Adam’s recent post on Free Press’s hysteria over media consolidation reminds me of the left’s general tendency to move the goalposts when it comes to market concentration in communications markets. Over the last quarter century, we’ve gone from a world in which there were honest-to-goodness monopolies in the telephone and cable markets to a “duopoly” where the former monopolies invaded one another’s turf, to a world of much greater competition as mobile companies entered the telephone market and satellite companies entered the video market. Yet as I noted last year, Free Press chairman Tim Wu characterizes the wireless market—with its four national carriers and several regional ones—as a “textbook oligopoly.” Indeed, one often gets the impression that the arguments of pro-regulation scholars are the same as those they would have made 20 years ago with “monopoly” replaced by “oligopoly.”
Now, I’m sympathetic to the argument that a “duopoly” is insufficient competition, and that regulators should at least take a close look at the behavior of firms that comprise one half of a duopoly. I think libertarians’ tendency to laud the broadband marketplace as a free-market nirvana is a bit misguided. However, I find the language of “oligopoly” much harder to swallow. While more competition is better, there are plenty of industries with 4-6 players that few people regard as problematic. The wireless business is extremely capital-intensive, so it’s not that surprising that there are relatively few restaurant chains.
But the tendency to shift from “monopoly” to “duopoly” to “oligopoly” while deploying essentially the same arguments does make one wonder if there’s any amount of competition that the good people at Free Press regard as sufficient. And it seems that Adam has found the answer. Having 55 major players in a market is the very definition of cutthroat competition. The notion that 55 firms can constitute a “bottleneck” that significantly curtail the flow of information to consumers is just silly. And of course the 55 figure is totally arbitrary. The media are a long tail business, and they could have included a lot more firms if they hadn’t set their cutoff at $100 million in revenues. For example, their chart misses Hubbard Broadcasting, which owns around a dozen broadcasting stations concentrated in the upper midwest. So if you’re a Twin Cities resident who doesn’t like what Clear Channel, News Corp and the rest are producing, you can tune in to KSTP’s TV station and talk radio station for a perspective that’s not controlled by the “Big 55.”
Ultimately, I think the moral of the story is that for some advocates of media regulation, it really has nothing to do with competition. Whether there are 1, 2, 4, 8, or 55 competitors, they continue to believe that there’s too little government regulation of communication. When the number is small, it makes a convenient talking point, but they go right on making the same arguments when the number of competitors gets ridiculously large.
I’m more sympathetic to EFF-style voluntary collective licensing than Mike Masnick is, but I have to say that the case he makes here is pretty compelling. I think this is really the key point:
What you’re doing is setting up a big, centrally planned and operated bureau of music, that officially determines the business model of the recording industry, figures out who gets paid, collects the money and pays some money out. The same record industry that has fought so hard against any innovation remains in charge and will have tremendous sway in setting the “rules.” The plan leaves no room for creativity. It leaves no room for innovation. It’s basically picking the only business model and encoding it in stone.
Oh, and did we mention it’s only for music? Next we’ll have to create another huge bureaucracy and “license” for movies. And for television. And, what about non-television, non-movie video content? Surely the Star Wars kid deserves his cut? And, newspapers? Can’t forget the newspapers. After all, they need the money, so we might as well add a license for news. And, if that’s going to happen, then certainly us bloggers should get our cut as well. Everyone, line right up!
This is a bad plan that will create a nightmare bureaucracy while making people pay a lot more, without doing much to actually reward musicians.
The key thing to remember here is that there’s nothing special about the music industry. The record labels have been hardest hit by peer-to-peer file sharing, but their fundamental problem actually has very little to do with BitTorrent. Rather, their problem is the same problem that’s befallen the newspaper industry: the marginal cost of content has dropped to zero, and so the price of content is also going to be driven to zero sooner or later. The only thing that’s different about the music industry is that BitTorrent has sped the process up: prices have dropped faster because in addition to competing with new entrants, labels are also “competing” with pirated copies of their own content.
