Lynne Kiesling of Knowledge Problem has investigated early commentary on the Federal Communications Commission order cartelizing the Internet.  (I'm being pejorative, perhaps unnecessarily, but if the report and order were full of seashells and balloons, wouldn't we have at least seen the executive summary already?)
What’s being sold as keeping the Internet neutral for innovation at the edge of the network is substantively doing so by encasing the existing Internet structure and institutions in amber, which yields rents for its large incumbents. Some of those incumbents, like Comcast and Time-Warner, have achieved their current (and often resented by customers) market power not through rivalrous market competition, but through receiving municipal monopoly cable franchises. Yes, these restrictions raise their costs too, but as large incumbents they are better positioned to absorb those costs than smaller ISPs or other entrants would be. It’s naive to believe that regulations of this form will do much other than softening rivalry in the Internet itself.

But is there really that much scope for innovation and dynamism within the Internet? Yes. Not only with technologies, but also with institutions, such as interconnection agreements and peering agreements, which can affect packet delivery speeds.
Put simply, there exists a market for fast lanes on the information superhighway.  Or, as Cato's Julian Sanchez puts it, there exist unexplored markets for more capacious packets (sorry, all this writing about ships compels me).
As Neutralites love to point out, the neutral or “end-to-end” model has served the Internet pretty well over the past two decades. But is the model that worked for moving static, text-heavy webpages over phone lines also the optimal model for streaming video wirelessly to mobile devices? Are we sure it’s the best possible model, not just now but for all time? Are there different ways of routing traffic, or of dividing up the cost of moving packets from content providers, that might lower costs or improve quality of service? Again, I’m not certain—but I am certain we’re unlikely to find out if providers don’t get to run the experiment. It seems to me that the only reason not to want to find out is the fear that some consumers will like the results of at least some of these experiments, making it politically more difficult to entrench the sacred principle of neutrality in law. After all, you’d think that if provider deviations from neutrality in the future prove uniformly and manifestly bad for consumers or for innovation, it will only be easier to make the case for regulation.

As I argued a few years back, common carrier regimes might make sense when you’re fairly certain there’s more inertia in your infrastructure than in your regulatory structure. Networks of highways and water pipes change slowly, and it’s a good bet that a sound rule today will be a sound rule in a few years. The costs imposed by lag in the regulatory regime aren’t outrageously high, because even if someone came up with a better or cheaper way to get water to people’s homes, reengineering physical networks of pipes is going to be a pretty slow process. But wireless broadband is not a network of pipes, or even a series of tubes. Unless we’re absolutely certain we already know the best way to price and route data packets—both through fiber and over the air—there is something perverse about a regulatory approach that precludes experimentation in the name of “innovation.”
A future generation of practitioners is going to encounter a virtual version of the Big John hopper car.

My younger readers will take such hopper cars in stride (there are strings of them in almost any general freight train that you wait for) and my older but not rail-aware readers might not know that these were disruptive in a way that lowered costs and improved the quality of service.  And that's not chicken feed.
The old boxcars in use by the railroads at the time were leaky and weevil-infested, causing losses en route. And rates were artificially high due to the capacity of the cars, the cost of unloading and a two-stage shipping process. As a result, grain had all but disappeared from the rails and was transported instead via barges along the Mississippi and Tennessee rivers.

The answer: hopper cars, also known as Big John cars. The aluminum covered cars developed by Southern Railway System revolutionized the way feedstock was transported to the South. The leader in development of the innovative cars was famed railroad leader D.W. Bronsan, [c.q.] a Georgian himself and a graduate of Georgia Tech. Bronsan had a keen interest in promoting Georgia industry, specifically its agribusiness.

“The hopper car was the key development that helped Georgia poultry growers feed their chickens efficiently so the industry could grow to the size it is today,” Giles says.

Bronsan’s hopper cars allowed Southern to haul as much grain in five big cars as was being hauled in 25 boxcars, which dramatically reduced the freight charges and, therefore, the cost of the grain. The cost of hauling grain from St. Louis to Gainesville, Ga., for example, dropped from $10.50 to $4.17 per ton in the five-car lots.
Southern's proposed rates, however, violated the regulatory modus vivendi among railroads, truckers, and barge operators, and the boatmen, not the chickens, squawked.
The Southern attempted to implement a 60% rate reduction in August 1961; but the Interstate Commerce Commission blocked these tarriffs [c.q.] on the grounds that they could drive truckers and other railroads out of business.  To qualify for the new rate, a shipment had be at least five cars and had to be shipped on one bill of lading, to one consignee, at one destination.  The railroad offered even lower 10-car and 20-car rates.  While the ICC postponed the implementation of the new rates, they were used on intrastate shipments.

ICC hearings began on 8 January 1962, and the issue went all the way to the Supreme Court twice!.  The Southern won both times.  By the time the new rates finally went into effect on 11 May 1963, the railroad was receiving the first of 500 bigger (4,948 CF) Big Johns, but less than two months later the ICC told the railroad it could only have a 53.5% reduction.  The Southern challanged [c.q.] the ICC again and the matter went back to the Supreme Court, which ruled in the railroad's favor in January 1965.  The next month the railroad received 500 even bigger (5,325 CF) cars, for a total of 1,075 Big Johns.
Today's freight trains comprise fifty-foot and larger cars; those of the steam era featured 28 foot covered hopper cars, 32 foot open hopper cars, and forty-foot boxcars.  Did regulation keep freight cars smaller than the railroad infrastructure could handle?  You decide.  Will regulation keep the internet slower than the information infrastructure can handle?  That will be the topic of doctoral dissertations in the 2020s.

No comments: