priced fruit shop next door? :)
I'm from Australia, similar i suppose. Monopolies are generally bad, but as I said before, monopolies that result because a firm is super efficient and keeps prices low, and remains that way because it is afraid of potential competitors, well that monopoly is hard-earned. Although these so called monopolies are generally small. (being the only firm in a niche market).
"there is nothing wrong with rightfully earned monopolies."
Market dynamics can create an anti-competitive, bad-for-the-economy monopoly without gov't help. Your assertion assumes that there exists no economic barrier so large that it is irrational for a potential competitor to try to overcome it.
In fact, there are easy counter-examples: imagine a town's public water source dries up and the only well is owned by a purely rational/greedy party.
Did you read "In search of stupidity"? It's a very interesting and amusing book, and it makes a point that Microsoft monopoly owes more to competitors' ineptitude (Borkand, Lotus, Novell, Digital Research, etc) than to Microsoft genius per se. Therefore Microsoft monopoly may be "rightfully earned" in some sense, as bad as it is - though it carried some positive effects too.
Actually, I'd see Amazon, Google and Facebook as three somewhat different types of monopolies.
Google's monopoly comes economies of scale and superior technology. No one is locked into the Google, you just prefer Google because it is superior.
Facebook's monopoly is network effect.
Amazon is something like traditional horizontal monopoly.
The only thing the companies have in common is that they're on the Internet. But the main conclusion is that Internet is where all future market dramas will be played out.
Google search is starting to suck a lot though. Now half the time I search for something I end up with garbage and give up. Tried searching up the name of a virus .exe and got a spam page instead of anything relevant as the first result, with more spam pages for the rest. This is not a one time or occasional occurrence anymore.
Their other commonality is that they are "new markets". As the article indicated, monopolies do tend to spring up in new markets. In new markets, what to do to be successful is not immediately obvious, so the difference between competitors is much greater.
As markets mature, the differentiation between competitors lessens. Does that mean that the competition will increase? In the case of Amazon or Google, it should. However, it's possible that Amazon or Google will gain governmental barriers to competitors (they're big enough to buy congressmen), or quasi-governmental barriers (a-la Paypal) or strong arm barriers (a-la Standard Oil). Let's hope that they don't gain those barriers.
"all online sales" is not really relevant. Nobody is anywhere close to monopolizing all sales on the internet. Saying Amazon is not a monopoly because they don't dominate all sales is like saying PacBell wasn't a monopoly because they didn't control all forms of interpersonal communication. Instead, look at their chunk of sales of their core items, such as books.
His conclusion is a little odd, but all in all an excellent summation of where we are.
I think we are fortunate that we have Google as one of the monopolies, as at the moment they seem to be willing to be vaguely non-evil, but have already drifted significantly from 'don't be evil'. As much as I like to rail against them, it could be worse.
Oh? How many other social networking sites do your mother or cousin or friends spend their time on? If you blocked Myspace and Facebook on your corporate network, which one would people pitch a fit about losing?
The very fact that people wouldn't pitch a fit about losing access to Myspace is a perfect example of why these network-effect-based "monopoly" claims are bullshit.
If they can be that dominating in 5 years, there's no reason they can't be eclipsed by another company in the next 5 years. Myspace is an apt comparison.
Google is the only one I use constantly. I've been on Amazon this out of curiosity to see what people said about a book I disliked, but haven't bought anything from them in years. Facebook? No. Twitter? No.
I wouldn't call it bullshit. The piece focused on purely online information networks so I really don't think there's anything devious or subversive going on here as your comment implies.
And the tool the new monopolists are using? The browser.
Somebody needs to redesign the browser so that it becomes vendor agnostic. That is, if I'm doing a search, I shouldn't see a brand or care who is providing me the results. All I care about is the quality of the results. Same goes for reading articles and buying things. Take the branding out of the web, and you'll kill the monopolies. (Yes, it would wreak havoc on the entire internet business model, but I have the magic wand, and I'm prepared to use it)
Is there some reason we have to have brands? Can't all the things we do and places we go be put into a configuration file somewhere and managed automatically?
Your reply looks correct but incomplete. Yes, the immediate dominance has to do with networking effects. But the network effects are all based on the branding, and the browser delivers the branding along with the content. Without the branding the network effects are negligible.
People use Google because it says "Google" on the page. People go to Facebook because their friends are there. None of the results they want -- search results or chatting-updating friends, has anything to do with the underlying companies involved, although it has everything to do with the network effect, as you point out.
To demonstrate, let's suppose all of my 300 friends all used different social networking software. If somebody kept a configuration file that allowed my new browser to access their information in real time and assimilate it for me in one corporate-free spot, I could still have the networking experience of being with my friends -- say picking out a new movie to rent -- without the branding having to be part of the networking effect.