But that’s just a transitory phenomenon. The long-run trend is that there’s going to be a much larger eco-system of free music, just as the blogosphere is a large eco-system of free punditry. And in that environment, business models that rely on content being expensive are doomed, just as Craig’s List doomed newspapers built on the premise of expensive classified advertising. I think Mike is probably right that implementing a de facto music tax would have the effect of cementing in place an increasingly anachronistic industry structure.
With that said, a music tax would have some short-term benefits. An effective collective licensing scheme would create a much more fertile environment for entrepreneurs to build innovative technologies on top of peer-to-peer technologies, so maybe a music tax is a price worth paying for the benefits of a peer-to-peer friendly legal environment. But before I get behind the idea, I’d want to see a clear explanation of how such an agreement would apply to other types of media, and what the long-term evolution of the industry would be.
TLF reader Timon makes a really good point about the choice between authenticated and unauthenticated networks:
Lawyers are trained to view complex questions and come up with balanced approaches to them — ie “balancing” privacy with police prerogatives and subpoenas. The technical world is rather the opposite; no matter how complex, an encryption algorithm, for example, either is or is not secure, and as soon as it isn’t it really isn’t. In a legal class it makes for good discussion to say, on the one hand, IP addresses should be private, except when they are used to commit a crime. In direct technical terms what this amounts to is a full surveillance state that is then ruled by court procedure: the law requires someone keep records of all mail or other communications, and then provide them to the authorities when told to. While under law there could be a protection of privacy, in technical terms there is absolutely no privacy, except that which the state decides to concede, the information it declines to look at but which is permanently stored on its orders and available for its inspection. It may seem to you that some people are just unwilling to split the difference and be reasonable, but it really is the case that where lawyers see gray others see black and white, with good technical reasons. There is no way to enforce copyright, for example, and allow anonymous speech online, as you seem to be picking up on. That is not because we are unwilling to be fair, it is a characteristic of information.
Demanding that government limit itself is never a very effective strategy, because if government can do something, it probably will. It’s far more effective to limit government by design. The Internet’s open architecture is, among other things, an important limitation on the government’s surveillance power. Precisely because there are so many ways to get on the Internet without authentication, the government has to do a lot of extra work if it wants to spy on people. For those of us who care about civil liberties, this is a feature, not a bug.
Of course, freedom isn’t free. The same characteristics that make the Internet resistant to a surveillance state also make it harder to enforce copyright, laws against child pornography, etc. This is unfortunate, but I don’t think it’s so unfortunate that we should be willing to toss out the liberty-enhancing benefits of the open Internet. Because as Timon says, there isn’t a middle ground here. Either ISPs have a reliable way to identify their users or they don’t. And if we require them to have this ability for what we regard as good reasons, it’s inevitable that the government will use that same power for bad purposes down the road.
Mike Masnick notes that a grassroots coalition seems to have killed Canada’s version of the DMCA. Of course, legislation backed by powerful interest groups is never dead for good, but for now, it looks like Canada will be DMCA-free for the foreseeable future.
It’s worth remembering that when the actual DMCA was passed here in the US 10 years ago, it faced very little serious opposition. Part of that is probably because Slashdot and the rest of the tech blogosphere was still in its infancy. But I think it’s also a sign of how much progress has been made in spreading awareness about copyright issues and getting non-geeks interested and engaged. I remember organizing an anti-DMCA organization at the University of Minnesota in the wake of the 2001 Sklyarov arrest and struggling to explain to non-techies what the DMCA was and why they should care about it.
The DMCA still isn’t a household concept yet, but knowledge and understanding of the DMCA and other copyright issues is a lot more widespread than it was a decade ago. And at least up North they’ve figured out how to translate that broader public interest into effective political advocacy. If the copyright reform movement continues growing over the next decade the way it has over the last decade, I think there’s a real chance that we’ll be able to stop the otherwise-inevitable Copyright Term Extension Act of 2018. I had sort of resigned myself to perpetual copyright extension, but ideas have consequences, and the debate may become so lopsided by 2018 that even the copyright industry’s millions may not be enough to buy them another 20 years of monopoly rents.
I want to associate myself with Cord’s excellent post about whether Google pays its fair share for bandwidth. Let me also add a more theoretical observation: we know Google is paying its fair share because if it weren’t, the companies that provide it with bandwidth would raise their rates. That may sound like a tautology, but I think it’s actually an important point that tends to get lost in these discussions. Nobody forces ISPs to interconnect, so we can be confident that each party to the web of interconnection agreements we call the Internet is getting at least as much value out as he puts in.
The obsession with whether Google is paying its “fair share” for bandwidth is nonsense for precisely the same reasons it’s silly to fret over trade deficits and “unfair” trade deals. In both cases, people fail to appreciate that we’re talking about positive-sum interactions: interactions in which both parties are made better off and no one is made worse off. Just as the buyer and seller in any given international transaction values what he’s getting more than what he’s giving up, so too does everyone in the chain of contracts between me and Google get more from carrying our traffic than the cost of doing so. Each is making a profit, or at least expected to make a profit ex ante when they agreed to carry it. Which means that in both cases, interfering is likely to only reduce
This means that if you add up all the value Google gets from the Internet and subtract Google’s costs, of course the number you get would be positive. The fact that Google benefits more from the Internet than it pays isn’t an indictment of Google, it’s a reflection of basic economics. I’ve written before that the commonplace idea that there’s no such thing as a free lunch is actually nonsense. To the contrary, a market economy is a free lunch for everyone who participates: almost everybody gets dramatically more value from their participation in the economy than their cost of participation. The same is true of the Internet. Google, Google’s users, Google’s customers, and various network owners are all “free riding” on each other. And this isn’t a problem, it’s the whole point of having positive-sum social institutions like the Internet.
It’s becoming increasingly clear to me that vigorous prosecution of the war on file sharing will lead to some deeply illiberal results. And our good friends at the Progress and Freedom Foundation’s Center for Digital Property periodically write things that confirm the point. Former PFFer Patrick Ross, for example, has compared the war on file sharing to America’s “lax” approach to drug law enforcement. And last year Jim DeLong made the argument that stopping file sharing will require copyright laws so draconian that they will make today’s laws, including the DMCA and lawsuits against 12-year-olds, look “ridiculously permissive.”
PFF’s new copyright guru, Tom Sydnor, seems to be equally enthusiastic about using ever-more-draconian legal penalties and restrictions on civil liberties in order to back up his vision of copyright. His latest target is online anonymity, as he argues that Boston University is guilty of “incompetence” for allowing anonymous communications on its network:
For those seeking to enforce federal laws or rights other than copyrights, this order is all bad news. London-Sire suggests that BU has made its campus network into a de-facto safe harbor for anyone using the Internet to commit any crime. It would seem that terrorists, pedophiles, phishing-scheme operators, hackers, identity thieves, and copyright pirates who can access the Internet through BU’s network now have a get-out-of-jail-free card–a judicial decision holding that any identifying data provided by BU is too hopelessly unreliable to support so much as the filing of a civil lawsuit.
What’s amazing about this argument is that it proves way, way too much: it applies to any network provider that allows customers to communicate without identifying themselves first. So, for example, the Panera down the street from me offers anonymous, free WiFi access. Terrorists, pedophiles, phishing-scheme operators, hackers, identity thieves, and copyright pirates can walk into Panera, commit a variety of crimes, and walk out, and in all likelihood Panera won’t be able to provide the police with any useful information about the culprit. (Panera might have logs showing the user’s MAC address, but these are not easy to match to an individual, and they can be spoofed anyway)
Continue reading →
Incidentally, the bureaucratic dynamic I wrote about in my last post also explains why efforts to “reinvent government” to reduce “waste, fraud, and abuse” never work very well. The problem isn’t that there’s no waste, fraud or abuse, or even that these policies don’t succeed in rooting some of it out. Rather, the problem is that policies designed to make government more efficient and accountable accumulate over time. So the new policies the Obama administration implements to deal with mismanagement that occurred under the Bush administration will largely co-exist with policies implemented under the Clinton, Reagan, Carter, Johnson, Kennedy, and Eisenhower administrations to deal with problems observed under their predecessors. There was probably a good reason for each set of rules by itself, but when you add them up, the result is a kind of death of a thousand cuts where federal bureaucrats can’t get anything done because doing anything requires a huge amount of paperwork. And then of course we have “paperwork reduction acts” where we hire a whole new set of bureaucrats to promulgate still more rules ostensibly designed to make the earlier rules less complicated. Not surprisingly, it doesn’t work especially well.