It's a trivial Greasemonkey script to suppress all the chrome on a search result page. If you really care that much, you or anyone else could whip one up and share it with the world.
I suspect very few people would use it, because I suspect that many people use a search engine because of the branding. It makes them feel comfortable and safe; they know that this search engine has given them good results in the past, and this makes them believe that this search engine will give them better results in the future.
In other words, this problem exists between keyboard and chair. It's not a technological problem: the technology exists right now to do exactly what you suggest. It's a psychological problem: people naturally trend toward the familiar, and that makes the familiar even more familiar.
But this was the exact same situation users were in before the very first search engines: there were many directory sites with lists of places to go. Each site had a brand and users became attached to that brand.
Generic search engines were successful because they abstracted the list-based directories into generic searches. Users were more than willing to replace the many brands they used with a single brand that offered meta functionality.
I see no reason to think that we are finished abstracting. Some new service should come along and abstract away more huge parts of the internet and put it under one brand name. Users will still have their brand-loyalty, it'll just be to a conceptually larger service.
Or is there something about the current state of affairs that means it should remain fixed?
I think that there's room for us to abstract away branded search under a new category, but I don't think new browsers are the answer.
People rely on brands when they have a product whose quality is important to them, but is difficult to discern from casual inspection. So for example, the strongest brands tend to be in markets like soda, foods, and cigarettes, because you're putting this stuff in your body and yet you can't tell what went into it. Branded medicine can fetch prices several times higher than generics, despite being exactly the same stuff. Things like software - Microsoft and Google - have middle-ground brand power, where quality is important (but not all-important) and you have few indications for overall software quality at purchase time. Things like airlines have virtually no branding power, because the trip is over in 5 hours, you've gotten to your destination, and you quickly forget about just how unpleasant the flight over was.
Following this, there are two ways to cut brand power of entrenched competitor:
1.) Make consumers care less about product quality. Cunard and White Star were done in by the airline industry, because by cutting trip duration from 5 days to 5 hours, they made consumers ambivalent about trip quality.
2.) Give consumers better information about the actual quality of the product. Google destroyed branded directory sites like Yahoo because they showed everyone just how many websites the directories were missing, and let them evaluate for themselves the quality of the sites.
To achieve #2, you could build some software - either a browser plugin or a meta-search engine - that "rates" each result on an unbranded results page, saying "This is filled with adsense" or "this information is false" or "this was copied from here". If you could convince consumers to trust the ratings more than the original search, you'd break the brand power of the original search engine. However, that's probably at least as hard a problem, if not harder, than building the search engine itself.
FaceBook and other social networks become dominant because of network effects, not because of brands. They don't let you export your friends, so you become trapped there, without much value anywhere else. FriendFeed tried to break that cycle with their meta-social-network, but it didn't work, probably because they didn't add any value besides what the social network itself had. In cases where new software really is superior, I think it's possible to convince whole cliques to move over at once, though, and that can break the back of the entrenched competitor. I saw it happen with LiveJournal, and then again with MySpace.
If everything was totally unbranded, just feeds of data, how would providers generate revenue? The obvious route would be by distorting the info in exchange for cash.
If Google couldn't show their ads on search result pages, they'd need to charge advertisers to prioritise certain search results in their feed.
If Facebook couldn't show ads on the sides of their pages, they'd have to start embedding marketing in your friends' actions - by emphasizing certain behaviours over others - for a fee.
I'm sure a bit of subtle manipulation already goes on, but I'll take branding and advertising any day rather than make that the ONLY option.
I only do about five things on the internet, although I "visit" many sites. Let's say those things are 1) read articles, 2) chat with friends, 3) search for media to consume, 4) get advice, and 5) tell people things I think are important.
So I make a black box with five buttons. Each button does one of the five functions in a completely brand-free manner. (Of course the box itself would have a brand)
Now I am doing all of the things I did before, and all without the concept of an internet "location" or the idea of a brand being associated with my natural activities. The system itself could keep track of available resources and adapt itself to translating those into the things I want.
Kind of like Ad-block on steroids.
This is all easily doable with current technology, btw. No magic required.
Amazon became big by selling books. Did publishers really say, hey, let's sell to the biggest online book store only? Would it be any more inconvenient for buyers to buy from a different online book store just because it was not the biggest? I don't think so. There is no network effect, just plain old economies of scale and the power of brands.
Myspace was the biggest. Friendster was the biggest. Both in areas that clearly do benefit massively from network effects.
So, I agree with the author on some points, but I think he plays down the "being better" part a little too much and he ignores substitution effects.