Another great essay from Paul Graham:
Checks on purchases [at large companies] will always be expensive, because the harder it is to sell something to you, the more it has to cost. And not merely linearly, either. If you’re hard enough to sell to, the people who are best at making things don’t want to bother. The only people who will sell to you are companies that specialize in selling to you. Then you’ve sunk to a whole new level of inefficiency. Market mechanisms no longer protect you, because the good suppliers are no longer in the market.
Such things happen constantly to the biggest organizations of all, governments. But checks instituted by governments can cause much worse problems than merely overpaying. Checks instituted by governments can cripple a country’s whole economy. Up till about 1400, China was richer and more technologically advanced than Europe. One reason Europe pulled ahead was that the Chinese government restricted long trading voyages. So it was left to the Europeans to explore and eventually to dominate the rest of the world, including China.
In more recent times, Sarbanes-Oxley has practically destroyed the US IPO market. That wasn’t the intention of the legislators who wrote it. They just wanted to add a few more checks on public companies. But they forgot to consider the cost. They forgot that companies about to go public are usually rather stretched, and that the weight of a few extra checks that might be easy for General Electric to bear are enough to prevent younger companies from being public at all.
One of the most challenging things about this kind of institutional bloat is that it’s extremely hard to articulate to those in positions of authority precisely how damaging this kind of institutional overhead can be. I’ve been involved in at least one organization (which shall remain nameless) that had created elaborate processes for reviewing and double-checking moderately expensive purchases. And the phenomenon Graham described applied with a vengeance in those cases. The minor cost was the dozens of hours devoted to making the case to the relevant decision-makers for our preferred option. But the more important, but harder to articulate, cost was the way the approval requirements distorted the decision-making process. Since we’d have to defend our choices to decision-makers who didn’t know very much about the options, we tended to overweight factors that could be clearly and easily explained to the decision makers (“This supplier is the market share leader,” “this candidate has a PhD”) and underweight more important but harder-to-articulate qualities of the various options. And because we had to gather a lot of mostly useless information about the options to present to the decision-makers, the most qualified suppliers would often opt out of dealing with us, because they could get business with a lot less hassle elsewhere.
And this organization was not especially large, in the grand scheme of things. These problems tend to get worse as an institution gets larger and older. I was just talking to a friend who works with a firm that provides services to large banks, and she was complaining about the large amounts of money large companies waste on this kind of overhead. And the condition is almost terminal in one of America’s largest and oldest institutions—the federal government. The reason we have $600 toilet seats isn’t that the people buying them are corrupt or incompetent. It’s that selling a toilet seat to the federal government costs $50 to manufacture the toilet seat and $550 to fill out the relevant paperwork.
It’s sad that it even needs to be said, but Mike Masnick reminds us that if you’re writing about “Digital Socialism and the Tyranny of the Consumer” then you’re deeply, deeply confused. The “tyranny of the consumer” is the distinctive feature of free-market economies. And if we were going to label someone in the copyright debate “socialist,” it would be those who advocate government-granted monopolies in the reproduction of creative works, not those who want to repeal them. The author of this piece seems not to grasp the distinction between collectively-owned resources and unowned resources. Here’s a handy cheat sheet:
| Collectively Owned |
Unowned |
The US Post Office
The British National Health Service
American public schools
AIG
Amtrak
The Cuban economy
| Air
Sunlight
The Bible
Tom Sawyer
War and Peace
The TCP/IP protocols |
If you’re a supporter of the kinds of institutions we find in the first column, it might be reasonable to call you a “socialist.” If you support those in the second column, not so much.
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.